Cancer biomarkers


Cancer biomarkers


Cancer biomarkers


Cancer biomarkers

28.1 Malignant diseases

Lothar Thomas

In industrial nations, cancer ranks the second most common cause of death following cardiovascular disease.

28.1.1 Epidemiology

The World Health Organization has reported that 14.1 million new cases of cancer occurred globally and caused approximately 8.2 million or 18.6% of all human deaths in 2012 /1/. Estimated numbers of new cancer cases in 2018 are 1,735,350 and cancer deaths 609,640 in the United States /2/. The corresponding annual numbers in 2013/2014 in Germany are 482,500 and 224,000 /3/, respectively. In the Western states, one in three individuals develops cancer during his or her lifetime, and cancer is the primary cause of death in one of four individuals.

The incidence of cancer increases dramatically with advancing age. Cancer with an incidence of 1–2 per million individuals aged 20 years can have an incidence of 20 per 1,000 individuals aged 80 years (Tab. 28.1-1 – Age dependent incidence of frequently encountered cancers in Great Britain).

Population based incidence of the individual types of cancer differs significantly /4/. For instance, the incidence of skin cancer in Queensland, Australia, and that of esophageal cancer in certain areas of Iran is higher than 20%. Environmental factors, dietary habits, ways of behavior/living, socioeconomic status and religion play a role. The incidence of most frequently encountered cancers in the world is shown in Tab. 28.1-2 – Incidence of most frequently encountered cancers; the percentage distribution of cancer in Germany is shown in Tab. 28.1-3 – Common types of cancer in Germany. Searches related to worldwide frequency of cancer see on the internet Global Cancer Observatory.

28.1.2 Causes of cancer

There is now sufficient evidence of carcinogenicity in humans according to the International Agency for Research on Cancer (IARC). Normal cells in tissue undergo genomic changes during cell division and exposure to agents that causes genotoxic stress /5/.

Genomic changes are point mutations, insertions, deletions, trans locations and changes in copy number.

Genomic stress caused by viruses

Viruses e.g., human T cell lymphotrophic virus, human immunodeficiency virus (HIV), hepatitis B virus (HBV), hepatitis C virus (HCV), Human papilloma virus (HPV), Epstein Barr virus (EBV), Human herpes virus 8, also known as Kaposi’s sarcoma-associated herpes virus (KSHV) /5/.

The Kaposi’s sarcoma and several types of aggressive B cell lymphomas occur most often in patients with CD4+ T cell lymphocytopenia. In patients with HIV induced immune deregulation, immunologic control of onco viruses (KSHV, EBV, HPV, HBV, HCV and Merkel-cell polyoma virus) and virus-infected cells is impaired, which permits the development of cancer /6/.

Genomic stress from other causes

The following causes have also been identified by the IARC: sunlight, tobacco, pharmaceuticals, hormones, alcohol, parasites fungi, bacteria, salted fish, wood dust and herbs /5/.

The World Cancer Research Fund and the American Institute for Cancer Research have determined the additional causes of cancer which include beta carotene, red meat, processed meats, low fibre diets, non breast feeding, obesity, increased adult height and sedentary life stiles /5/.

Cancer Stem cells

In tissues where cancers normally arise only a minority population of stem cells (1 of 5–100 thousand)maintains and repairs the tissue, whereas most replication competent cells have a limited life span. With Inflammation and activation of the innate immune system by toll-like receptor signaling, stem cells are activated by toll-like receptor signaling pathway to repair the tissue damage. Stem cells are the longest-living cells in many tissues, and multiple mutations must accumulate in cancer. Investigations with hematopoietic stem cells indicate that cancer arises from sequential mutations accumulating in tissue stem cells /7/.

28.1.3 Classification of malignant tumors

Tumors are classified according to the TNM (tumor nodes, metastases) staging system recommended by the Union International Contre le Cancer (UICC) in 1968 and amended in 1986. The TNM system thus allows, classification according to two distinct systems the clinical TNM and pathological TNM (Tab. 28.1-4 – Tumor classification according to the TNM system).

Clinical TNM:

  • T describes the size of the primary tumor and whether it has invaded nearby tissue
  • N describes nearby (regional) lymph nodes that are involved
  • M describes distant metastases (spread cancer).

Pathological TNM:

The G category (grading) describes the degree to which the tumor has differentiated. The following tumor grades are generally used:

  • G1, well differentiated, similar to normal cells
  • G2, moderately differentiated
  • G3, poorly differentiated
  • G4, undifferentiated
  • GX, differentiation grade cannot be assessed.

28.1.4 Metastatic spread

The aggressiveness of tumors is based on their ability for metastatic spread. Investigations on colon cancer have shown that:

  • Tumors still confined to the colon wall do not cause a reduction in life expectancy post surgery
  • Life expectancy is normal in only one third of patients if nearby lymph nodes are affected
  • Chances of 5-year survival are extremely poor in the presence of metastasis.

Generally, the ability of a tumor for metastatic spread is influenced by the tumor origin, grade of differentiation and size. Certain primary tumors are predisposed to metastasize as follows:

  • Prostate cancer; early bone metastasis
  • Ovarian cancer; development of malignant ascites
  • Breast cancer; multiple metastasis (bone, lung, brain)
  • Small cell lung cancer; multiple metastasis (bone, lung, brain)
  • Colorectal cancer, gastric cancer, pancreatic cancer; primarily liver metastasis
  • Melanoma; primarily brain, liver, lung, ovary.

The determination of circulating tumor cells, tumor markers and enzymes plays an important role in the detection and monitoring of metastatic tumors /8/, with the determination of enzymes being especially significant in bone and liver metastasis. For instance, the diagnostic sensitivity of ALP is 20% in solitary bone metastasis and 70% in multiple bone metastasis.

Two thirds of malignant tumors in the liver are due to metastasis and one third is caused by hepatocellular carcinoma. Besides the enzymes GGT, ALP and LD, AFP is the most important diagnostic marker to diagnose hepatocellular carcinoma. On the other hand CEA is the marker of choice in liver metastasis. In the presence of liver metastases CEA is elevated in 80% of pancreatic cancer, 71% of lung cancer, 71% of colorectal cancer and 54% of breast cancers, respectively. CEA concentrations above 20 μg/L have a markedly higher positive predictive value for liver metastasis than for hepatocellular carcinoma.

Three phases determine the process of metastasis /910/:

  • The pre-colonization phase involves intravasation of tumor cells within the vasculature of the tumor. Circulating tumor cells (CTCs) travel through the vasculature of the body as single cells or aggregates, pass in target organs, and arrest in capillaries to start colonization. One of the main theories regarding the mechanism for such invasion is epithelial-to-mesenchymal transition (EMT).
  • Colonization starts with extravasation of CTCs into the mesenchyme, which set up resistance to host-tissue immunity, and may encounter a favorable micro environment known as metastatic niche. Subsequently the cells carry out a long standing mesenchymal survival as micro metastasis or single cells.
  • Finally the cells end latency and undergo a reversal step, the mesenchymal-to-epithelial transition (MET). Building new metastases, cancer progression continues at multiple different sites.


1. World Cancer Report 2014. World Health Organization, Chapter 1.1; 2014.

2. American Cancer Society. Cancer facts and figures 2018.

3. Bericht zum Krebsgeschehen in Deutschland 2016.

4. Blackadar CB. Historical review of the causes of cancer.World J Clin Oncol 2016; 7: 54–86.

5. Blackadar CB. Historical review of the causes of cancer. World J Clin Oncol 2016; 7: 54–86.

6. Yarchoan R, Uldrick TS. HIV-associated cancers and related diseases. N Engl J Med 2018, 378: 1029–41.

7. Clarke MF. Clinical and therapeutic implications of cancer stem cells. N Engl J Med 2019; 380: 2237–45.

8. Lamerz R, Reithmeier A, Stieber P, Eiermann W, Fateh-Moghadam A. Role of blood markers in the detection of metastases from primary breast cancer. Diagn Oncol 1991; 1: 88–97.

9. Burz C, Pop VV, Buiga R, Daniel S, Samasca G, Aldea C, Lupan I. Circulating tumor cells in clinical research and monitoring patients with colorectal cancer. Oncotarget 2018; 9: 24561–571.

10. Thiele JA, Bethel K, Kralickova M, Kuhn P. Circulating tumor cells: fluid surrogates of solid tumors. Annu Rev Pathol Mech Dis 2017; 12: 419–47.

11. 40 Jahre epidemiologisches Krebsregister Saarland. Ministerium für Gesundheit, Justiz und Soziales. Saarbrücken 2007.

12. Parkin DM, Stjernsward J, Muir CS. Estimates of the worldwide frequency of twelve major cancers. Bull World Health Org 1984; 62: 163–82.

28.2 Oncogenes, tumor suppressor genes

Lothar Thomas

28.2.1 Carcinogenesis

According to estimates of the American Cancer Society, one in two male and one in three female individuals develop cancer during his or her lifetime /1/. Cancer is a genomic disease. To date, 70 oncogenes resulting from germ line mutations and 342 associated with somatic mutations have been detected. Cancer is a group of more than 100 different diseases causing uncontrolled growth. The switch from a normal cell to cancer cells is caused in the majority by mutations of their genes. Cancer cells show a certain phenotype including:

  • Increased cell division
  • Reduced cell differentiation
  • Suppression of apoptosis.

Three groups of genes and their encoded proteins are typically mutated in the presence of cancer:

  • Oncogenes
  • Tumor suppressor genes
  • Repair genes.

Physiologically all three gene groups are responsible for regulating cell proliferation and differentiation. Oncogenes promote cell growth, while tumor suppressor genes have an inhibiting effect. The cell loses its rhythm of growth and enters an uncontrolled growth phase if the effects of oncogenes and tumor suppressor genes become imbalanced /2/. Repair genes take action if genetic damage occurs. They detect and can repair the damage in many cases. If the damage cannot be repaired, the cell will get a signal telling it to die in a process called apoptosis. This prevents the genetic damage from being passed on during cell division.

Mutations in two basic classes of genes, proto oncogenes and tumor suppressor genes, are what leads to cancer. Driver mutation are mutations in (proto) oncogenes that initiate the growth of the respective tumor entity.

A gene and its function are characterized by the sequence of bases (nucleotides). Changes in the base sequence (mutations) of a gene lead to defects and affect a cell in many ways. Specific gene defects activates the gene to produce more protein while others genes cause the opposite effect. Some mutations have no effect.

Approximately 90% of cancers result from sporadic DNA mutations over the course of life, especially in older age. Sporadic mutations are also called somatic or acquired mutations. They can be triggered by events in the environment such as exposure to toxins or radiation.

Hereditary mutations, also referred to as germ line mutations, are responsible for tumor development and progression in 5–10% of cancers. It is not the cancer itself that has been inherited, but the predisposition of development. Contrary to spontaneous cancers, the genetic damage in germ line mutations is present in the spermatocytes or oocytes. Transmitted gene defects in germ cells do not immediately cause cancer development because there is still the gene of the healthy parent. Cancer may, however, develop if this gene is damaged later in life. A list of hereditary cancer syndromes has been published under https://themedicalbiochemistrypage.org/oncogene.php.

Cancer develops from a single genetically impaired cell. The damage cannot be repaired and apoptosis does not take place. As a consequence, cancer develops, especially if a gene involved in cell division or apoptosis is affected. During malignant transformation, an emerging clone must circumvent the antineoplastic countermeasures that usually regulate the activity of proto oncogenes. Tumor cells can then use the beneficial properties of an oncogene.

Examples of tumorigenesis /3/:

  • Proto oncogenes such as Myc induce cellular proliferation. However, intrinsic “fail-safe” apoptotic mechanisms such as the ARF/MDM2/p53 pathway counteract the mitotic stimulus mediated by Myc and suppress transformation. Loss of function mutation of p53 or over expression of BCL2 are required to facilitate Myc driven tumorigenesis.
  • Oncogenes such as BCL2 are potent inhibitors of apoptosis but poor inducers of cell proliferation. They are, therefore, insufficient to drive tumorigenesis. The cooperative activity of oncogenes driving cell proliferation (Myc, Ras) and BCL2 suppresses apoptosis, thus promoting tumorigenesis.
  • Certain oncogenes such as BCR-ABL can activate intracellular signaling pathways that simultaneously induce cell proliferation and suppress apoptosis, thereby leading to transformation.

Cancer is now cosidered a genetic disease. Frequently genes of cancer patients are found to malfunction. This may be due to gene translocation (gene movement to abnormal positions), amplification (acquiring two or more gene copies) mutation (change of normal sequence), deletion (loss of a gene or section of a gene), or abnormal regulation (under- or overexpression) /4/.

28.2.2 Proto oncogenes

Proto oncogenes and proto oncogene products are involved in pathways related to cell division and differentiation. Oncogene products presumably affect these same proto oncogene pathways but in an aberrant fashion /4/.

28.2.3 Oncogenes

Per definition, an oncogene is a gene whose abnormal expression or altered gene product leads to malignant transformation of the cell.

Oncogenes are mutations of proto oncogenes. Typically, they are dominant and counteract the function of proto oncogenes. Oncogenes increase the expression of the proto oncogene responsible for the production of cell regulating proteins, resulting in increased cell division, reduced differentiation and suppression of apoptosis. This generally leads to cell transformation and the cell is out of control /5/.

There are various mechanisms of oncogene activation based on /5/:

  • Up regulated expression of a normal gene
  • Expression of a mutant protein with increased stability or changed function
  • Changed subcellular localization of a normal gene product due to interaction with a false or mutant binding protein.

Molecular changes that transform or activate a proto oncogene to become an oncogene include point mutation, deletion, insertional activation, amplification and translocation. The Philadelphia chromosome is a characteristic example of chromosomal translocation, in which the ends of chromosome 9 and chromosome 22 swap places. The broken end of chromosome 22 contains the BCR gene which fuses with a fragment of chromosome 9 that contains the ABL1 gene. When these two chromosome fragments fuse, the genes also fuse creating a new gene BCR-ABL which, due to the ABL1 half of the gene, encodes for a protein that displays high tyrosine kinase activity. The unbridled expression of the protein of the ABL1 half activates a range of other proteins involved in cell cycle regulation, thereby stimulating cell division. As a result, the Philadelphia chromosome is associated with chronic myelogenous leukemia as well as other forms of leukemia.

Mutations are a common cause of cancers. High throughput oncogene mutation profiling across 1,000 samples of 17 different cancer types had the following result /6/: of 17 oncogenes analyzed, 14 were found to be mutated at least once, and 298 samples carried at least one mutation.

Examples of proto oncogenes and oncogenes and their clinical significance are listed in Tab. 28.2-1 – Examples of proto oncogenes and oncogenes and their clinical significance.

The functional grouping of oncoproteins is shown in Tab. 28.2-2 – Functional grouping of oncoproteins.

28.2.4 Tumor suppressor genes

Tumor suppressor genes are anti-oncogenes. These genes, when mutated or deleted, deprive the cell of their protein products. They normally function as physiologic barriers against clonal expansion or genomic mutability and are able to hinder the growth and metastasis of cells to uncontrolled proliferation by oncogenes.

Important tumor suppressor genes include TP53, BRCA1, BRCA2, APC and RB1. An important difference between oncogenes and tumor suppressor genes is that oncogenes result from the activation (turning on) of proto oncogenes, but tumor suppressor genes cause cancer when they are inactivated (turned off) /2/.

Oncogenes and tumor suppressor genes in malignant tumors are listed in Tab. 28.2-3 – Oncogenes and tumor suppressor genes in malignant tumors.

28.2.5 Molecular targeting in cancer therapy

Drugs have been developed which attack oncogenes such as the protein HER2/neu /7/, reduce cell proliferation and prolong the life of patients with, for example, breast cancer. A selection of such drugs is shown in Tab. 28.2-4 – Oncogene addiction to particular agents. The treatment of cancer inducing tumor suppressor genes is difficult. In many cases, profiler PCR assays are used. They enable simultaneous analysis of a large number of genes.

28.2.6 Causes of genetic mutations

Carcinogenic substances can cause the transformation of a proto oncogene to an oncogene. Three groups of inducing mechanisms (chemical substances, viruses, radiation) have been defined.

Chemical substances

The most important source of chemical carcinogens is tobacco smoke. It is a combination of agents (nitrosamines, polycyclic aromatic hydrocarbons, benzenes) together leading to cell damage. Tobacco smoke is responsible for more than 5% of cancer cases in women and more than 20% in men. For instance, tobacco smoke causes the majority of lung, pancreatic, bladder and esophageal cancers. Moreover, the combination of tobacco smoke and alcohol is thought to be responsible for most oral, pharyngeal, laryngeal and esophageal cancers. Important carcinogenic substances besides tobacco smoke are mold toxins (aflatoxins) and nitrosamines in food.


Tumor cells can form due to specific actions of onco viruses. There are two types of onco viruses:

  • Viruses with DNA genome (Papillomavirus, Adenovirus). Proteins encoded by DNA onco viruses (also called tumor antigens/T antigens) interact with other cell proteins. This primarily refers to tumor suppressor type proteins. The interaction interferes with the normal function of the proteins (i.e., cell homeostasis).
  • Viruses with RNA genome, also called retroviruses. They are common in birds, cats and mice and rare in humans. Human retroviruses are, for example, the Human T lymphotropic virus (HTLV) and Human immunodeficiency virus (HIV). The RNA genome of a retrovirus infested cell is transformed to DNA by reverse transcriptase and integrated into the genome of the host cell. The exchange of DNA between the host and the virus (transduction) can lead to uncontrolled proliferation of the host cell and, thus, tumorigenesis.


Radioactive contamination can cause ruptures of DNA strands and result especially in leukemia. Exposure to sunlight can induce the development of melanoma.

28.2.7 Liquid biopsy

Liquid biopsy is the analysis of biomarkers in a non-solid biological tissue, mainly blood, which has remarkable advantages over the traditional tumor biopsies /17/.

Advantages are /18/:

  • Minimally invasive nature
  • No risk (e.g., associated with significantly less morbidity)
  • Painless
  • Does not require surgery
  • Provides temporal measurement of tumor burden and early evidence of recurrence or resistance
  • Provides a personalized snapshot of disease at successive time points
  • Reflects the genetic profile of all tumor sub clones present in a patient, unlike tissue biopsies which are obtained from only one tumor region
  • Reduces cost and diagnosis time.

Circulating cells and molecules that can be determined in liquid biopsies are /19/:

  • Circulating tumor cells (CTCs)
  • Cell-free DNA (cfDNA) that can be used to measure DNA levels, integrity, methylation, tumor DNA (ctDNA) mutational status and copy number of aberration.
  • Circulating RNA classes (e.g., miRNAs). Key sources of circulating RNA include CTCs and tumor cell exosomes. Refer to Section 28.5 – Micro RNA.

CTCs are intact cells and represent entire tumor-derived genome equivalents. ctDNA originates from apoptotic cells and is highly fragmented, ranging approximately 160 base pairs. CTCs and circulating and cfDNA are valuable tools and both distinct and complementary types of information can be derived from them:

  • CTCs are used for phenotyping, genotyping, primary cell-line culture and patient-derived xenografts.
  • ct-DNA represents the tumor burden and provides real-time molecular information for monitoring treatment response and relapse.

Both CTCs and cfDNA represent powerful resources for deducing the genome wide copy-number status, providing insights into the biology and evolution of cancer and their impact in the management of patients /19/.

28.2.8 Technologies of genetic diagnostics

Monogenic malignant tumors, where a single genetic change leads to cancer are rare and affect tumor predispositions such as ovarian cancer and the hereditary non-polyposis colorectal cancer. The sequence of the single stranded immobilized nucleic acid is determined according to Sanger. Sanger sequencing, also known as the chain termination method, is a technology for determining the nucleotide sequence of DNA.

Of malignant tumors the far higher number is caused with changes in many hereditary dispositions. The sequential individual analyis according to Sanger makes little sense in such cases. It is important to determine many selected genes at the same time with high throughput (gene panel) of all DNA sequences encoding proteins (exome). Next-generation sequencing makes large-scale whole-exome sequencing accessible. It enables scientists to analyze the entire human genome in a single sequencing experiment, or sequence thousands to tens of thousands of exomes in a short time /11/.


1. American Cancer Society. Cancer facts and figures 2008.

2. American Cancer Society. Oncogenes, tumor suppressor genes and cancer. American Cancer Society database 2011.

3. Shortt J, Johnstone RW. Oncogenes in cell survival and death. Cold Spring Harb Perspect Biol 2012; 4: a009829.

4. Diamandis EP. Oncogenes and tumor suppressor genes: new biochemical tests. CRC Laboratory Sciences 1992; 29: 269–305.

5. Chin L, Gray JW. Translating insights from the cancer genome into clinical practice. Nature 2008; 452: 553–63.

6. Croce CM. Oncogenes and cancer. N Engl J Med 2008; 358: 502–11.

7. Thomas RK, Baker AC, DeBiasi RM, Winckler W, LaFramboise T, Lin WM, et al. High throughput oncogene mutation profiling in human cancer. Nature Genetics 2007; 39: 347–51.

8. Brown J, O’Carrigan B, Jackson SP, Yap TA. Targeting DNA repair in cancer. Cancer Discovery 2017; 7 (1): 20–37.

9. Die Meo A, Bartlett J, Cheng Y, Pasic MD, Yousef GM. Liquid biopsy: a step forward towards precision medicine in urologic malignancies. Molecular Cancer 2017; 16: 80. https://doi.org/10.1186/s12943-017-0644-5.

10. Marrugo-Ramirez J, Mir M, Samitier J. Blood based cancer biomarkers in liquid biopsy: a promising non-invasive alternative to tissue biopsy. Int J Mol Sci 2018; 19: 2877. https://doi.org/10.3390/ijms19102877.

11. Shendure J, Balasubramanian S, Church CM, Gilbet W, Rogers J, Schloss Ja, Waterston RH. DNA sequencing at 40: past, present and future. Nature 2017; 550: 345–53.

12. Thomas J St J. Quality assurance in Her-2 testing: redefining the gold standard. Clin Chem 2009; 55 (7): 1265–7.

13. Kobayashi Y, Mitsudomi T. Not all epidermal growth factor receptor mutations in lung cancer are created equal: perspectives for individualized treatment strategy. Cancer Science 2016; 107: 1179–86.

14. Xiao Y, Gao X, Maragh S, Telford WG, Tona A. Cell lines as candidate reference materials for quality control of ERBB2 amplification and expression assays in breast cancer. Clin Chem 2009; 55 (7): 1307–15.

15. Bose M, Cabanillas ME, Cohen EFW, Wirth LJ, Riehl T, Yue H, Sherman Si, et al. Vemurafenib in patients with BRAFV600E-positive metastatic or uresectable papillary thyroid cancer refractory to radioactive iodine: a non-randomised, multicentre, open-label, phase 2 trial. Lancet Oncol 2016; 17: 1272–82.

16. Iyer R, Wehrmann L, Golden RL, Naraparaju K, Croucer JL, MacFarland SP, et al. Entrectinib is a potent inhibitor of Trk-driven neuroblatomas in a xenograft mouse model. Cancer Letters 2016; 372: 179–86.

17. Itatani Y, Kawada K, Yamamoto T, Sakai Y. Resistance to anti-angiogenic therapy in cancer–alterations to anti-VEGF pathway. J Mol Sci 2018; 19: 1232. doi: 10.3390/ijms19041232.

18. Chandra V, Kim JJ, Mittal B, Rai R. MicroRNA aberrations: An emerging field for gallbladder cancer management. World J Gastroenterol 2016; 22: 1787–99.

19. Graziani G, Szabo C. Clinical perspectives of PARP inhibitors. Pharmacol Res 2005; 52: 109–18.

20. Van Lanschot JJB, Polkowski W, Obertop H, Offerhaus GJH. What is known about growth factors, oncogenes, and tumor suppressor genes in the evolution of Barrett’s dysplasia and carcinoma. www.oeso.org/OESO/books/Vol_5_Eso_Junction/Articles/art295.html

21. Sizemore GM, Pitarresi JR, Balakrishnan S, Ostrowski C. The ETS family of oncogenic transcription factors in solid tumours. Nature Reviews Cancer 2017; 17: 337–51.

22. Mohamed AA, Tan SH, Xavier CP, Katta S, Huang W, Ravindranath L, et al. Synergistic activity with NOTCH inhibition and androgen ablation in ERG-positive prostate cancer. Mol Cancer Res 2017; 15 (10): 1308–17.

23. Cseh AM, Fabian Z, Sümegi B, Scorrano L. Poly(adenosine diphosphate-ribose) polymerase as therapeutic target: lessons learned from its inhibitors. Oncotarget 2017; 8 (30): 50221–50239.

24. Harris CC, Hollstein M. Clinical implications of the p53 tumor-suppressor gene. N Engl J Med 1993; 329: 1318–27.

25. Burstein HJ. Systemic therapy for estrogen receptor-positive, HER2-negative breast cancer. N Engl J Med 2020; 383 (26): 2557–70.

26. Clark GM, Osborne CK, McGuire WL. Correlations between estrogen receptor, progesterone receptor, and patient characteristicsin human breast cancer. J Clin Oncol 1984; 2: 1102–9.

27. Baretta Z, Mocellin S, Goldin E. Olopade OI Huo D. Effect of BRCA germline mutations on brest cancer prognosis. A systemic review and meta-analysis. Medicine 2016; 95: 40(e4975).

28. Antoniou AC, Casadei S, Heikkinen T, Barrowdale D, Pylkäs K, Roberts J, et al. Breast cancer risk in families with mutations in PALB2. N Engl J Med 2014; 371 (6): 497–506.

29. Papaemmanull E, Gerstung M, Bullinger L, Gaidzik VI, Paschka P, Roberts ND, et al.Genomic classification and prognosis in acute myeloid leukemia. N Engl J Med 2016; 374 (23): 2209–21.

30. Lu KH, Broaddus RR. Endometrial cancer. N Engl J Med 2020; 383 (21): 2053–64.

31. Zheng F, Zhang Y, Chen S, Weng X, rao Y, Fang H. Mechanisms and current progress of poly ARP-ribose polymerase (PARP) inhibitors in the treatment of ovarian cancer. Biomedicine & Pharmacotherapy 2020; 123: 109661.

32. Romaszko AM, Doboszynska A. Multiple primary lung cancer: aliterature review. Adv Clin Exp Med 2018; 27 (5): 725–30.

33. Sandoval GJ, Pulice JL, Pakula H, Schenone MA, Takeda DY, Pop M, et al. Binding of TMPRSS2 to BAF chromatin remodeling complexes mediates prostate oncogenesis. Mol Cell 2018; 71 (4): 554–66.

34. De Bone J, Mateo J, Fizazi K, Saad F, Shore N,Sandhu S, et al. Olaparib for metastatic castration-resistant prostate cancer. N Engl J Med 2020; 382 (22): 2091–2102.

35. Sriramulu S, Ramachandran M, Subramanian S, Kannan R, Gopinath M, Sollano J, et al. A review on role of ATM gene in hereditary transfer of colorectal cancer. Acta Biomed 2018; 89: 463–9.

28.3 Circulating tumor cells

Lothar Thomas

In most solid tumors, it is distant metastases rather than the primary tumor which limit the prognosis. Distant metastases are caused by circulating tumor cells (CTCs) which actively invade the blood stream, attach to the endothelium in the target organ, invade the surrounding parenchyma, and form new tumors /1/. Tumor cells present in the blood flow are named CTCs, the ones that reach and implant in the bone marrow are called disseminated tumor cells (DTCs) /6/.

Most CTCs are intact cells and represent entire tumor genome equivalents. Blood samples and specific reagents are required to analyze CTCs, circulating free DNA (cfDNA) and exosomees /2/.

28.3.1 Indication

In total: predictor of overall survival (OS) and progression free survival (PFS) e.g.,

  • Importance in early stage and metastatic cancer.
  • In non-metastatic cancer, preoperative as prognostic marker.
  • In patients that underwent surgical resection of a primary evaluation of an increased risk of postoperative metastasis.

28.3.2 Method of determination

CTCs are distinguished from surrounding hematopoietic cells by their size and the expression of epithelial cancer specific markers. Prominent CTC isolation and detection techniques are presented in Ref. /2/.

CTCs are characterized by:

  • The presence of a cell nucleus
  • A diameter > 5 μm
  • Positive selection for the cytoplasmic expression of cytokeratin (CK). CK is labeled with a fluorescence dye and the cell nucleus is stained using 4,6 diamidino-2-phenyl indole (DAPI). CTCs are defined as CK+/CD45/DAPI+ /4/.
  • Negative selection for the expression of the leukocyte antigen CD45.

From the methods, only the CellSearch system was approved by US Food and Drug Administation for breast, prostate and colorectal cancer in 2004. According to this method CTCs are defined as Cell adhesion molecule (EpCAM)+ , Cytokeratin+ and CD45.

Strategy for CTC detection

CTCs detection techniques can be realized through enrichment strategies depending on their physical or biological properties or without cell enrichment /3/.

Enrichment step

The CTC enrichment method is based on the application of magnetic nano particles. These nano particles are coated with a polymer layer of biotin analogs loaded with anti-epithelial cell adhesion molecule (EpCAM) antibodies for CTC binding in the sample. The cells are immunomagnetically enriched before they are characterized and counted. This method has a recovery rate of 93% with a detection limit of one CTC in 7.5 mL of blood /4/.

After enrichment, the CTC fraction usually still contains a substantial number of leukocytes, and CTCs need to be identified at a single cell level by a method that can distinguish tumor cells from normal blood cells.

Identification step

There are many methods of detection of CTCs such as immunochemistry, flow cytometry, reverse transcriptase polymerase reaction, immunomagnetic separation, microchips, fiber optic array scanning technology, and density based cell mechanism combined with digital scanning microscope.

28.3.3 Specimen

Container with a specific solution

28.3.4 Threshold

Approximately 99.7% of healthy people and a limited set of people with benign disease show < 2 CTCs per 7.5 mL blood sample /5/.

28.3.5 Clinical significance

At early stages, from the primary tumor after epithelial-to-mesenchymal transition of primary tumor cells, CTCs begin to flow into blood stream at a rate of roughly 106 cells per tumor gram /6/. CTCs travel through the vasculature as single cells or aggregates and contribute to forming a distant metastasis. Although not every CTC represents a potential future metastasis, many distant metastases are considered to be established by hematogenous spread of these cells /7/. If CTCs are present in a peripheral blood collection, they account for a fraction of 1 × 10–4 percent of all nucleated cells (1–10 cells/mL) /8/.

The diagnostic sensitivity of CTC detection in patient blood is highly dependent on the method of detection. For the CellSearch system sensitivity of 85% has been reported with other systems the sensitivity can be as high as 99.9%. The diagnostic specificity can be as high as 100% /9/.

Despite the large amount of CTCs released daily, they are found in low concentrations in the peripheral blood. This dilemma is caused by platelets cloaks or coagulation factors the surround the CTCs, shielding them from the immune surveillance. As a result, a fraction of cells may remain undetectable /6/.

In early tumor stage CTC detection is correlated with tumor progression and poor prognosis, whereas in metastatic tumors the CTCs high levels indicate disease progression, along with an overall poor outcome /6/.

The CTC prevalence in patients with malignant disease differs with carcinoma type. In early breast cancer CTCs are detected in 18–30% of patients compared with detection rates of approximately 70% in patients with metastatic disease. A study /10/ using CellSearch and blood samples of metastatic patients with colon, breast, rectal, gastric, ovarian, and prostate cancer resulted in 54% of total samples with detected CRCs, including 71% positive CTC samples for breast cancer, 64% for colon cancer, 33% for gastric cancer, 66% for rectal cancer, 60% for ovarian cancer, and 20% for prostate cancer.

In patients that underwent surgical resection of the primary colorectal tumor CTCs revealed an increased risk of postoperative metastasis /11/.

Higher levels of CTCs are correlated with tumor relapse due to their conversion in cancer stem cells that start recurrence /12/.

CTCs fulfill a better disease monitoring and prognostic marker in colorectal cancer than tumor markers like CEA /13/. Monitoring of patients with metastatic cancer is the most practical application of CTCs. In this regard, it has been demonstrated that 50–70% of patients with breast, colon and prostate cancers have elevated CTC levels. A meta analysis /14/ provides evidence that the presence of CTCs in peripheral blood is significantly associated with poorer prognosis both in early stage and metastatic breast cancer (Tab. 28.3-1 – Prognostic value of circulating tumor cells in breast cancer).

Moreover, elevated CTC levels prior to initiation of a new systemic therapy are associated with a worse prognosis than those below the cutoff value. Persistently elevated CTCs or rising CTC levels under systemic therapy strongly suggest that the therapeutic regimen with which the patient is being treated is not working /15/. Various national and international studies have demonstrated the relevance of early CTC detection in metastatic breast cancer for determining the efficacy of systemic chemotherapy and testing new treatment strategies using, for example, anti-HER2 drugs /16/.

Disadvantages in the assessment of CTCs are /17/:

  • CTCs are highly heterogenous, including epithelial tumor cells, epithelial-to mesenchymal cells, hybrid epithelial/EMT (epithelial-to mesenchymal transition) cells and cancer stem cells.
  • It is not possible to directly obtain a pure CTC population, thus mandating a further purification step to overcome blood-cell contamination for the downstream genetic analysis of single cells, which is hampered by the high error rate of the genome amplification procedures required for single-cell genomics.
  • Single-cell DNA sequencing data from CTCs obtained after whole-genome amplification are often characterized by low-coverage stretches, nonuniform coverage, false positive errors introduced by PCR, false negative errors due to insufficient coverage, allele dropout, and allelic imbalance.


1. Garcia SA, Weitz J, Schölch S. Circulating tumor cells. Methods Mol Biol 2018; 1692: 213–9.

2. Batth IS, Manier S, Ghobrial IM, Menter D, Kopetz S, Li S. Circulating tumor markers: harmonizing the yin and yang of CTCs and CtDNA for precision medicine. Ann Oncol 2017; 28: 468–77.

3. Cubero MJA, Lorente JA, Robles-Fernandez I, Rodriguez-Martinez A. Puche JL, Serrano MJ. Methods Mol Biol 2018; 1692: 282–303.

4. Marrugo-Ramirez J, Mir M, Samitier J. Blood based cancer biomarkers in liquid biopsy: a promising non-invasive alternative to tissue biopsy. Int J Mol Sci 2018; 19: 2877; https://doi.org/10.3390/ijms19102877.

5. Nagrath S, Sequist LV, Maheswara S, Bell DW, Irima D, et al. Isolation of rare circulating tumor cells in cancer patients by microchip technology. Nature 2007; 450 (7173): 1235–9.

6. Burz C, Pop VV, Buiga R, Daniel S, Samasca G, Aldea C, Lupan I. Circulating tumor cells in clinical research and monitoring patients with colorectal cancer. Oncotarget 2018; 9: 24561–571.

7. Thiele JA, Bethel K, Kralickova M, Kuhn P. Circulating tumor cells: fluid surrogates of solid tumors. Annu Rev Pathol Mech Dis 2017; 12: 419–47.

8. Miller MC, Doyle GV, Terstappen LWMM. Significance of circulating tumor cells detected by theCEllSearch system in patients with metastatic breast, colorectal and prostate cancer. J Oncol 2010; 617421.

9. Allard WJ, Matera J, Miller MC, Repollet M, Conelly MC, et al. Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with nonmalignant diseases. Clin Cancer Res 2004; 10: 6897–904.

10. Balic M, Dandachi N, Hofmann G, Samonigg H, Loibner H, et al. Comparison of two methods for enumerating circulating tumor cells in carcinoma patients. Cytom B Clin Cytom 2005; 68: 25–30.

11. Wang JY, Wu CH, Lu CY, Hsieh JS, Wu DC, Huang SY, Lin SR. Molecular detection of circulating tumor cells in the peripheral blood of patients with colorectal cancer using RT-PCR: significance of the prediction of postoperative metastasis. World J Surg 2006; 30: 1007–13.

12. Mitra A, Mishra L, Li S. EMT, CTCs and CSCs in tumor relapse and drug resistance. Oncotarget 2015; 6: 10697–711.

13. Huang MY, Tsai HL, Huang JJ, Wang JY. Clinical implications and future perspectives of circulating tumor cells and biomarkers in clinical outcomes of colorectal cancer. Transl Oncol 2016, 9: 340–7.

14. Zhang L, Riethdorf S, Wu G, Wang T, Yang K, Peng G, et al. Meta-analysis of the prognostic value of circulating tumor cells in breast cancer. Clin Cancer Res 2012; 18: 5701–10.

15. Hayes DS, Smerage JB. Circulating tumor cells. Prog Mol Biol Translat Sci 2010; 95: 95–112.

16. Bidard FC, Fehm T, Ignatiadis M, Smerage JB, Alix-Panabieres C, Janni W, et al. Clinical application of circulating tumor cells in breast cancer: overview of the current interventional trials. Cancer Metastasis Rev 2013; 32: 179–88.

17. Heitzer E. Circulating tumor DNA for modern cancer management. Clin Chem 2020; 66 (1): https://doi.org/101373/clin chem.2019.304774.

28.4 Cell-free DNA in cancer patients

Lothar Thomas

Similar to circulating tumor cells (CTCs) cell free DNA has attract attention in clinical research and cancer treatment. Cell free DNA (cfDNA), which contains tumor derived DNA fragments (circulating DNA; ctDNA) in patients with cancer, is a valuable tool for monitoring recurrence, resistance and metastasis. In comparison to CTCs cfDNA represents more of the tumor burden and provides real-time molecular information for monitoring treatment response and relapse /12/.

28.4.1 Circulating cell-free nucleic acids in plasma and serum

The term cfDNA refers to fragmented DNA found in the non-cellular component of the blood. DNA, mRNA and micro RNA are released into the blood through apoptosis or necrosis and are rapidly cleared from the circulation with a half-life time of an hour or less. The cfDNA is typically found as double stranded fragments of approximately 150 to 200 base pairs in length, corresponding to nucleosome-associated DNA /3/.

CfDNA results from tumor cells or from membrane vesicles (exosomes).

Exercise, trauma, myocardial infarction, stroke, and end stage renal failure represent some of the situations that increase the concentration of cfDNA. In healthy persons the concentration of cfDNA is approximately 10 to 15 μg/L /3/.

In cancer patients, a small portion of cfDNA results from tumor shedding and is referred to as circulating tumor DNA (ctDNA). The fraction of ctDNA in tumor patients can vary from 0.1% to more than 90% /1/. In cancer patients, ctDNA carries tumor-related genetic and epigenetic alterations that are relevant to cancer development, progression and resistance to therapy. These alterations include loss of heterozygosity (LOH)and mutations of tumor suppressor genes (such as TP53) and oncogenes (such as KRAS and BRAF) /4/. The fraction of ctDNA tends to parallel tumor burden within an individual patient. However, there is substantial variability among patients with the same tumor type.

Ultrasensitive methods like next-generation sequencing /5/ are required to detect alterations (e.g., mutations or copy number changes).

28.4.2 Diagnosis of cancer

Tumor derived mutant DNA can be detected in the cell-free fraction of the blood in individual cancers. High concordance rates of 80 to 90% between cfDNA results in plasma and tissue samples are obtained. Non concordance of alterations is most often observed in patients with low ctDNA levels /1/. In a study with an optimal proportion of ctDNA /6/ no mutations were identified in cfDNA of healthy controls, whereas exactly half of the patients with metastatic breast cancer had at least one mutation or amplification in cfDNA across a total of 13 genes.

Application of cfDNA for detection of cancer in patients without clinical evident disease fails from low analytical sensitivity of ctDNA because of the low levels. In a study a high sensitive approach was used to assess tumor dynamics /7/. In patients with colorectal cancer ctDNA was detectable in all subjects before surgery, and serial blood sampling revealed oscillations in the ctDNA that correlated with the extend of surgical resection. Subjects who had detectable ctDNA after surgery generally relapsed within 1 year. The ctDNA seemed to be a much more reliable and sensitive indicator than the biomarker CEA.

28.4.3 Molecular profiling

Circulating cell free DNA has the potential to assess the molecular profile of a tumor in identifying genetic alterations. Because of the tumor heterogeneity cfDNA determination may identify concurrent resistance alterations residing in distinct tumor metastases. In a study /8/ molecular profiling of metastatic colorectal tumors using next-generation sequencing uncovered alterations beyond the well-characterized RAS/RAF mutations associated with anti-EGFR (epidermal growth factor receptor) resistance.

Molecularly targeted agents have been reported to have anti-tumor activity in patients whose tumors harbour the matching molecular alteration. Off-label use of molecularly targeted agents should be discouraged without enrollment in clinical trials /9/.

28.4.4 Monitoring of acquired resistance

Acquired resistance versus conventional therapy arises from the outgrowth of multiple resistant sub clones. The sub clones may reside in the primary tumor or distant metastases. Multiple unique resistance alterations frequently coexist in different metastases /10/. In the primary tumor all cells may harbour the original clonal target alteration, preexisting sub clonal alterations that provide a fitness advantage under the selective pressure of therapy may exist in some cells. As therapy is initiated, it may exert a cytotoxic effect on most tumor cells, but an outgrowth of resistant sub populations may occur, leading to dynamic shifts in clonal abundance and eventual disease progression /1/.

Circulating free DNA is shed from the tumor sites and resistance can be detected by the determination of ctDNA. In addition ctDNA can be used to monitor the dynamics of distant resistant sub clones /1/.


1. Corcoran RB, Chabner BA. Application of cell-free DNA analysis to cancer treatment. N Engl J Med 2018; 379: 1754–65.

2. Heitzer E. Circulating tumor DNA for modern cancer management. Clin Chem 2020; 66 (1): https://doi.org/101373/clin chem.2019.304774.

3. Swaminathan R, Butt AN. Circulating nucleic acids in plasma and serum: recent developments. Ann N Y Acad Sci 2006; 1075: 1–9.

4. Schwarzenbach H, Hoon DSB, Pantel K. Cell-free nucleic acids as biomarkers in cancer patients. Nature Reviews Cancer 2011; 11: 426–37.

5. Adams CR, Eng CM. Next generation sequencing to diagnose suspected genetic disorders. N Engl J Med 2018; 379: 1353–62.

6. Page K, Guttery DA, Fernandez-Garcia D, Hills A, Hastings RK Luo J, et al. Next generation sequencing of circulating cell-free DNA for evaluating mutations and gene amplification in metastatic breast cancer. Clin Chem 2017; 63: 532–41.

7. Diehl F, Schmidt K, Choti MA, Romans K, Goodman S, Li M, T et al. Circulating mutant DNA to assess tumor dynamics. Nature Medicine 2008; 14: 985–90.

8. Gong J, Cho M, Sy M, Salgia R, Fakih M. Molecular profiling of metastatic colorectal tumors using next-generation sequencing; a single-institution experience. Oncotarget 2017; 8: 42198–213.

9. Le Tourneau C, Delord JP, Goncalves A, Gavoille C, Dubot C, Isambert N, et al. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised controlled phase 2 trial. Lance Oncol 2015; 13: 1324–34.

10. Garraway LA, Jänne PA. Circumventing cancer drug resistance in the era of personalized medicine. Cancer Discov 2012; 2: 214–26.

28.5 Micro ribonucleic acids (miRNAs)

Lothar Thomas

Micro ribonucleic acids (miRNAs) are small single-stranded non-coding RNAs of 19 to 25 ribonucleotides in length, that are encoded in genomes. In mammals, they account for 1–3% of the genome. Transcription by RNA polymerase II from a DNA helix generates primary precursor miRNA which is transported into the cytoplasm of the cell and there transformed to miRNA. Each miRNA is assigned a specific prefix (hsa for humans) and a numerical identifier.

The normal functions of miRNAs include regulation of cell differentiation, cell cycle progression and apoptosis. They down regulate their target genes by either inducing messenger RNA (mRNA) degradation or by translation inhibition at the post-translational level /1/.

The expression of miRNA can be increased or reduced in cancer. Dysregulation of miRNA expression has an influence on tumorigenesis with some to behave as oncogen and others as tumor suppressor gene. The binding to 3’untranslated region (3’UTR) of target mRNAs through base-pairing results in degradation or transcriptional inhibition and possible negative regulation of gene expression.

In general, the dysregulation of miRNA expression contributes to loss of control and development of malignancy. Reversely, malignancy related changes in miRNAs can facilitate the detection and characterization of cancer development, metastatic processes and therapy /2/. A summary of tumor associated miRNAs as potential biomarkers for cancers has been published in Ref. /3/.

miRNA detection

Analysis for miRNA from tissue is performed in cell lines, tissues serum and plasma using Northern blotting, in situ hybridization methods, hybridization based micro array platforms and individual determination by quantitative reverse transcriptase-polymerase chain reaction (RT-PCR).

Clinical significance

Key sources of circulating RNA include circulating tumor cells (CTCs) and tumor cell exosomes. The exosomes are actively secreted vesicles and transmit information between cells. Besides proteins and enzymes exosomes contain DNA and RNAs. Cancer cells secrete excessive amounts of exosomes compared to normal cells. The interplay via exchange of exosomes between cancer cells and the tumor stroma may promote the transfer of miRNA from one cell to another, leading to the reprogramming of the recipient cell /4/. Tumor vesicles can be used as diagnostic markers. Aberrant miRNA levels in exosomes reflect the physiological state of cancer cells and can be detected by miRNA expression profiling and harnessed for the purpose of diagnosis and prognosis /5/.

Altered miRNA expression profiles have been identified in several malignancies /13, 67/.


1. Nugent M. MicroRNA function and dysregulation in bone tumors: the evidence to date. Cancer Manag Res 2014; 6: 15–25.

2. Ul Hussain M. Micro-RNAs:genomic organisation, biogenesis and mode of action. Cell Tissue Res 2012; 349: 405–13.

3. Bouyssou JM, Manier S, Huynh D, Issa S, Roccaro AM, Ghobrial IM. Regulation of microRNAs in cancer metastasis. Biochim Biophys Acta 2014; 1845: 255–65.

4. Kharaziha P, Ceder S, Li Q, Panaretakis T. Tumor cell-derived exosomes: a message in a bottle. Biocim Biophys Acta 2012; 1826: 103–11.

5. Lujambio A, Lowe SW. The microcosmos of cancer. Nature 2012; 482: 347–54.

6. Wang J, Zhang KY, Liu SM, Sen S. Tumor-associated circulating microRNAs as biomarkers of cancer. Molecules 2014; 19: 1912–38.

7. Wei T, Ye P, Peng X, Wu LL, Yu GY. Prognostic value of miR-22 in various cancers: a systematic review and meta-analysis. Clin Lab 2016; 62: 1387–95.

28.6 Tumor markers

Lothar Thomas

Tumor markers are macromolecular substances circulating in the blood and other body fluids. Their presence and changes in concentration correlate with the growth of malignant tumors. Tumor markers have carbohydrate or lipid portions and are expressed in or on tumor cells. Their formation can also be induced by other cells. They are released to the blood and other body fluids as circulating molecules. Markers formed by tumor cells may be onco fetal antigens, carbohydrate epitopes detectable with monoclonal antibodies, enzymes, isoenzymes, oncogenic products and receptors /1/.

28.6.1 Classification of tumor markers

  • Onco fetal and onco placental antigens (e.g., CEA, AFP, hCG)
  • Carbohydrate epitopes detectable with monoclonal antibodies (e.g., CA 19-9, CA 125, CA 15-3)
  • Differentiation and proliferation antigens (e.g., NSE, PSA, β2-microglobulin)
  • Ectopically produced hormones (e.g., ACTH in lung cancer, calcitonin in thyroid cancer)
  • Ectopically produced proteins (e.g., monoclonal immunoglobulin or free monoclonal light chains in multiple myeloma).

Clinically important tumor markers are listed in Tab. 28.6-1 – Clinically important tumor markers.

28.6.2 Variables of tumor marker concentration

The concentration and/or changes in concentration of tumor markers in blood and other body fluids are subject to the following variables /2/:

  • Total amount of marker producing cells (i.e., mass, spread and stage of a tumor)
  • Rate of tumor marker synthesis
  • Rate of tumor marker release from the tumor cell or cell surface
  • Marker expression; an increase in tumor marker concentration will not occur unless this marker is expressed by an individual tumor
  • The tumor is a non secretor type tumor; the marker is expressed by the tumor cell, but not released to body fluids
  • Blood supply to the tumor; if the supply is poor, less tumor marker is released to the circulation
  • Degree of tumor necrosis; strong tumor necrosis causes an over proportionate increase in tumor marker compared to tumor size in patients receiving cytostatic and/or irradiation treatment
  • Rate of tumor marker degradation; excretory dysfunction (e.g., due to renal insufficiency, hepatic dysfunction or cholestasis) causes an increase in tumor marker concentration
  • The effect of antibodies can cause the formation of immune complexes with a size dependent elimination rate.

28.6.3 Indication

  • Screening (restricted to prostate, ovarian and pancreatic cancer)
  • Risk groups (liver cirrhosis, germ cell tumor and thyroid cancer)
  • Cancers of unknown primary
  • At the time of primary diagnosis of a tumor
  • Approximately several days or weeks after therapy for baseline determination
  • Differential diagnosis (e.g., indeterminate liver or lung tumors)
  • Prognosis
  • Monitoring and assessment of therapy of a tumor.

28.6.4 Reference interval

The threshold value is the upper reference interval value of a tumor marker in healthy individuals or in patients with benign diseases and is expressed as the 95th or 97.5th percentile. In the presence of a tumor, the upper reference interval has no significance in the further course of the disease.

28.6.5 Clinical significance

A tumor marker concentration within the reference interval does not exclude malignancy; the presence of a tumor is not indicated unless values are well above the reference interval /12, 34/. Screening using tumor markers

All tumor markers are physiologically present in the blood, but since their release and elimination vary individually, each individual has his or her own typical baseline. The baseline values of healthy individuals overlap with those of patients with non malignant diseases and/or cancer. Therefore, screening using tumor markers is not generally recommended and should be limited to a few diagnoses. Moreover, tumor markers are not suited for monitoring risk patients and controlling patients with familial clustering of tumors (e.g. familial polyposis coli, postmenopausal ovarian cancer) /1/.

In the presence of cancer of unknown primary, the ESMO guidelines and American NCCN guidelines recommend that tumor markers be determined to search for the site of origin /5/:

  • hCG (choriocarcinoma) and AFP (ovarian germ cell tumor) should be measured in women with mediastinal tumors and CA 125 (ovarian cancer) in women with inguinal lymph nodes or peritoneal disease
  • AFP and hCG (germ cell tumor of the testes) and PSA (prostate cancer) should be measured in men. Significance of tumor markers in differential diagnosis

In suggestive of a tumor based on imaging techniques, tumor markers can provide additional detailed information for the differential diagnosis /5/:

  • Benign nodules in the liver may be due to metastasis or a primary tumor (hepatocellular carcinoma, cholangiocellular carcinoma). AFP concentrations above 1,000 μg/L very likely indicate hepatocellular carcinoma if germ cell tumor has been excluded. Liver metastasis of an adenocarcinoma (colon, breast) is likely in CEA levels above 50 μg/L
  • Pulmonary tumors of unclear dignity detected with imaging techniques can be benign or the metastases of another primary tumor or lung cancer. ProGRP concentrations above 500 ng/L point to small cell lung cancer.
  • HER-2 levels ≥ 80 μg/L point to breast cancer
  • S100 concentrations above 1 μg/L are likely associated with metastatic malignant melanoma. Determination of tumor markers before primary cancer therapy

Following the primary diagnosis of a malignancy, it is important to determine the concentration of the released tumor marker before therapy starts. The objective is to select a useful marker from the large number of possible markers. It is usually not very useful to determine several markers. Useful markers are listed in Fig. 28.6-1 – Use of tumor marker combinations and Tab. 28.6-2 – Indication for tumor marker determination.

The following information can be obtained from the tumor marker release pattern /1/:

  • The tumor marker expressed at the time of primary diagnosis is very likely also the relevant marker for follow-up and serves as the baseline for post therapy monitoring
  • Prognosis is possible; high concentrations indicate the presence of previously undetected distant metastasis. They can be used to predict the expected course of the disease and help to make the decision for treatment (surgery, conservative or aggressive therapy).
  • hCG, AFP and the enzyme LD must be determined in germ cell tumors because these values are used for tumor staging. Monitoring of post-treatment effectiveness

The pattern of tumor marker concentrations after surgery and/or radio- or chemotherapy provides information on post treatment effectiveness (Tab. 28.6-3 – Biological half life and threshold values of tumor markers). After one month, most tumor markers decline by at least four half lives, reaching baseline concentration in patients who fully responded to treatment (Tab. 28.6-2 – Indication for tumor marker determination). Generally, the following applies:

  • Continuously declining concentrations indicate that therapy is effective
  • Constant concentrations point to stable disease
  • Increasing concentrations indicate non response and the need to change the treatment concept
  • Elevated tumor marker concentrations are found following manipulations of the tumor and/or in the first days after the start of radio- or chemotherapy.

Recurrence-free interval, remission

The objective of surgery is the complete removal of the tumor (R0 resection), achieving normalization of the high pre surgery tumor marker level to an individual baseline. The baseline should be below the 95th percentile of healthy controls or even within the median interval. Values within this range merely showing methodology related variations indicate a recurrence free interval, but can never exclude concomitant clinical progression of the disease.

Residual tumor and ensuing tumor progression

Tumor marker concentrations above the 95th percentile of healthy controls are characteristic of the presence of a residual tumor after resection. After a short time interval, a continuous, strong increase in concentration can be observed. Tumor marker levels following primary therapy

During follow-up for approximately 1 month after primary treatment (surgery, chemotherapy or radiotherapy), the tumor marker concentration is an important indicator as the patient’s individual baseline value. The tumor marker concentration measured during this period is the individual standard value of a patient for the following years of monitoring. A percentage increase during monitoring is a sensitive parameter of recurrence and requires further investigation (clinical symptoms, imaging techniques). For instance, a 100% increase in CEA or CA 15-3 is a tumor specific criterion and indicates recurrence or the presence of metastasis or another cancer /1/.

After primary treatment, monitoring is only guided by the individual baseline value; the upper reference interval value of the tumor marker is irrelevant. The tumor marker must also be determined for monitoring if the pretreatment concentration was within the reference interval. Tumor markers during follow- up

Tumor markers are determined within defined time intervals to detect recurrence or progression before they manifest clinically. An increase in tumor marker concentration following seemingly curative surgery indicates tumor recurrence or distant metastasis. However, the options for curative treatment decrease with increasing lead time.

Lead time

After treatment, the lead time plays an important role in the early detection of tumor recurrence. It refers to the time interval between the initial increase in tumor marker and the clinical or imaging based detection of tumor recurrence and/or metastasis. The following distinction is made:

  • Positive lead time where clinical symptoms or imaging based tumor detection are preceded by an increase in marker concentration (may be 4–26 months)
  • Negative lead time where an increase in marker concentration is preceded by clinical diagnostics. High diagnostic sensitivity and a high positive predictive value of a tumor marker are not useful in such a situation.

The lead time depends on the timing of sample collection. For efficient follow-up, sampling must be performed at intervals of 6–8 weeks.

Tumor free period and ensuing tumor progression

A recurrence free interval is followed by an initially slow, continuous increase in tumor marker. The increase may remain within the reference range across several sampling intervals and is as significant as a concentration above the upper reference interval value.

A continuous increase indicates disease progression. In the event of a one time increase, even a pronounced one, without any associated clinical symptoms, no invasive or therapeutic action should be taken. The increase should always be verified by short-term follow-up analysis.

Tab. 28.6-4– Tumor markers in cancers show cancers and the practical use of tumor markers.

28.6.6 Comments and problems

Method of determination

Tumor marker concentration depends on the individual kit used. Kits from different manufacturers measure different values even if the same methodology and the same antibodies are used. Therefore, the kit manufacturer must be specified along with a result.

Within subject variation

The intraindividual variation in tumor markers is higher in cancer patients than in healthy individuals and amounts to /9/ 8.4% in healthy individuals and 19.3% in tumor patients for CEA and 6.0% in healthy individuals and 17.3% in tumor patients for CA 15-3.

Interference factors

The following must be observed:

  • A time interval > 60 min. until serum separation results in an increase in NSE due to NSE release from the platelets; the time interval for the determination of free PSA should not exceed 3 h
  • Skin contact with the inside of specimen cups results in an increase in SCCA
  • Contamination of the sample with saliva causes an increase in the concentrations of SCCA, CA 19-9 and, to a lesser extent, also CEA
  • Hemolysis increases the NSE values due to release of NSE from erythrocytes and platelets
  • Jaundiced serum results in higher PSA values
  • Drugs, for example high concentrations of bivalent and trivalent metal ions as well as purine, indole, guanidine analogs (Isoket, Isoptin), vitamin C, cisplatin, mitomycin, estradiol or epirubicin can lead to falsely high PSA values
  • Human anti-mouse IgG antibodies (HAMA) form in patients who received mouse immunoglobulins for diagnostic or therapeutic reasons within the scope of immunoscintigraphy or immunotherapy. This results in a false positive signal in test systems using monoclonal mouse antibodies. The heterophilic anti-immunoglobulin antibodies also occur in patients following live cell therapy, thus simulating falsely high tumor marker concentrations.

Influencing factors

The following must be observed:

  • Age; the tumor marker value is influenced by age. For instance, in a study /10/ subjecting healthy individuals aged 66–99 years to routine determination of the tumor markers CA 19-9, CEA, CA 72-4, CA 15-3, AFP and PSA, 40% were found to have at least one elevated marker level
  • Renal insufficiency or cholestasis can cause an increase in tumor markers due to reduced excretion. Consistently elevated levels of the tumor markers are characteristic in such cases.
  • Depending on the consumption of cigarettes, CEA can be 10 μg/L and in rare cases even 20 μg/L.


1. Stieber P, Heinemann V. Sinnvoller Einsatz von Tumormarkern. J Lab Med 2008; 32: 339–60.

2. Lamerz R. Tumormarker. Dtsch Med Wschr 1984; 109: 1219–20.

3. Wolter C, Luppa P, Breuel J, Fink U, Hanauske AR, Präuer HW, Sendler A, Wilhelm O, Neumeier D. Humorale Tumormarker. Praxisorientierte Vorschläge für ihren effizienten Einsatz. Dt Ärztebl 1996; 93: A-3286–52.

4. Zhou Q, Hu H, Hou L. Discover, develop & validate- advance and prospect of tumor biomarkers. Clin Lab 2015; 61: 1589–99.

5. Savage P. Tumour markers in cancers of unknown primary: a clinical perspective. Ann Clin Biochem 2006; 43: 1–2.

6. Yarchoan R, Uldrick TS. HIV-associated cancers and related diseases. N Engl J Med 2018, 378: 1029–41.

7. Hermanek P, Sobin LH. UICC: TNM classification of malignant tumors. Heidelberg: Springer, 1987.

8. Lamerz R, Reithmeier A, Stieber P, Eiermann W, Fateh-Moghadam A. Role of blood markers in the detection of metastases from primary breast cancer. Diagn Oncol 1991; 1: 88–97.

9. Dittadi R, Peloso L, Gion M. Within-subject biological variation in disease: the case of tumour markers. Ann Clin Biochem 2008; 45: 226–8.

10. Lopez LA, del Villar V, Ulla M, Fernandez F, Ferandez LA, Sandoz I, et al. Age and Ageing 1996; 25: 45–50.

11. Sturgeon CM, Duffy MJ, Stenman UH, Lilja H, Brünner N, Chan DW, et al. National academy of clinical biochemistry laboratory medicine practice guidelines for use of tumor markers in testicular, prostate colorectal, breast and ovarian cancers. Clin Chem 2008; 54: e11–e79.

12. Locker GY, Hamilton S, Harris J, Jessup JM, Kemeny N, Macdonald JS, et al. ASCO 2006 update recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 2006; 24: 5313–27.

13. European Group on Tumour Markers (EGTM): Consensus recommendations. Anticancer Res 1999; 19: 2785–820.

14. Nolen BM, Lokshin AE. Screening for ovarian cancer: old tools, new lessons. Cancer Biomarkers 2011; 8: 177–86.

15. Germa-LLuch JR, Garcia del Muro X, Maroto P, Paz-Ares L, Arranz JA, Guma J, et al. Clinical pattern and therapeutic results achieved in 1490 patients with germ-cell tumours of the testis: the experience of the Spanish Germ Cell Cancer Group. Eur Urol 2002; 42: 553–62.

16. Holdenrieder S, Stieber P. New challenges for laboratory diagnostics in non-small cell lung cancer. Cancer Biomarkers 2010; 6: 119–21.

28.7 Alpha-fetoprotein (AFP)

Rolf Lamerz

AFP is one of the few tumor markers recommended for the screening of malignancy and used for therapy monitoring and follow-up.

28.7.1 Indication

  • Suspected hepatocellular carcinoma (e.g., in patients with liver cirrhosis)
  • Diagnosis and differential diagnosis of germ cell tumors (testes, ovaries, extra gonadal location)
  • Therapy monitoring and follow-up of patients with AFP positive germ cell tumors or primary hepatocellular carcinoma (e.g., post surgery or during or after radiotherapy or chemotherapy).

28.7.2 Method of determination

Immunoassays such as enzyme linked immunoassays and immunometric assays using enzyme labeled, fluorescence labeled or luminescence labeled tracers. Monoclonal antibodies are usually the primary antibody in assays compared to polyclonal antibodies.

28.7.3 Specimen

Serum, pleural exudate, ascitic fluid, cerebrospinal fluid: 1 mL

28.7.4 Reference interval

Serum/plasma: up to 10 μg/L (about 7 IU/mL)*

* After the 1st year of life.

28.7.5 Clinical significance

The determination of AFP is relevant for the screening and monitoring of risk groups with suspected development of primary hepatocellular carcinoma or germ cell tumor.

AFP can be elevated in benign and malignant disease (Tab. 28.7-1 – Elevated serum AFP concentrations) /12/. Liver cirrhosis

Patients with liver cirrhosis often show low abnormal AFP levels with constant or transitory increases during the course of the disease. The positivity rates are 10–62% /2/. Elevated mean concentrations are 15–100 μg/L in 17% of the patients, > 500 μg/L in 20% and > 500 μg/L in 1%. According to long term studies, liver cirrhosis patients with elevated AFP are at greater risk for the development of hepatocellular carcinoma /3/. Viral hepatitis

Elevated AFP concentrations are found in acute and chronic active viral hepatitis and are transitory in acute viral hepatitis. The overall rate of abnormal values is 31%, but only 1% of these 31% are higher than 500 μg/L /2/. In acute viral hepatitis, there is a temporal correlation between AFP and the peak values of aminotransferases. The aminotransferase peaks (necrotic stage) precede the AFP peak (regeneration stage) by two weeks. AFP returns to normal after 6–10 weeks. In fulminant viral hepatitis, elevated AFP has a favorable prognostic significance (regeneration). Elevated AFP is significantly correlated with lower albumin levels and genotype 1b chronic hepatitis C and patients with advanced fibrosis, have a poorer prognosis /4/. Other liver diseases

Increased AFP is rare in toxic liver disease, hepatitis caused by non hepatotrophic viruses, biliary atresia, cholestasis, hereditary liver cirrhosis and metabolic disease with liver involvement. Hepatocellular carcinoma (HCC)

In Europe, North America and Japan, HCC occurs mainly in people with preexisting liver cirrhosis. In regions with a high prevalence (Asia and Africa: 30–40/100,000 in males and 10–14/100,000 in females), primary liver cancer more frequently affects people < 40 years of age than in regions with low prevalence. In Europe, North America and Canada, the prevalence in men is 2–8/100,000 while in women it is 1–4/100,000; there is a predilection for the male sex (male/female = 1.4–3.0) and incidence rates in general are increasing /5/.

As a result of improved imaging techniques, primary diagnosis of HCC is now made at a smaller tumor size and only about 60% of tumors will be associated with elevated serum AFP levels at the time of diagnosis. In about 20% of these cases, AFP levels are > 10 mg/L with maximal concentrations of up to 10,000 mg/L, in 32% they are > 1 mg/L, and in about 50% > 100 μg/L.

Histologically confirmed HCC of significant size and extent will show normal AFP concentrations in 5–10% of cases. In the much rarer hepatoblastomas, normal or elevated levels are observed whereas cholangiocellular cancers are AFP negative.

While the AFP concentration and other biomarkers or clinical signs of disease do not correlate with tumor size, tumor growth rate, stage or degree of malignancy, there is a correlation with the AFP content of the tumor and the rise in undifferentiated cells within the tumor /6/.

AFP determinations are considered to be very useful for monitoring patients who are at risk (liver cirrhosis, carriers of HBsAg) and for the early detection of HCC /7/. Patients with liver disease positive for AFP show a higher rate of developing HCC as well as a poorer prognosis at 5-year follow-up /3/.

According to the tumor marker (TM) guideline of the National Academy of Clinical Biochemistry (NACB) of 2010 /8/, AFP is currently the only marker that can be recommended for clinical use in HCC and should be measured in screening of patients at high risk for HCC (liver cirrhosis, chronic hepatitis B/C) and ultrasound should be performed at 6-month intervals. AFP concentrations > 20 μg/L are suspicious and should prompt further investigation even if ultrasonography is negative.

In patients at risk for HCC, sustained increases in serum AFP may be used in conjunction with ultrasonography to aid early detection, while tumor sizes < 1 cm and/or 1–2 cm should be investigated by CT/MR imaging modalities and biopsy may be considered. If lesions are > 2 cm in size, AFP is > 200 μg/L and the ultrasound appearance is typical of HCC, biopsy is not necessary.

For HCC treatment monitoring and follow-up, measurement of AFP is recommended every 3 months for 2 years and then every 6 months to assess disease status after liver resection or liver transplantation for detection of recurrence or after ablative therapies and application of palliative treatment. Germ cell tumors

The synthesis of AFP, as observed in germ cell tumors of the testes, ovaries or of extra gonadal location (sacrococcygeal, mediastinal, intracranial), can be traced back to yolk-sac entodermal like structures based on the distribution of AFP synthesizing cells in the fetus /9/. This has had an impact on the histological classification especially in testicular germ cell tumors (incidence: 4–6/100,000 male population/year) and in conjunction with improved therapy (cisplatin, bleomycin, etoposide).

Besides the three customary histological nomenclatures (UK, USA, WHO) /10/, the histogenetic classification by Teilum /9/ also includes the two tumor markers AFP and hCG. According to this classification, the primordial germ cell may develop into germinomas (seminomas, dysgerminomas; always AFP negative) and embryonic carcinomas (AFP/hCG negative or positive). These may undergo the following types of differentiation:

  • Extraembryonic (trophoblast choriocarcinoma: AFP negative/hCG positive, and/or yolk sac-entodermal sinus tumor: AFP positive/hCG negative)
  • Embryonic (mature/immature teratoma: AFP/hCG negative).

Based on this, the following conclusions can be made (Fig. 28.7-1 – Differentiation between germ cell tumors by AFP and hCG determination):

  • Pure seminomas, dysgerminomas and differentiated teratomas are always AFP negative
  • Pure yolk sac tumors are always AFP positive
  • Embryonic carcinomas and combination tumors may be positive or negative depending on the amount of entodermal structures present.

The diagnostic sensitivity of AFP in non seminomatous tumors of the testes is 50–80%. For instance, elevated AFP values were observed /11/: in 70–72% of patients with undifferentiated malignant teratoma (MTU), in 60–64% of patients with intermediate malignant teratoma (MTI), in 64% of patients with yolk sac or combination tumors, but in none of 130 patients with seminoma. High concentrations above 1,000 μg/L are found in 53% of MTU cases, 16% of MTI cases and 26% of combination tumors /12/.

Depending on the clinical stage, the prevalence of AFP elevations in association with malignant teratomas and combination tumors is: 76% in stage I, 63% in stages II–III and 81% in stage IV. The diagnostic specificity is 100% for pure seminomas and healthy individuals. According to other authors /13/, the combined determination of the markers AFP and hCG has a diagnostic sensitivity of 86% for detecting tumor recurrences and partial responses, in conjunction with 100% diagnostic specificity as well as positive and negative predictive values of 100% and 87%, respectively. In retroperitoneal lymphadenectomy, staging errors are reduced from about 50% to less than 15% in stages I and II by employing AFP and hCG determinations /11/.

According to the TM guideline of the National Academy of Clinical Biochemistry (NACB) of 2008 /14/ and the International Germ Cell Cancer Collaborative Group /15/, measurement of hCG, AFP and the enzyme LD is mandatory for staging and risk stratification. If raised before therapy, AFP, hCG and/or LD should be monitored weekly until concentrations are within the reference interval. Wherever possible, the marker half-life should be determined. Marker levels exceeding the upper reference interval after therapy suggest residual disease, which should be confirmed or excluded by other methods. Serial monitoring with AFP, hCG and LD is recommended even when these are not raised before therapy because marker expression can change during therapy.

Cerebrospinal fluid (CSF) AFP determination

AFP and hCG determinations in the CSF are valuable for diagnosing and monitoring intracranial germ cell tumors /16/. Serum AFP is often not elevated in these cases. Other tumors

Elevated AFP concentrations are measured in rare cases of non hepatic gastrointestinal tumors in 21% (gastric, colonic, biliary and pancreatic cancers). AFP elevations are very rarely observed in non gastrointestinal tumors (e.g., lung cancer usually in association with liver metastasis) /2/. The levels mainly fall within a range < 500 μg/l, concentrations > 500 μg/L are only observed in about 4% of cases. Monitoring of the clinical course of AFP

Along with increasing tumor spread, untreated tumors (HCC, germ cell tumors) lead to an initially slow, then exponential rise in the AFP level which, in HCC, may reach values of 100–1,000 mg/L or more. This rise, especially during the preterminal stage, is not necessarily correlated with tumor growth but, instead, may be disproportionately high, due to disorders of catabolism (liver failure). However, in cases without therapy, preterminal decreases in the AFP concentration have also been observed (necrosis). Surgical tumor reduction or tumor resection

In tumors composed of one or more cell types of which one produces the tumor marker, the decrease in AFP concentration reflects the amount of tumor tissue which was resected. Thus, in complete tumor resection, after a frequently encountered, transient postoperative increase due to surgical tumor manipulation, the AFP concentration (with a half-life < 5 days) decreases to a level within the reference interval (Tab. 28.7-2 – AFP concentrations during therapy). Irradiation and chemotherapy

A transient increase in AFP concentration may occur due to the release of AFP in conjunction with acute tumor cell destruction and the tumor lysis syndrome.

The AFP concentration profile is dependent on the composition of the tumor. In tumors with a homogeneous cell population (HCC) with widespread capacity for AFP synthesis, the decrease in the serum AFP concentration reflects the behavior of the entire tumor. AFP response to treatment

In tumors with a heterogeneous cell population (combination tumors of the testis), the decrease in AFP exclusively represents the response of the marker producing cell type; this may differ from that of the other cell types. This is the reason why it is mandatory to use various tumor markers for monitoring mixed-cell tumors (e.g., AFP and hCG in testicular tumors). Furthermore, in testicular tumors, it is also possible that despite concordant normalization of both tumor markers (AFP and hCG), residual tumor and/or progressive disease may be detected later on by other methods, (e.g., CT or X-ray). This phenomenon is occasionally observed in patients undergoing chemotherapy and is caused by transformation of the histological type (e.g., transition into a mature, marker-negative teratoma).

There is general agreement that increasing levels of one or several tumor markers are related with a tumor non responding to treatment and that they often indicate progressive disease with a lead time of 1–6 months prior to the detection by other methods, thus possibly warranting a change in the therapeutic approach. Prognostic significance of AFP

The determination of the half life of AFP and hCG /17/ is recommended for monitoring purposes; it is considered to be a prognostically favorable sign if found to be within the physiological range (less than 5 days).

Patients with a plasma half life > 7 days for AFP or > 3 days for hCG have a significantly lower survival rate after two cycles of chemotherapy than patients showing normal tumor marker clearance. It has been suggested that this group of patients at high risk should receive more aggressive chemotherapy /1819/.

The International Germ Cell Cancer Collaboration Group issued the International Germ Cell Consensus Classification scheme which incorporates prognostic factor based staging for metastatic germ cell tumors /15/. For non seminomas, it includes:

  • Three prognostic groups (good, intermediate, poor) for primary testicular tumor, retroperitoneal or mediastinal tumor without or with non pulmonary visceral metastasis
  • Three interval groups of increasing AFP concentration AFP (< 1,000; 1,000–10,000; > 10,000 μg/L), hCG (< 5000; 5000–50,000; > 50,000 IU/L) and LD (< 1.5-fold; > 1.5–10-fold; > 10-fold the upper reference interval). The rates for progression free survival or overall survival are: good 89/92%; intermediate 75/80%; poor 41/48%) /15/.

The AFP levels measured in 2253 liver transplantation patients (stratified by < 20; 20–399; ≥ 400 μg/L) before therapy were a prognostic predictor for survival following transplantation /20/.

Early AFP response to localized chemotherapy and radiotherapy in patients with advanced HCC correlated with significantly improved progression free survival and overall survival in AFP responders /21/. Early AFP response (decline > 20% from baseline after 2 to 4 weeks of treatment) predicts treatment efficacy of anti-angiogenic systemic therapy in patients with advanced HCC.

AFP responders, compared with non responders, also had longer median progression free survival and longer median overall survival and AFP response remained a significant independent predictor of better progression-free survival and median overall survival /22/.

28.7.6 Comments and problems

Method of determination

Depending on the immunoassay employed, the detection limit is 0.1–1 μg/L. A WHO standard is available for calibration of the assays (AFP 72/225). The international unit is recommended for the reporting of results (1 IU = 1.21 ng).

Reference interval

From the 10th gestational week, pregnant women show elevated serum AFP depending on the gestational weeks (Section 38.4 – Alpha-fetoprotein). Serum AFP peak values are reached between the 32nd and 36th gestational week (maximally 400–500 μg/L); concentrations are 40–250 μg/L before birth and decrease during delivery followed by a further decline with a half life of 3.8 ± 0.9 days until normalization /23/.

In newborns, AFP decreases physiologically from a mean cord serum concentration of 70 mg/L with a half life of 4.0 ± 1.8 days, reaching very variable levels of 500–4000 μg/L after 2–3 weeks and finally adult values after about the 10th month of life /23/.


Samples are stable up to 1 week if stored in the refrigerator; longer storage at –30 °C /24/ has been observed /25/.

28.7.7 Pathophysiology

Human AFP is a glycoprotein (4% carbohydrate content) with a molecular weight of about 70 kDa and with α1-electrophoretic mobility /26/.

During fetal life, AFP is synthesized in the gastrointestinal tract, the liver and the yolk sac and released into the fetal blood and other body fluids. Depending on gestational age and the composition of the placental barrier, AFP reaches the maternal serum via the transplacental route.

The peak fetal serum AFP concentrations of 300–400 mg/L are observed during the 13th–15th gestational week, the concentrations in the amniotic fluid (maximally up to 35 mg/L around the 16th gestational week) and in maternal serum (with peaks of about 500 μg/L in the 32nd and 36th gestational week) are significantly lower.

Discussion continues regarding the physiological roles of AFP during fetal life; they include protection of the fetus against maternal estrogens or against immunological rejection and as a substitute for albumin until its concentration increases later in gestation.

At birth, the mean AFP levels in cord blood are 70 mg/L. Subsequently, in the following weeks of life, there is a gradual decline in the concentration of AFP, marked by large, individual fluctuations, at a half life rate of approximately 4 days. As a result of this, normal adult AFP levels < 15 μg/L are not reached until the 10th month of life.

During adulthood, elevated AFP concentrations of constant or transient nature occur in benign liver diseases and regenerative liver processes; highly abnormal levels are associated with the development of hepatocellular carcinoma or germ cell tumor of the testes, ovaries or of extra gonadal localization.

The detection of AFP levels in serum comparable to those observed during fetal life is explained by derepression of genomes responsible for the synthesis of AFP, which are normally repressed at birth.

In hepatocellular carcinoma, irreversible changes of certain hepatocyte precursors or persistent hepatoblasts, among other factors, are thought to be responsible for the synthesis of AFP whereas, in germ cell tumors, cells with the capability of AFP synthesis derive from the yolk sac epithelium (entodermal sinus structures).


1. Grob PJ, Dati F, Joller-Jemelka HI. Diagnostische Relevanz von Alpha-Fetoprotein in der Onkologie. Laboratoriumsblätter 1982; 2: 59–68.

2. Lamerz R. Die klinische Bedeutung von Alpha-Fetoprotein bei Lebererkrankungen. Bayer Internist 1980; 1: 19–23.

3. Harada T, Shigata K, Noda K, Fukumoto Y, Nishimura N, Mizuki M, Takemoto T. Clinical implications of alpha-fetoprotein in liver cirrhosis: 5 years’ follow-up study. Hepato-Gastroenterol 1980; 27: 169–75.

4. Chu CW, Hwang SJ, Luo JC, Lai CR, Tsay SH, Li CP, Wu JC, Chang FY, Lee SD. Clinical, virologic, and pathologic significance of elevated serum alpha-fetoprotein levels in patients with chronic hepatitis C. J Clin Gastroenterol 2001; 32: 240–4.

5. Stevens RG, Merkle EJ, Lustbader ED. Age and cohort effects in primary liver cancer. Int J Cancer 1984; 33: 453–8.

6. Lamerz R, Fateh-Moghadam A. Carcinofetale Antigene. I. Alpha-Fetoprotein. Klin Wschr 1975; 53: 147–69.

7. Geb KA, Chander G, Jenckes MW, Ghanem KG, Her-long HF, Torbenson MS, El-Kamary SS, Bass EB. Screening tests for hepatocellular carcinoma in patients with chronic hepatitis C: a systematic review. Hepatology 2002; 36: S84–S92.

8. Lamerz R, Hayes P, Hoffmann-RT, Löhe F, Taketa K. NACB medicine practice guidelines for use of tumor markers in liver cancer. Clin Chem 2010; 56 (6): e1–e18 (ref. 4-215, e35–e40).

9. Teilum G, Albrechtsern R, Norgaard-Pedersen B. The histogenetic embryologic basis for reappearance of alpha-fetoprotein in endodermal sinus tumors (yolk sac tumors) and teratomas. Acta Pathol Microbiol Scand, Sect A 1975; 83: 80–6.

10. Mostofi FK. Pathology of germ cell tumors of testis: a progress report. Cancer 1980; 45: 1735–41.

11. Javadpour N. The role of biologic tumor markers in testicular cancer. Cancer 1980; 45: 1755–61.

12. Mann K. Tumormarker beim Hodenkarzinom. Urologe [A] 1990; 29: 77–86.

13. de Bruijn HWA, Sleijfer DT, Schraffordt-Koops H, Suurmeiher AJ, Marrink J, Ockhuizen T. Significance of human chorionic gonadotropin, alpha-fetoprotein, and pregnancy-specific beta1-glycoprotein in the detection of tumor relapse and partial remission in 126 patients with nonseminomatous testicular germ cell tumors. Cancer 1985; 55: 829–35.

14. Stenman UH, Lamerz R, Looijenga LH. NACB medicine practice guidelines for use of tumor marker in testicular cancer. Clin Chem 2008; 54 (12): e12–e23, (ref. 26-117, e65–e67).

15. International germ cell cancer collaborative group. International germ cell consensus classification: a prognostic factor-based staging system for metastatic germ cell cancers. J Clin Oncol 1997; 15: 594–603.

16. Allen JC, Nisselbaum J, Epstein F, Rosen G, Schwartz MK. Alpha-fetoprotein and human chorionic gonadotropin determination in cerebrospinal fluid. J Neurosurg 1979; 51: 368–74.

17. Kohn J. The value of apparent half life assay of alpha-fetoprotein in the management of testicular teratoma. In: Lehmann FG (ed). Carcinoembryonic proteins. Amsterdam: Elsevier, 1979: 383.

18. Mazumdar M, Bajorin DF, Bacik J, Higgins G, Motzer RJ, Bosl GJ. Predicting outcome to chemotherapy in patients with germ cell tumors: the value of the rate of decline of human chorionic gonadotropin and alpha-fetoprotein during therapy. J Clin Oncol 2001; 19: 2528–41.

19. Bosl GJ, Chaganti RSK. The use of tumor markers in germ cell malignancies. Hematol Oncol Clin North Am 1994; 8: 573–87.

20. Mailey B, Artinyuan A, Khalili J, Deritz J, Sanchez-Luege N, Sun CL, et al. Evaluation of absolute serum alpha-fetoprotein levels in liver transplant for hepatocellular cancer. Arch Surg 2011; 146: 26–33.

21. Kim BK, Ahn SH, Seong JS, Park jY, Kim DJ, Kim JK et al. Early alpha-fetoprotein response as a predictor for clinical outcome after localized concurrent chemoradiotherapy for advanced hepatocellular carcinoma. Liver International 2011; 31: 369–76.

22. Shao AA, Lin ZZ, Hsu C, Shen YC, Hung C, Cheng AL. Early alpha-fetoprotein response predicts treatment efficacy of antiangiogenic systemic therapy in patients with advanced hepatocellular carcinoma. Cancer 2010; 116: 4590–6.

23. Lamerz R, Stein G, Belohradsky B, Kümper HJ, Wirtz A, Brandt A. Changes of alpha-fetoprotein levels during pregnancy and in the puerperium. In: Weitzel HK, Schneider J (eds). Alpha-Fetoprotein in clinical medicine. Stuttgart: Thieme, 1979: 78–83.

24. Pollard DR, Gupta K. Stability of alpha-fetoprotein in stored and frozen-thawed aliquots. Clin Biochem 1982; 15: 266–71.

25. Wu JT, Knight JA. In-vitro stability of human alpha-fetoprotein. Clin Chem 1985; 31: 1692–7.

26. Morinaga T, Sakai M, Wegmann TG, Tamaoki T. Primary structures of human alpha-fetoprotein and its mRNA (cDNA clones/three domain structures/molecular evolution). Proc Nat Acad Sci USA Biol Sci 1983; 80: 4604–8.

28.8 Gastrointestinal cancer antigen (CA 19-9, GICA)

Rolf Lamerz

CA 19-9 has the highest diagnostic sensitivity and specificity in the differential diagnosis of pancreatic cancer in patients with other gastrointestinal tumors. Moreover, the main diagnostic significance of C A19-9 determination is in the monitoring of treatment of pancreatic, hepatobiliary and gastric cancers.

28.8.1 Indication

Absolute indications

  • Suspected pancreatic, hepatobiliary (liver cancer, biliary cancer) or gastric cancer
  • Monitoring and follow-up of patients with these cancers.

Relative indications

  • Diagnosis and monitoring/follow-up of colorectal cancer (second line tumor marker after CEA) and ovarian cancer (second line tumor marker after CA 125).

28.8.2 Method of determination

Immunometric assay /1/ or enzyme linked immunoassay (ELISA) using the same monoclonal anti-CA 19-9 antibody as for capture and detection. The tests usually require two short incubation and washing steps, the measurement range is 5–1,000 U/mL.

28.8.3 Specimen

Serum, plasma, pleural exudate, ascitic fluid: 1 mL

28.8.4 Reference interval

Serum, plasma: ≤ 37 U/mL /1/

(1 U = 0.8 ng)

28.8.5 Clinical significance

CA 19-9 can be elevated in benign and malignant diseases (Tab. 28.8-1 – Diagnostic sensitivities of CA 19-9 in benign and malignant diseases). Benign disease

Given a threshold value of 37 U/mL, relatively high frequencies of elevated levels have been found in 20–30% of cases of cholecystitis, cholangitis as well as liver cirrhosis, mucoviscidosis and massive liver cell necrosis. Levels > 1,000 U/mL are frequently measured in benign obstructive jaundice. Other benign conditions with mildly or, more rarely, pronouncedly elevated CA 19-9 concentrations include spleen, liver, pancreas and bronchogenic cysts, benign multi cystic peritoneal mesothelioma, pulmonary fibrosis, diverticulitis and benign hydronephrosis and pyonephrosis /2/.

Elevated concentrations are only found in 0–6% of chronic inactive pancreatitis whereas in acute pancreatitis and during an acute episode of chronic pancreatitis, they are seen in 15–20% of cases, usually at values < 100 U/mL but up to a maximum of 500 U/mL /3/.

In order to better differentiate pancreatic cancer from benign diseases, a cutoff value of 100 U/mL is recommended as a cost saving initial test, rather than sonography, in patients presenting with weight loss and abdominal pain. Using this approach, the diagnostic sensitivity for the diagnosis of cancer is 62% at a specificity of 97% /45/. Pancreatic cancer

In excretory ductal pancreatic adenocarcinoma (incidence 8/100,000 population/year), the diagnostic sensitivity of CA 19-9 is 70–95% at a specificity of 72–90%, with maximum levels > 100,000 U/mL /136, 7, 8, 910/.

According to systematic reviews of 22 and 30 studies on patients with pancreatic cancer, CA 19-9 (cutoff 35–40 U/mL) has a median diagnostic sensitivity of 79% (70–90%) at a median specificity of 82% (68–91%) and a diagnostic sensitivity of 59–85% at a specificity of 60–100%, respectively /1112/. The diagnostic sensitivity of CEA is only approximately half as much /7/.

Given a prevalence of 58% for pancreatic cancer versus pancreatitis and a CA 19-9 cutoff value of 50 U/mL, the following data were obtained: diagnostic sensitivity of 81% at a specificity of 94%, positive and negative predictive values of 95% and 78%, respectively /39/.

The level and the incidence of abnormal CA 19-9 levels correlate with:

  • Tumor localization (cutoff 37 U/mL: pancreatic head 80%, body, tail 57%)
  • Tumor extent; cutoff value 120 U/mL: T2/3 33%, T + N1 71%, TN + M1 85%
  • Tumor diameter (cutoff value 37 U/mL): < 3 cm 57%, 3–6 cm 80%, > 6 cm 100%
  • Resectability and metastasis but not with histological differentiation /9/.

Studies on patients with pancreatic cancer confirm the prognostic value of CA 19-9 concentration before treatment and the significant role of CA 19-9 monitoring for the assessment of treatment outcome with and without imaging techniques /7810/:

  • In 142 patients with intraductal papillary mucinous neoplasm of the pancreas (IPMN), some 74% of patients with invasive IPMN had elevated levels of CA 19-9, compared with only 14% who had non invasive tumors. Given a cutoff of 37 U/mL, CA 19-9 had a diagnostic specificity of 85.9%, a negative predictive value of 85.9%, a positive predictive value of 74% and an accuracy of 81.7% /13/
  • Likewise, IPMN patients had a postoperative recurrence rate of 12.6% with a mean postoperative survival of 17 months in the recurrence group and 41.4 months in the non recurrence group. The independent risk factors of recurrence were invasive carcinoma, main location in the pancreatic head and baseline CA 19-9 concentrations above 38 U/mL /14/
  • In other IPMN patients, CA 19-9 values > 37 U/mL after surgical intervention were associated with an advanced stage of the disease. It was shown that the CA 19-9 level had good predictive value for malignant or invasive IPMN and postoperative survival. Concentrations > 63.6 U/mL indicated poor postoperative prognosis and disease specific recurrence /15/. Liver cancer and biliary cancer

The diagnostic sensitivity of CA 19-9 is reported to be 22–51% for hepatocellular and cholangiocarcinoma whereas for biliary cancer it was 55–79% /3/.

A CA 19-9 cutoff of 100 U/L was utilized for the distinction between patients with cholangiocarcinoma without primary sclerosing cholangitis (PSC) and those with benign liver disease and biliary obstruction at a diagnostic sensitivity of 53% and a specificity of 76% (liver disease) and 92% (biliary obstruction), respectively /16/.

A cutoff of 100 U/L has 75% diagnostic sensitivity and 80% specificity in the distinction between PSC and the combination of PSC with cholangiocarcinoma /17/. Moreover, a King’s College Score was described for the same issue using CA 19-9 and CEA determination (40 × CEA + CA 19-9 ≥ 400) /18/; the accuracy of the score, however, is disputed. Gastric cancer

For gastric cancer, CA 19-9 shows a diagnostic sensitivity of 26–60%, depending on the tumor stage. Due to the complementary findings of CA 19-9 and CEA, the diagnostic sensitivity increases to twice that level when both tumor markers are used. Furthermore, the combined determination of CA 19-9 and CEA proved to be an independent prognostic factor for survival, besides the depth of invasion, liver metastasis, peritoneal spread and tumor stage /19/. Colorectal cancer (CRC)

For CRC, the diagnostic sensitivity of CA 19-9 is 18–58%, with a strong dependence on the tumor stage (Dukes A 0–7%, B 17%, C 47%, D 75%) /13/. In comparison to CEA (diagnostic sensitivity 38–58%, Dukes D 65–94%), the incidence rates for CA 19-9 (18–31%, Dukes D 29–59%) are only half as high. Furthermore, the preoperative CA 19-9 (cutoff > 60 U/mL), besides the tumor stage according to Dukes, is a predictor for survival /20/. Other tumors

Much lower diagnostic sensitivity of CA 19-9 has been reported for lung cancer (7–42%) and breast cancer (10%) /126/. In ovarian cancer, the diagnostic sensitivity is 15–38%, with the mucinous type usually having 68–88% and the non mucinous one 25–29%; the diagnostic sensitivity for uterine cancer is only 13% /32122/. Monitoring of the clinical course

Findings observed during monitoring determinations made after an interval of at least 14 days include:

  • Benign disease, if at all, shows either transitory elevation or persistently low concentration, usually < 200 U/mL
  • Untreated malignant disease shows a gradual concentration increase which may be as high as 1,000 U/mL.

In pancreatic cancer, hepatobiliary cancer, gastric and colorectal cancer, the CA 19-9 values usually correlate well with the clinical course associated with surgical treatment, chemotherapy and irradiation /3/ with a lead time of up to 7 months /23/. The clinical course can be:

  • Normalization (< 15 U/mL) within 2–4 weeks after complete surgical resection (stage I)
  • Only a slight, transient decrease in concentration without normalization with palliative therapy
  • Renewed or further increase in the case of tumor recurrence, metastasis and/or progression /3/.

By CA 19-9 monitoring during follow-up, changes in the clinical disease course can be observed with the following accuracy /3/:

  • In gastric cancer (38–70% diagnostic sensitivity, 89–91% specificity, 83% accuracy)
  • In colorectal cancer (53–73% diagnostic sensitivity, 91–94% specificity, 80% accuracy).

Elevated CA 19-9 levels prior to treatment point to a poor prognosis; a preoperative level > 1,000 U/mL is most useful for differentiating between good and poor outcome. A decline in the marker post surgery is the optimal predictor for overall survival. Levels of 200 U/mL and/or following adjuvant chemoradiotherapy > 90 or 180 U/mL predict a poor outcome /2425/.

In a review of 7 studies on pancreatic cancer (two post-surgery, three following chemoradiotherapy and two following chemotherapy) regarding different CA 19-9 pretreatment threshold values (between 200 and 680 and between 958 and 1212 U/mL, respectively), patients with levels below the median had a higher median survival (9.5–22 months) than patients with levels above the median (4.4–8.0 months) /26/.

In further three chemotherapy studies, the median overall survival was longer in responders (10.6–23 months) than in non responders (4.1–8 months) depending on the definition of response (CA 19-9 reduction) as ≥ 75% or ≥ 50% or merely decreasing /26/.

In eight chemotherapy studies with definitions to response (CA 19-9 reduction) as ≥ 15–25%, ≥ 50% or ≥ 75%, responders had a longer median overall survival (4.7–13.8 months) than non responders (2.9–8.1 months) /26/.

In a multicenter chemotherapy study involving 115 patients with pancreas cancer and including the determination of CA 19-9 kinetics from at least three measurements in a time varying covariate model, the multivariate analysis yielded the following evidence /27/: the CA 19-9 kinetics (log CA 19-9) after start of treatment is a significant predictor for the time-to-progression (hazard ratio 1.45) and overall survival (hazard ratio 1.38).

28.8.6 Comments and problems

Method of determination

Comparability between CA 19-9 values obtained by test kits from different manufacturers is moderate even if the same antibody and test methods are used. The intra assay variation is 3–13% and the inter-assay variation is 4–16%. Therefore, in serial monitoring, the same test is always to be used.

Reference interval

The upper reference interval value in a very large group of healthy blood donors was 37 U/mL (x = 8.4 ± 7.4 U/mL) /1/. There is no correlation with age or smoking although women show slightly higher values.

In healthy individuals and patients with the rare blood group constellation Lewis (a–b–) (3–7% of the population), no measurable CA 19-9 elevations should be anticipated since such genotypes lack both a sialyltransferase which is important for the expression of the CA 19-9 epitope and a fucosylated precursor chain.

There is a biological variation for serum CA 19-9 concentration in normal individuals based on Lewis and secretor genotypes in Caucasians /28/. The upper reference interval values for CA 19-9 are between 8 U/mL (Le/le; Se/Se) and 51 U/mL (Le/Le; se/se).

During menstruation and pregnancy, mild CA 19-9 elevations of up to 70 U/mL and 120 U/mL have been described in 15% of non pregnant and in 10% of pregnant women respectively, regardless of gestational age /2/.

With regard to pleural and ascitic fluid, significant differences between benign and malignant diseases have only been found in the case of ascitic fluid (50% diagnostic sensitivity, 100% specificity), given a cutoff value (30 U/mL) comparable to serum /29/.

Obstructive jaundice /30/ based on benign or malignant disease with marker levels obtained before and after endoscopic biliary drainage was associated with elevated CA 19-9 (cutoff value 37 U/mL) in 61% of benign cases and 86% of malignancies. After biliary drainage, decrease in CA 19-9 was observed in 50% of the malignant cases and in almost all benign cases. A cutoff value of 90 U/mL was associated with improved diagnostic accuracy after biliary drainage regarding CA 19-9 elevation in malignant and benign conditions (clinical sensitivity 61%, specificity 95%).


CA 19-9 in serum is stable for 24 h at 4 °C; increase in marker by 5% after 72 h. If stored in sampling tube containing separator gel at 4 °C, an increase in marker by 4% is noted within 24 h and by 21% after 72 h /31/.

Interference factors

False positive CA 19-9 levels in immunoassays may result from antibody synthesis in patients undergoing dry cell extract therapy or after the administration of monoclonal antibodies (mouse or rat).

CA 19-9, as an epitope of a blood group antigen, is a normal component of many mucosal cells and their secreted products (e.g., meconium, feces). Therefore, special precautions must be taken to avoid contamination with secretions. In fact, extremely high CA 19-9 concentrations up to and above 100,000 U/mL may be measured even in healthy individuals, for example in milk, sputum, saliva, bronchial secretions, seminal fluid, cervical mucus, gastric secretions, amniotic fluid, urine and fluid contained in ovarian cysts /3/.

28.8.7 Pathophysiology

In 1979, the first report was published on a monoclonal antibody (1116NS-19-9) which reacted with a human colorectal cancer cell line /32/. Following isolation, the antigen was ascribed to a molecule of 36 kDa and it was named CA 19-9 or GICA (gastrointestinal cancer antigen).

Chemically, CA 19-9 is a monosialoganglioside (glycolipid); specifically, it is the sialyl derivative of lacto-N-fucopentaose II, a hapten of the human Lewis-a blood-group antigen /33/. Its immunochemical detection can be impeded by neuraminidase because in CA 19-9 neuraminic acid is integrated into the epitope /34/.

CA 19-9 has mainly been detected in cancer of the colon (59%), the stomach (89%) and the pancreas (86%) and, more rarely, in cancer of the liver, biliary tract, lung, breast and in mucinous ovarian cancer /35/. In addition, CA 19-9 can also be determined in blood as a high molecular weight mucin (above 106 kDa) /36/.

CA 19-9 (sialyl Lea) like Lex, sialyl-Lex and Ley, derives from the Lewis blood group system (Lea, Leb) and may occur in the form of a glycolipid or mucin which contains neuraminic acid as well as fucose.

As a normal component of the blood group antigen Lea, extremely high, circulating CA 19-9 concentrations represent a physiological finding in secretions. Such concentrations in serum and related body fluids are only considered to be abnormal when exceeding the reference interval of 30–40 U/mL. Increasing concentrations are pathognomonic of tumor disease.


The National Academy of Clinical Biochemistry (NACB, USA) does not recommend measuring CA 19-9 for the diagnosis of pancreatic cancer. However, if used for diagnostic purposes, the marker should be assessed in conjunction with results from other modalities such as CT, MR or endoscopic ultrasound (EUS). Appropriately interpreted CA 19-9 results can guide further invasive testing such as endoscopic retrograde cholangiopancreatography (ERCP), laparoscopy or EUS fine needle aspiration and can also be used in conjunction with other procedures for risk stratification /37/.

According to the ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer, for pancreatic cancer, CA 19-9 can be measured every 1–3 months in patients with locally advanced or metastatic disease receiving active therapy. Elevations in serial CA 19-9 determinations suggest progressive disease but confirmation with other studies should be sought /38/.

According to the National Cancer Comprehensive Network (NCCN, USA), CA 19-9 concentration can be useful for differentiating between pancreatic adenocarcinoma and inflammatory pancreatic lesions. In non metastatic disease, CA 19-9 is recommended for use as preoperative baseline as well as in serial monitoring in conjunction with CT using the same test every 3–6 months for 2 years following surgical resection /39/.


1. del Villano BC, Brennan S, Broch P, Buche C, Liu V, McClure M, Rake B, Space S, Westrick B, Shoemaker H, Zurawski VR. Radioimmunometric assay for a monoclonal antibody-defined tumor marker, CA 19-9. Clin Chem 1983; 29: 549–52.

2. Katsanos KH, Kitsanou M, Christodoulou DK, Tsianos EV. High CA 19-9 levels in benign biliary tract diseases. Report of four cases and review of the literature. Eur J Int Med 2002; 13: 132–35.

3. Lamerz R. CA 19-9, GICA (gastrointestinal cancer antigen). In: Sell St, ed. Serological cancer markers. To-towa: Humana Press, 1992: 309–39.

4. Ritts RE, Nagorney DM, Jacobsen DJ, Talbot RW, Zurawski VR. Comparison of preoperative serum CA 19-9 levels with results of diagnostic imaging modalities in patients undergoing laparotomy for suspected pancreatic or gallbladder disease. Pancreas 1994; 9: 707–16.

5. Richter JM, Christensen MR, Rustgi AK, Silverstein MD. The clinical utility of the CA 19-9 radioimmunoassay for the diagnosis of pancreatic cancer presenting as pain or weight loss. Arch Int Med 1989; 149: 2292–7.

6. Lamerz R. Role of tumour markers, cytogenetics. Ann Oncol 1999; 10, suppl 4: 145–9.

7. Halm U, Schumann T, Schiefke I, Witzigmann H, Mössner J, Keim V. Decrease of CA 19-9 during chemotherapy with gemcitabine predicts survival time in patients with advanced pancreatic cancer. Br J Cancer 2000; 82: 1013–6.

8. Ikeda M, Okada S, Tokuuye K, Ueno H, Okusaka T. Prognostic factors in patients with locally advanced pancreatic carcinoma receiving chemoradiotherapy. Cancer 2001; 91: 490–5.

9. Lamerz R. General laboratory tests and tumor markers in exocrine pancreatic cancer. Diagn Oncol 1993; 3: 147–7.

10. Rocha Lima CMS, Savarese D, Bruckner H, Dudek A, Eckardt J, Hainsworh J, et al. Irinotecan plus gemcitabine induces both radiographic and CA 19-9 tumor marker responses in patients with previously untreated advanced pancreatic cancer. J Clin Oncol 2002; 20: 1182–91.

11. Goonnetilleke KS, Siriwardena AK. Systematic review of carbohydrate antigen (CA 19-9) as a biochemical marker in the diagnosis of pancreatic cancer. Eur J Surg Oncol 2007; 33: 266–270.

12. Bünger S, Laubert T, Roblick UJ, Habermann JK. Serum biomarkers for improved diagnostic of pancreatic cancer. J Cancer Res Clin Oncol 2011; 137: 375–389.

13. Fritz S, Hackert T, Hinz U, Hartwig W, Büchler MW, Werner J. Role of serum carbohydrate antigen 19-9 and carcinoembryonic antigen in distinguishing between benign and invasive intraductal papillary mucinous neoplasm of the pancreas. Br J Surg 2011; 98: 104–110.

14. Park J, Lee KT, Jang TH, Seo YW, Lee KH, Lee JK, et al. Risk factors associated with the postoperative recurrence of intraductal papillary mucinous neoplasms of the pancreas. Pancreas 2011; 40: 46–51.

15. Xu B, Zheng WY, Jin DY, Ding WX, Lou WH, Ramsohok L. Predictive value of serum carbohydrate antigen 19-9 in malignant intraductal papillary mucinous neoplasms. World J Surg 2011; 35: 1103–9.

16. Patel AH, Harnois DM, Klee GG, LaRusso NF, Gores GJ. The utility of CA 19-9 in the diagnosis of cholangio-carcinoma in patients without primary sclerosing cholangitis. Am J Gastroenterol 2000; 95: 204–7.

17. Chalasani N, Baluyut A, Ismail A, Zaman A, Sood G, Ghalib R, et al. Cholangiocarcinoma in patients with primary sclerosing cholangitis: a multicenter case-control study. Hepatology 2000; 31: 7–11.

18. Ramage JK, Donaghy A, Farrant JM, Iorns R, Williams R. Serum tumor markers for the diagnosis of cholangio-carcinoma in primary sclerosing cholangitis. Gastroen-terol 1995; 108: 865–9.

19. Ikeda Y, Oomori H, Koyanagi N, Mori M, Kamakura T, Minagawa S, et al. Prognostic value of combination assays for CEA and CA 19-9 in gastric cancer. Oncology 1995; 52: 483–6.

20. Reiter W, Stieber P, Reuter C, Nagel D, Lau-Werner U, Lamerz R. Multivariate analysis of the prognostic value of CEA and CA 19-9 serum levels in colorectal cancer. Anticancer Res 2000; 20: 5195–8.

21. Canney PA, Wilkinson PM, James RD, Moore M. CA 19-9 as a marker for ovarian cancer: alone and in comparison with CA 125. Brit J Cancer 1985; 52: 131–3.

22. Bast RC, Klug TL, Schaetzl E, Lavin P, Niloff JM, Greber TG, et al. Monitoring human ovarian carcinoma with a combination of a CA 125, CA 19-9, and carcinoembryonic antigen. Amer J Obstet Gynecol 1984; 149: 553–9.

23. Glenn J, Steinberg WM, Kurtzman SH, Steinberg SM, Sindelar WF. Evaluation of the utility of a radioimmunoassay for serum CA 19-9 levels in patients before and after treatment of carcinoma of the pancreas. J Clin Oncol 1988; 6: 462–8.

24. Ferrone CR, Finkelstein DM, Thayer SP et al. Perioperative CA 19-9 levels can predict stage and survival in patients with resectable pancreatic adenocarcinomas. J Clin Oncol 2006; 24: 2897–902.

25. Berger AC, Gardia M, Hoffmann JP et al. Postresection CA 19-9 predicts overall survival in patients with pancreatic cancer treated with adjuvant chemoradiation: a prospective validation by ROTG 9704. J Clin Oncol 2008; 26: 5918–22.

26. Boeck S, Stieber P, Holdenrieder S, Wilkowski R. Prognostic and therapeutic significance of carbohydrate antigen 19-9 as tumor marker in patients with pancreatic cancer. Oncology 2006; 70: 255–64.

27. Boek S, Haas M, Laubender RP, Kullmann F, Klose C, Bruns CJ, et al. Application of a time-varying covariate model to the analysis of CA 19-9 as serum biomarker in patients with advanced pancreatic cancer. Clin Cancer Res 2010; 16: 986–94.

28. Vestergaard EM, Hein HO, Meyer H, Grunnet N, Jörgsen J, Wolf H, Orntoft TF. Reference values and biological variation for tumor marker CA 19-9 in serum for different Lewis and secretor genotypes and evaluation of secretor and Lewis genotyping in a Caucasian population. Clin Chem 1999; 45: 54–61.

29. Lamerz R, Mezger J, Gerbes AL. Der diagnostische Wert von Tumormarkern in Aszites und Pleurapunktaten. In: Wüst G, ed. Tumormarker. Darmstadt: Steinkopff, 1986: 109–17.

30. Marrelli D, Caruso S, Pedrazzani C, Neri A, Fernandes E, Marini M, et al. CA 19-9 serum levels in obstructive jaundice: clinical value in benign and malignant conditions. Am J Surg 2009; 198: 333–9.

31. Banfi G, Parma P, Pontillo M. Stability of tumor markers CA 19.9, CA 125 and CA 15.3 in serum obtained from plain tubes and tubes containing thixotropic gel separator. Clin Chem 1997; 43: 2430–1.

32. Koprowski H, Steplewski Z, Mitchell K, Herlyn M, Fuhrer P. Colorectal carcinoma antigens detected by hybridoma antibodies. Somatic Cell Genetics 1979; 5: 957–72.

33. Magnani JL, Nilsson B, Brockhaus M, Zopf D, Steplew-ski Z, Koprowski H, et al. A monoclonal antibody-defined antigen associated with gastrointestinal cancer in a ganglioside containing sialylated lacto-N-fucopentaose II. J Biol Chem 1982; 257; 14365–9.

34. Bechtel B, Wand AJ, Wroblewski K, Koprowski H, Thurin J. Conformational analysis of the tumor-associated carbohydrate antigen 19-9 and its Lea blood group antigen component as related to the specificity of monoclonal antibody CA 19-9. J Biol Chem 1990; 265: 2028–37.

35. Atkinson BF, Ernst CS, Herlyn M, Steplewski Z, Sears HF, Koprowski H. Gastrointestinal cancer-associated antigen in immunoperoxidase assay. Cancer Res 1982; 42: 4280–3.

36. Magnani JL, Steplewski Z, Koprowski H, Ginsburg V. Identification of the gastrointestinal and pancreatic cancer-associated antigen detected by monoclonal antibody 19-9 in the sera of patients as a mucin. Cancer Res 1983; 43: 5489–92.

37. Duffy MJ, Sturgeon C, Lamerz R, Haglund C, Holubec VL, Klapdor R, et al. Tumor markers in pancreatic cancer: a European Group on Tumor Makers (EGTM) status report. Ann Oncol 2010; 21: 441–7.

38. Locker GY, Hamilton S, Harris J ,Jessup JM, Kemeny N, McDonald JS et al. ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 2006; 24: 5313–27.

39. NCCN Clinical Practice Guidelines in Oncology, Pancreatic Adenocarcinoma. Version 2.2011; https://nccn.org/professionals/physician_gls/PDF/pancreatic.pdf

28.9 CA 125

Rolf Lamerz

The major diagnostic relevance of CA 125 is in assisting the diagnosis of ovarian cancer, evaluating the success of treatment and the disease course. Furthermore, it may be used as a second line marker, after CA 19-9, for pancreatic cancer. It cannot be recommended for other malignant diseases due to its low diagnostic sensitivity and specificity.

28.9.1 Indication


  • Suspected ovarian cancer
  • Monitoring the treatment and course of ovarian cancer.


  • Suspected pancreatic cancer; second line marker after CA 19-9.

28.9.2 Method of determination

Immunoassay; the monoclonal antibody M11 is the capture antibody and OC125 the tracer antibody. The detection limit ranges from 0.5–5 U/mL depending on the test used /123/.

28.9.3 Specimen

Serum, plasma, ascitic fluid, cerebrospinal fluid: 1 mL

28.9.4 Reference interval

Serum, plasma:

0–35 U/mL* /3/

0–65 U/mL** /3/

* 99% confidence interval for healthy individuals

** 99.8% confidence interval for healthy individuals and patients with benign disease

28.9.5 Clinical significance

CA 125 can be elevated in benign and malignant disease (Tab. 28.9-1 – Diagnostic sensitivities of CA 125 in benign and malignant diseases). Benign disease

Elevated serum CA 125 concentrations are measured (threshold > 65 U/mL)in: acute adnexitis (17–25%), endometriosis-related cysts, pelvic inflammatory disease, peritonitis (59%), intestinal obstruction, benign gastrointestinal disease (2–8%), acute pancreatitis (25–36%), cholelithiasis (7% ), cholecystitis (23% ), acute and chronic active hepatitis (2–5%), chronic liver disease (57%), liver cirrhosis (35–64%), jaundice without liver cirrhosis (21–35%), hepatic granulomatosis (47%), autoimmune disease (7%), cardiac and renal insufficiency (11%), external endometriosis (30% > 35 U/mL). Increases in CA 125 were also found in benign adnexal tumors, Meigs’ syndrome, leiomyoma, pericarditis, pleuritis, ascites, venoocclusive disease, and post bone marrow transplantation /345/. Ovarian cancer

Incidence of elevated concentrations

In primary ovarian cancer with an incidence of 15/100,000 women/year, the highest diagnostic sensitivity for CA 125 was 82–96% (cutoff 35 U/mL) and 74–78% (cutoff 65 U/mL). The maximum concentrations were > 5,000 U/mL /23, 4, 6, 7, 89/.

The search for screening methods for the early detection of ovarian cancer was performed based on symptoms (abdominal tension, urination rate, abdominal pain) in conjunction with trans abdominal or trans vaginal ultrasound and use of tumor markers (mainly CA 125). The search was assisted by:

  • Including patients with familial predisposition (hereditary ovarian cancer syndrome, Lynch syndrome)
  • Genetic consulting and testing (BRCA1/2, MLH1, MSH2) in high risk patients
  • Examining post menopausal women /1011/. For women of this group, the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) has provided the most valid results to date /12/.

Results of the UKCTOCS study

In the UKCTOCS study, 202,638 post menopausal women aged 50–74 years were randomly assigned to a no treatment control group (n = 101,359), an annual CA 125 screening group (interpreted using a risk of ovarian cancer algorithm) with transvaginal ultrasound scan as a 2nd line test (multi modal screening, MMS; n = 50,640) and a second screening group (n = 50,639) with annual screening with trans vaginal ultrasound (USS) alone. Women with abnormal screens had repeat tests. Women with persistent abnormality on repeat screens underwent clinical evaluation and, where appropriate, surgery.

Overall, 8.7% of the women in the MMS group and 12% in the USS group required a repeat test, 0.3% of the MMS group and 3·9% of the USS group required clinical evaluation, and 0.2% of the MMS group and 1.8% in the USS group underwent surgery. 42 MMS and 45 USS primary ovarian and tubal cancers were detected, including 28 borderline tumors. 28 of 58 (48.3%) of the invasive cancers were stage I/II, with no difference in stage distribution between the groups, and a further 13 women developed primary ovarian cancer during the year after the screening.

The diagnostic sensitivity, specificity and positive predictive values for all primary ovarian and tubal cancers were 89.4%, 99.8% and 43.3% for MMS and 84.9%, 98.2% and 5.3% for trans vaginal ultrasound, respectively.

For primary invasive epithelial ovarian and tubal cancers, the diagnostic sensitivity, specificity and positive predictive values were 89.5%, 99.8% and 35.1% for MMS and 75%, 98.3% and only 2.8% for ultrasound, respectively.

There was a significant difference in diagnostic specificity but not sensitivity between the two screening groups for all ovarian and tubal cancers.

Special significance was attributed to the percentage of detected stage I/II cancers (47.1% MMS, 50% ultrasound), the pronouncedly higher positive predictive value (35.1% vs. 2.8%) and, consequently, the lower number of interventions per screen detected cancers for MMS (2.9 : 1 vs. 35.2 : 1).

In the PLCO Cancer Screening Randomized Controlled Trial, an intervention group (n = 39,105, annual screening with CA 125 for 6 years and trans vaginal ultrasound for 4 years) was compared with a usual care control group (n = 39,111). The women participating in the trial were aged 55–74 years. Participants were followed up for a maximum of 13 years. Based on the screening, ovarian cancer was detected in 212 and 176 women in the relevant groups. There were 118 and 100 deaths caused by ovarian cancer, respectively. Moreover, diagnostic evaluation of the numerous false positive results obtained was associated with complications. However, the screening did not reduce ovarian cancer mortality /13/. Correlation between CA 125 concentration and tumor mass

According to a combination of data from 15 studies /13/, elevated CA 125 concentrations above 35 U/mL were seen in 49/96 (50%) during FIGO stage I, in 55/61 (90%) during stage II, in 199/216 (92%) during stage III and in 77/82 (94%) during stage IV.

According to the data from 12 different studies /13/, elevated CA 125 levels were seen in 254/317 (80%) in the serous type tumors, in 35/51 (69%) in the mucinous type, in 39/52 (75%) in the endometrioid type, in 28/36 (78%) in the clear cell type and in 56/64 (88%) in the undifferentiated type. Preoperative CA 125 determination

The preoperative level is of prognostic significance in epithelial ovarian cancer. Mild elevations are found more frequently during the early clinical stage, in conjunction with minimal tumor loads, successful therapy responses and low recurrence rates. In contrast, high preoperative values are more likely associated with advanced disease and a low response rate to chemotherapy /14/. Correlation between CA 125 and disease course

Following surgery or chemotherapy, serum CA 125 levels correlated well with the disease course in 87–94% of cases of ovarian cancer /26, 7, 89/; in several cases, a lead time of 1–7 months has been observed /78/.

An exponential decrease of 75–90% of the initial level frequently occurs within the first 7 days after complete tumor resection, followed by normalization within 1–3 months /8/ (half life of 4.8–6.4 days). Therefore, in patients apparently free of tumor, abnormal CA 125 levels are detected in only 1–4% of cases /46/. However, since residual tumor tissue, usually < 1 cm in diameter, is found during second-look surgery in up to 50% of patients after curative surgery despite normalization of CA 125 levels, patients cannot be presumed to be free of tumor based on normalized CA 125 levels /47, 815/. The same applies to normalization of CA 125 under chemotherapy /8/.

On the other hand, in patients with increasingly or persistently elevated CA 125 levels 1–3 months after surgery, residual tumor > 1–2 cm in diameter is still present. Thus, repeated laparotomy to decide on further cytostatic therapy is not necessary in these cases /15/.

In patients with stage III/IV ovarian cancer, the CA 125 half life determined after initial chemotherapy is a very important prognostic indicator for survival. A half life of less than 20 days suggests a good prognosis, while 20–40 days indicate an average and > 40 days a poor prognosis. The 2-year survival rates are 76, 48, and 0%, respectively /16/. This also applies to CA 125 determination one month after chemotherapy /17/.

In tumor recurrence and during clinically progressive disease, CA 125 concentrations above 65 U/mL were seen in 74–89% of cases /4/. The CA 125 level three months after surgery and initial chemotherapy is therefore considered to be a critical predictor of tumor response to therapy /8/.

In order to achieve higher diagnostic sensitivity and specificity, CA 125 was used with a combination of markers to obtain more reliable information for early diagnosis, prognosis, course of treatment and recurrence detection /18/. Of these markers, HE4 (whey-acidic protein human epididymis protein 4) appears to be the most promising. A combination of HE4 and CA 125 used with a different Risk of Ovarian Malignancy Algorithm (ROMA) in pre and post menopausal patients is said to contribute to significantly improved sensitivity for the detection of epithelial ovarian cancers /19/. This finding was disputed in a prospective study comparing the use of CA 125 in single marker assay vs. multiple marker assay /20/. Response criteria following initial chemotherapy

Response criteria include /21/:

  • CA 125 decrease after two initially elevated follow-up values with ensuing 50% decline in marker confirmed by a 4th follow-up sample within 28 days after taking the previous one
  • 75% decline in marker across three follow-up measurements with the last sample taken as above
  • A decline in CA 125 by at least 50% for 20 days or more based on a previous value above two-fold the upper reference interval value before start of therapy (Gynecologic Cancer Intergroup, GCIG) /22/. CA 125 monitoring following initial therapy

The objective of CA 125 monitoring after completed initial therapy is to detect recurrence or metastatic progression of the disease with an empirical lead time of 1–15 months between increase in CA 125 and clinical progression (median 3–4 months). Confirmed increase of CA 125 to more than twice the upper reference interval during first line chemotherapy is considered to be a predictor of recurrence with a diagnostic sensitivity of 84% and a false positive rate below 2% /23/. This also includes confirmed doubling of the CA 125 level compared to baseline (marker minimum) with 94% sensitivity and 100% specificity /24/. Both definitions were adopted by the GCIG for defining progression following first line chemotherapy.

In a study (MRC OVO5/EORTC 55955) /25/, one of the above rules was applied to verify the outcome of early recurrence treatment based on elevated CA 125 (early) compared to second line chemotherapy after about 4.8 months (delayed). Symptoms and clinical findings suggested recurrent ovarian cancer in the patients after complete remission following platinum based first line chemotherapy with normalized CA 125 and clinical examinations and CA 125 measurements performed every three months. In the study, patients were randomly assigned to 264 early cases and 264 delayed cases. With a mean follow-up of 56.9 months from randomization and 370 deaths (186 early, 184 delayed), there was no evidence of a difference in overall survival between early and delayed treatment. Median survival from randomization was 25.7 months (early treatment) and 27.1 months (delayed treatment). The findings show no evidence of a survival benefit with early treatment of relapse based on elevated CA 125 alone and, therefore, the value of routine determination of CA 125 in the follow-up of patients with ovarian cancer who attain complete remission after first line treatment is not proven. However, numerous details of this study do not stand up to critical assessment of the study findings /26/.

A US-American study analyzed 74 patients after primary treatment of epithelial ovarian cancer and secondary cyto reductive surgery in recurrence of the disease. The patients received secondary line chemotherapy due to suspected recurrence because of an increase in CA 125 to two-fold the baseline and CT examination findings. The time interval between the first CA 125 elevation and surgery was 5.3 weeks in an optimal intervention group (microscopic residual disease (MRD) ≤ 0.5 cm, n = 41) and thus significantly shorter than the 16.4 weeks in a suboptimal intervention group (MRD > 0.5 cm, n = 33) (hazard ratio 1.03); the median overall survival of 47 vs. 23 months was even more significant /28/. Guideline recommendations

According to the NACB recommendations for ovarian cancer /27/ and the European Group on Tumor Markers (EGTM) recommendations /28/, CA 125 is not recommended for screening asymptomatic women, but – in conjunction with trans vaginal ultrasound – is recommended for the early detection of ovarian cancer in women with hereditary syndromes who benefit from early intervention.

Moreover, CA 125 determination is recommended as a complementary diagnostic test to assist differentiation between suspected benign and malignant pelvic tumors especially in post menopausal women.

CA 125 determination can also be used for monitoring chemotherapy. The first sampling should be performed within 2 weeks before treatment, followed by sampling every 2–4 weeks during treatment and at intervals of 2–3 weeks thereafter, always using the CA 125 assay from the same manufacturer and excluding patients receiving anti-CA 125 antibody treatment.

Determination of CA 125 during follow-up is only recommended in initially elevated concentrations. A valid definition of follow-up intervals has not been established in current practice to date. Therefore, patients should be monitored every 2–4 months for 2 years and less often thereafter.

It is also recommended to determine CA 125 during primary treatment because both preoperative and postoperative CA 125 levels can be of prognostic significance. Persistently elevated marker concentrations suggest a poor prognosis.

Despite the availability of other promising markers for ovarian cancers, CA 125 is the only marker recommended for use in serous ovarian cancer. Other gynecological cancers

The following diagnostic sensitivities have been reported for CA 125:

  • Breast cancer 8–13% (cutoff 35/65 U/mL) /26/
  • Cervical cancer 13–54% (cutoff 35 U/mL) /929/
  • Endometrial cancer 9–41% (cutoff 35 U/mL) with elevated levels in stage I/II only in conjunction with extrauterine spread, in stage III–IV 55–86% /30/, also in tubal cancer. Gastrointestinal cancers

The following diagnostic sensitivities have been reported for CA 125: pancreatic cancer 45–79% (cutoff 35 U/mL) /25/, liver metastasis 70% (cutoff 65 U/mL), hepatoma 40–77% (cutoff 35/65 U/mL), biliary cancer 46% (cutoff 35 U/mL), colorectal cancer 20–39% (cutoff 35 U/mL) /25/, gastric cancer 39% (cutoff 35 U/mL). The CA 125 concentration correlates well with the tumor stage, especially in pancreatic cancer /5/. Other cancers

Detection rates of 30–57% have been observed for concentrations > 35 U/mL in lung cancer /2/, as well as in lung and pleural metastasis, peritoneal and pleural mesothelioma and non-hodgkin lymphoma.

28.9.6 Comments and problems

Method of determination

Test kits from different manufacturers show a moderate to poor correlation despite the use of the same monoclonal antibody and similar methodology.

Very high CA 125 levels may occur in the serum of tumor patients. In order to avoid a high dose hook effect, levels > 350–400 U/mL should be verified by repeating the analysis using a 1 : 10 serum dilution.

In patients following radioimmunoscintigraphy with OC 125, the frequent occurrence of human anti-murine CA 125 antibodies (HAMA) should be anticipated in the serum, possibly resulting in falsely high or falsely low CA 125 levels. This can be ruled out by using CA 125-II tests (epitope different capture antibody M11) or similar heterologous tests and by using the 2-step procedure (intermediary washing step).

Reference interval

Healthy male blood donors have a mean concentration of 8.0 ± 9.4 U/mL, female blood donors have a mean concentration of 9.9 ± 8.0 U/mL, and both groups together have a total mean concentration of 8.7 ± 8.9 U/mL /3/.

The values from the reference population show a log-normal distribution, with a significant gender related difference (women slightly higher) while the age related difference is less pronounced (slightly decreasing levels with advancing age) in both sexes regardless of smoking habits /3/.

Nonpregnant women occasionally have slightly elevated CA 125 during menstruation /31/. In some pregnant women, elevated CA 125 may occur; specifically, levels during the first trimester (16–268 U/mL) are higher than those during the second (12–25 U/mL) or third (17–44 U/mL) trimesters /32/. This is due to a significant CA 125 gradient between amniotic fluid and serum, with amniotic fluid CA 125 levels of 8800–82,000 (7th–12th gestational week), 3000–13,000 U/mL (13th–25th gestational week) and 640–3400 U/mL (33rd–42nd gestational week) in contrast to cord blood values of 10–50 U/mL and urinary levels of 15–33 U/mL in the newborn.


Increase in marker by 6% within 24 h if stored at 4 °C; increase by 8% after 72 h. If stored in a sampling tube containing separator gel at 4 °C, an increase in marker by 6% is noted within 24 h and by 20% after 72 h /33/. According to the NACB recommendation /27/, temporary storage of the serum samples is possible at 4 °C (1–5 days); prolonged storage requires freezing at – 20 °C (2 weeks to 3 months) and long term storage requires –70 °C.

Other body fluids

Since CA 125 is a high molecular-weight glycolipid or glycoprotein, clinically useful determinations can only be performed in serum/plasma or in body fluids such as pleural exudate/transudate, cerebrospinal fluid, ascitic fluid and certain secretions.

The following CA 125 concentrations were measured:

  • 14,200–15,300 U/mL in cervical secretions of healthy women
  • Mean of 24,600 U/mL in non malignant ovarian fluid
  • In the fluid of benign ovarian cystadenomas, serous type: 50–371,000 U/mL, mucinous type: 845–116,000 U/mL
  • In malignant epithelial ovarian cancers, serous type: < 50–73,200 U/mL, mucinous type: 1130–113,000 U/mL /34/.

28.9.7 Pathophysiology

In 1981, the first report was published on the monoclonal antibody OC 125 which was directed against an epithelial ovarian tumor type, a serous cystadenocarcinoma /35/. This antibody reacts with epithelial human ovarian cancer cell lines and tumor cells from the ascitic fluid of patients with ovarian cancer although not with tissue from fetal or adult ovaries, other adult normal tissue such as uterus, breast, salpinx, skin, lung, liver, spleen, kidneys or with tissue from cancer of non ovarian origin.

Tissue from benign and borderline serous ovarian tumors reacts positively, as do 83% of tissues obtained from serous adenocarcinomas and ovarian cancer of the endometrioid, clear cell and the undifferentiated non mucinous types /36/. OC 125 also reacts with /37/:

  • Normal and carcinomatous epithelial cells of the salpinx, endometrium and endocervix (decidua)
  • Samples of coelomic epithelium (Müllerian ductal epithelium) within fetal tissues
  • Parietal cells and mesothelial cells from the peritoneum, pleura and pericardium.

Therefore, CA 125 is considered to be a normal component of the surface epithelium lining the female genital tract.

The highly purified and dis aggregated protein from tumors, milk, and patient’s serum has a molecular weight of around 200 kDa. The biochemical properties and its unequivocal differentiation from CA 19-9 characterize the OC125-binding determinant as an independent protein-carbohydrate-associated, configuration dependent epitope /38/.

More recent investigations led to the molecular cloning of CA 125 as mucin MUC16 in chromosome 19p13.3-p13.2 /39/. Moreover, in an international workshop (TD 1), numerous antibodies and their correlations were studied by comparative analysis with subsequent publication /40/.


1. Kenemans P, Verstraeten AA, von Kamp GJ, von Mens-dorff-Pouilly S. The second generation CA 125 assays. Ann Med 1995; 27: 107–13.

2. Bast RC, Klug TL, John ES, Jenison E, Niloff JM, Laza-rus H, Berkowitz RS, Leavitt T, Griffiths CT, Parker L, Zurawski VR, Knapp RC. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983; 309: 883–7.

3. Klug TL, Bast RC, Niloff JM, Knapp RC, Zurawski VR. Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA 125) associated with human epithelial ovarian carcinomas. Cancer Res 1984; 44: 1048–53.

4. Kaesemann H, Caffier H, Hoffmann FJ, Crombach G, Würz H, Kreienberg R, Möbus V, Schmidt-Rhode P, Sturm G. Monoklonale Antikörper in Diagnostik und Verlaufskontrolle des Ovarialkarzinoms. CA 125 als Tumormarker. Klin Wschr 1986; 64: 781–5.

5. Klapdor R, Klapdor U, Bahlo M, Dallek M, Kremer B, van Ackeren H, Schreiber HW, Greten H. CA-125 bei Karzinomen des Verdauungstraktes. Dtsch Med Wschr 1984; 109: 1949–54.

6. Crombach G, Zippel HH, Würz H. Erfahrungen mit CA 125, einem Tumormarker für maligne epitheliale Ovarialtumoren. Geburtsh Frauenheilk 1985; 45: 205–12.

7. Rustin GJS, Marples M, Nelstrop AE, Mahmoudi M, Meyer T. Use of CA-125 to define progression of ovarian cancer in patients with persistently elevated levels. J Clin Oncol 2001; 19: 4054–7.

8. Vergote TB, Bormer OP, Abeler VM. Evaluation of serum CA 125 in the monitoring of ovarian cancer. Am J Obstet Gynecol 1987; 157: 88–92.

9. Chi DS, Venkatraman S, Masson V, Hoskins WJ. The ability of preoperative serum CA-125 to predict optimal primary tumor cytoreduction in stage III epithelial ovarian carcinoma. Gyn Oncol 2000; 77: 227–31.

10. Schorge JO, Modesitt SC, Coleman RL, Cohn DE, Kauff ND, Duska LR, et al. SGO White paper on ovarian cancer: etiology, screening and surveillance. Gyn Oncol 2010; 119: 7–17.

11. Pearson DLC. Screening for ovarian cancer. N Engl J Med 2009; 361: 170–177.

12. Menon U, Gentry-Maharaj A, Hallett R, Ryan A, Bumell M, Sharma A, et al. Sensitivity and specificity of multimodal and ultrasound screening for ovarian cancer, and stage distribution of detected cancers: results of the prevalence screen of the UK collaborative trial of ovarian cancer screening (UKCTOCS). Lancet Oncology 2009; 10: 327–340.

13. Buys SS, Partridge E, Black A, Johnson CC, Lamerato L, Isaacs C, et al. Effect of screening on ovarian cancer mortality. PLCO cancer screening randomized controlled trial. J Am Med Ass 2011; 305: 2295–303.

14. Jacobs I, Bast RC. The CA 125 tumour-associate antigen: a review of the literature. Hum Reprod 1989; 4: 1–12.

15. Cooper BC, Sood AK, Davis CS, Ritchie JM, Sorosky JI, Anderson B, Buller RE. Preoperative CA 125 levels: an independent prognostic factor for epithelial ovarian cancer. Obstet Gynecol 2002; 100: 59–64.

16. Rubin SC, Hoskins WJ, Hakes TB, Markman M, Reichman BS, Chapman D, Lewis JL. Serum CA 125 levels and surgical findings in patients undergoing secondary operations for epithelial ovarian cancer. Amer Obstet Gynecol 1989; 160: 667–71.

17. Hawkins RE, Roberts K, Wiltshaw W, Mundy J, Fryatt IJ, McCready VR. The prognostic significance of the half-life of serum CA 125 in patients responding to chemotherapy for epithelial ovarian carcinoma. Br J Obstet Gynaecol 1989; 96: 1395–9.

18. Rustin GJS, Gennings JN, Nelstrop AE, Covarrabias H, Lambert HE, Bagshawe KD. Use of CA-125 to predict survival of patients with ovarian carcinoma. J Clin Oncol 1989; 7: 1667–71.

19. Husseinzadeh N. Status of tumor markers in epithelial ovarian cancer: has there been any progress? A review. Gyn Oncol 2011; 120: 152–7.

20. Moore RG, Jabre-Raughley M, Brown AK, Robison KM, Miller MC, Allard WJ, et al. Comparison of a novel multiple marker assay vs the risk of malignancy index for the prediction of epithelial ovarian cancer in patients with a pelvic mass. Am J Obstet Gynecol 2010; 203: 228e1–6.

21. Van Gorp T, Cadron I, Despierre E, Daemen A, Leunen K, Amant F, et al. HE4 and CA 125 as a diagnostic test in ovarian cancer: prospective validation of the risk of ovarian malignancy algorithm. Br J Cancer 2011; 104: 863–70.

22. Rustin GJS. Use of CA 125 to assess response to new agents in ovarian cancer trials. J Clin Oncol 2003; 21: 187s–93s.

23. Rustin GJS, Quinn M, Thigpen T, et al. Re: New guidelines to evaluate the response to treatment in solid tumors (ovarian cancer). J Natl Cancer Inst 2004; 96: 487–8.

24. Rustin GJS, Nelstrop AE, Tuxen MK, Lambert HE. Defining progression of ovarian cancer during follow-up according to CA 125: a North Thames Ovary Group Study. Ann Oncol 1996; 5: 361–4.

25. Rustin GJS, Marples M, Nelstrop AE, et al. Use of CA 125 to define progression of ovarian cancer in patients with persistently elevated levels. J Clin Oncol 2001; 10: 4054–7.

26. Rustin GJS, van der Burg MEL, Griffin CL, Guthrie D, Lamont A, Jayson GC, et al. Early versus delayed treatment of relapsed ovarian cancer (MRC OV05/EORTC 55955): a randomised trial. Lancet 2010; 376: 1155–63.

27. Bast RC. Commentary. CA 125 and the detection of recurrent ovarian cancer. A reasonably accurate biomarker for a difficult disease. Cancer 2010; 116: 2850–3.

28. Fleming ND, Cass I, Walsh S, Karlan BY, Li AJ. CA 125 surveillance increases optimal respectability at secondary cytoreductive surgery for recurrent epithelial ovarian cancer. Gyn Oncol 2011; 121: 189–92

29. Chan DW, Bast RC, Shih IM, Sokoll LJ, Söletormos G (ovarian cancer sub-committee). National academy of clinical biochemistry laboratory medicine practice guidelines for use of tumor markers in testicular, prostate, colorectal, breast, and ovarian cancers. Clin Chem 2008; 54: e11-e79.

30. Duffy MJ, Bonfrer JM, Kulpa J, Rustin GJS, Soletormos G, Torre GC, et al. CA 125 in ovarian cancer: European group on tumor markers guidelines for clinical use. Int J Gynecol Cancer 2005; 15: 679–91.

31. Duk JM, de Bruijn HWA, Groenier KH, Fleuren GJ, Aalders JG. Adenocarcinoma of the uterine cervix. Prognostic significance of pretreatment serum CA 125, squamous cell carcinoma antigen, and carcinoembryonic antigen levels in relation to clinical and histopathologic tumor characteristics. Cancer 1990; 65: 1830–7.

32. Duk JM, Aalder JG, Fleuren GJ, de Bruijn HWA, CA 125: a useful marker in endometrial carcinoma. Amer J Obstet Gynecol 1986; 155: 1097–1102.

33. Bon GG, Kenemans P, Dekker JJ, Hompes PG, Ver-straeten RA, van Kamp GJ, Schoemaker J. Fluctuations in CA 125 and CA 15-3 serum concentrations during spontaneous ovulatory cycles. Human Reproduction 1999; 14: 566–70.

34. Jacobs IJ, Fay TN, Stabile I, Bridges JE, Oram DH, Grudzinkas JG. The distribution of CA 125 in the reproductive tract of pregnant and non-pregnant women. Br J Obstet Gynaecol 1988; 95: 1190–4.

35. Banfi G, Parma P, Pontillo M. Stability of tumor markers CA 19.9, CA 125 and CA 15.3 in serum obtained from plain tubes and tubes containing thixotropic gel separator. Clin Chem 1997; 43: 2430–1.

36. Fleuren GJ, Nap M, Aalders JG, Trimbos JB, de Bruijn HWA. Explanation of the limited correlation between tumor CA 125 content and serum CA 125 antigen levels in patients with ovarian tumors. Cancer 1987; 60: 2437–42.

37. Bast RC, Feeney M, Lazarus H, Nadler LM. Reactivity of a monoclonal antibody with human ovarian carcinoma. J Clin Invest 1981; 68: 1331–7.

38. Kabawat SE, Bast RC, Welch WR, Knapp RC, Colvin RB. Immunopathologic characterization of a monoclonal antibody that recognizes common surface antigens of human ovarian tumors of serous, endometrioid, and clear cell types. Amer J Clin Pathol 1983; 79: 98–104.

39. Kobayashi F, Sagawa N, Nanbu Y, Nakamura K, No-nogaki M, Ban C, Fujii S, Mori T. Immunohistochemical localization and tissue levels of tumor-associated glycoproteins CA 125 and CA 19-9 in the decidua and fetal membranes at various gestational ages. Amer J Obstet Gynecol 1989; 160: 1232–8.

40. Davis HM, Zurawski VR, Bast RC, Klug TL. Characterization of the CA 125 antigen associated with human epithelial ovarian carcinomas. Cancer Res 1986; 46: 6143–8.

28.10 CA 72-4 (TAG-72)

Rolf Lamerz

CA 72-4 belongs to the group of tumor associated glycoproteins (TAG). Monitoring treatment and the disease course in patients with gastric cancer is the main indication for CA 72-4 as a first line tumor marker and, for increased diagnostic sensitivity, in conjunction with a second line marker (CEA or CA 19-9). Furthermore, CA 72-4 has a relative indication as a second line marker, after CA 125, in ovarian cancer due to complementary findings as well as a higher diagnostic sensitivity for mucinous ovarian cancer.

28.10.1 Indication


First line tumor marker for monitoring of treatment and disease course in patients with gastric cancer; CA 19-9 or CEA as second line markers.


Second line tumor marker for mucinous ovarian cancer.

28.10.2 Method of determination

Immunometric assay using two monoclonal antibodies /1/. The assay includes the capture antibody CC49 bound to the solid phase as well as the detector/tracer antibody B72.3 which detects the bound portion of TAG-72.

28.10.3 Specimen

Serum, plasma, cerebrospinal fluid, pleural exudate, ascitic fluid: 1 mL

28.10.4 Reference interval

Serum/plasma: ≤ 6 U/mL /12, 3, 4, 5, 67/

28.10.5 Clinical significance

CA 72-4 can be elevated in benign and malignant disease (Tab. 28.10-1 – Diagnostic sensitivity of CA 72-4 in benign and malignant diseases). Benign disease

Elevated serum CA 72-4 levels are found in patients with various benign diseases (2–11%) /12/: pancreatitis (3%) /3/, liver cirrhosis (4%) /3/, pulmonary disease (17–19%) /3/, rheumatic disease (21%) /3/, gynecological disease (0–10%) /35/, benign ovarian disease (adenoma, cyst 3–4%) /5/, ovarian cysts (25%) /3/, breast disease (10%), benign gastrointestinal disease (5%) /3/. Overall, however, in comparison to other tumor markers (CEA, CA 19-9), the high clinical specificity in benign diseases is noteworthy /12, 3, 4, 5, 67/. Gastric cancer

Incidence of elevated concentration

Given a diagnostic specificity above 95% in the case of benign gastrointestinal diseases, the diagnostic sensitivity is reported to range from 28–80% and is usually around 40–46% /12, 3, 57/. In general, these diagnostic sensitivities are significantly higher than those of CA 19-9 (mean around 32%) and CEA (20–24%). Correlation between CA 72-4 and spread of disease

Dependence on the tumor stage exists, showing the following positivity rates (cutoff 6 U/mL) in comparison to CA 19-9 (cutoff 37 U/mL) and CEA (cutoff value 5 μg/L): 11/33/0% (stage IA), 20/20/13% (IB), 13/6/19% (II), 46/42/25% (IIIA), 41/28/21% (IIIB), 58/42/37% (IV) and 56/32/11% in tumor recurrence /5/.

Other investigators /7/ found the following positivity rates for CA 72-4 (cutoff 4 U/mL), CA 19-9 (cutoff 37 U/mL) and CEA (cutoff 5 μg/L): 0/25/13% (stage I), 25/13/25% (II), 50/41/23% (III), 57/50/50% (IV), no evidence of residual disease (NED) 4/4/13%, tumor recurrence 61/77/31%.

For CA 72-4, there is correlation with lymph node involvement but none with infiltration of the serosa /5/.

During the postoperative period, CA 72-4 levels normalize within 1–2 weeks and in NED cases will remain within the reference interval, whereas in 70% of recurrent tumor cases (CA 19-9 by 50%, CEA by 20%) they will rise prior to, or concurrently with, clinical detection. In combination, the diagnostic sensitivity of CA 72-4 (42%) in conjunction with CA 19-9 increases to 57%, while together with CEA it only rises to 51% /5/.

In a study on operable gastric cancer and compared to CA 19-9 and CEA, low 3-year cumulative survival was significantly associated with elevated serum concentrations of all markers. However, age, tumor stage and CA 72-4 provided prognostic information in the multivariate analysis. Patients with elevated preoperative serum CA 72-4 concentrations showed a 4.2 times higher risk of death than patients with low levels of the marker /6/. According to other investigations /7/, a multivariate analysis of 167 patients resected for gastric cancer showed that preoperative positivity for CEA, CA 19-9 or CA 72-4 is an independent risk factor for hematogenous recurrences with a relative risk of 4.82, in addition to lymph node involvement (relative risk 3.82).

In another investigation of preoperative CEA, CA 19-9, CA 72-4 and AFP in 95 patients with gastric cancer, clinical sensitivities were 41%, 32.6%, 24.2% and 8.4%, respectively. CEA was more frequently positive in patients with liver metastasis, CA 19-9 was more frequently positive in patients with lymph node, peritoneal and serosal involvement and CA 72-4 was more frequently positive in patients with lymph node, peritoneal and liver involvement. Low 3-year cumulative survival was significantly associated with elevated serum concentrations of CEA, CA 19-9, CA 72-4 and AFP. In the multivariate analysis, age, tumor stage and CA 72-4 were the only independent prognostic factors. Being positive for CA 72-4 was associated with a 3.8-fold higher risk of death in gastric cancer /8/.

In other investigations, preoperative serum levels of CEA, CA 19-9 and CA 72-4 (cutoff 10 μg/L, 60 U/mL, 6 U/mL) were evaluated in 52 patients with gastric cancer. Diagnostic sensitivities were 35%, 52% and 58%, respectively. When all three markers were used, the diagnostic sensitivity increased to 75%. Concerning prognostic value for non metastatic patients, none of the markers were significant. In the metastatic patients, only high values of CA 19-9 and gender were indicators of poor prognosis in the univariate analysis; multivariate analysis, however, revealed that both CA 19-9 and CA 72-4, adjusted for gender, were independent prognostic factors /9/.

In a prospective study on preoperative CEA (above 5 μg/L), CA 19-9 (above 37 μg/L), and CA 72-4 (above 4 μg/L) in 66 patients with gastric cancer (27 patients at stages I–II, 39 patients at stages III–IV), positivity rates were 0/5 for CEA, 7/12 for CA 19-9 and 0/28 for CA 72-4 in patients at stages I–II/III–IV. The serum levels of these markers did not correlate with the histological type or tumor grade of gastric cancer, but CA 72-4 was found to have the best predictive value in indicating advanced disease /10/.

Another study on the preoperative serum levels of CEA, CA 19-9, CA 72-4, CA 242 and hCGβ (cutoffs 5 μg/L, 37 U/mL, 6 U/mL, 20 U/mL, 2 pmol/L, respectively) in 146 patients with gastric cancer of stages I–IV and low cumulative 2-year survival of 40% had the following results: the overall positivity rates of the tumor markers were 18%, 31%, 34%, 34% and 36%, respectively, and all markers except for CA 19-9 showed significant association between elevated concentrations and stage. In the univariate analysis, all markers except for CEA were found to be prognostic factors. In the multivariate analysis, however, stage had the most significant association with prognosis, followed by tumor histology and only by hCGβ and CA 72-4 as independent prognostic factors /11/. Colorectal cancer

Incidence of elevated concentrations

Diagnostic sensitivities of 20–41% have been reported /25/. There is association with the clinical tumor stage according to the Dukes classification, featuring CA 72-4 positivity rates (cutoff 4/6 U/mL), in comparison to CEA (cutoff 5 μg/L) of 3–29% (A), 30–31% (B), 22–53% (C ), 55–70% (D) with an overall rate of 43% and a clinical specificity of 98% for benign colonic diseases /35/.

Correlation between CA 72-4 and disease course

After complete resection, within 18 days, there is a decrease in the tumor marker concentration, but not after palliative surgery. During long term monitoring, CA 72-4 remains elevated or increases further in the presence of residual tumor and often increases in advance of clinical evidence in 78% of tumor recurrences /5/ and/or in 39% of patients with local and in 52% with distant relapse of disease /12/.

When CA 72-4 is combined with CEA, the diagnostic sensitivity increases from 43% to 60% for establishing the primary diagnosis and from 78% to 87% for detecting postoperative tumor recurrence with early or simultaneous increases /5/.

According to a study /13/, in the multivariate analysis of colorectal cancer of all stages, the strongest prognostic factor was stage, followed by tumor location and, as independent prognostic factors, the preoperative serum markers hCGβ, CA 72-4 and CEA. Other cancers

Elevated CA 72-4 levels were found in biliary cancer (35–52%), pancreatic cancer (17–35%) /3512/ and esophageal cancer (4–25%) /2/; however, CA 19-9 appeared to be significantly superior in these cases.

In a study on the prognostic value of CEA, CA 19-9, CA 242, CA 72-4 and hCGβ in 160 patients with pancreatic cancer of all stages, CA 19-9 (cutoff 37 U/mL) was found to have the highest positivity rate of all markers (87%). In the univariate analysis, stage, tumor location and size, curability as well as tumor markers CEA, CA 72-4 and hCGβ were observed to be prognostic factors. In the multivariate analysis, when evaluated individually and adjusted for stage, all markers were found to convey significant independent prognostic information. The strongest prognostic factor was hCGβ followed by CA 72-4 and tumor stage /14/.

In an investigation of CEA, CA 19-9, CA 72-4, SCCA and NSE in the pericardial fluid of patients with various malignant and non malignant etiologies, CA 72-4 was found to be the strongest discriminant factor, followed by CEA. Of 29 effusion fluids, 21 (72%) were non malignant and 8 were malignant; CA 72-4 positivity was only found in one non malignant effusion fluid (4%); it is therefore recommended for use in the evaluation of pericardial effusion of inconclusive cytology /15/. Ovarian cancer

Diagnostic sensitivities for CA 72-4 of 47–80% have been reported /345/ with higher rates in stages III–IV (56%) than in stages I–II (10%) /3/ and a higher diagnostic sensitivity for the mucinous type of cancer in comparison to CA 125 /4/.

The diagnostic specificity for benign ovarian diseases is 97% for CA 72-4 and 85% for CA 125. In the case of tumor recurrence, patients undergoing second look surgery had elevated marker concentrations in 40% (CA 72-4) and 60% (CA 125) whereas in NED cases, no elevations were found for CA 72-4 but in 30% for CA 125. A combination of both tumor markers resulted in an additive increase in the diagnostic sensitivity from 60% (CA 125 alone) to 73% for primary diagnosis and from 60% to 67% for detection of tumor recurrence.

In patients with serous and mucinous ovarian cancers and given a diagnostic specificity of 95% for benign ovarian diseases, a comparison of the tumor markers CA 72-4 (cutoff 6.8 U/mL), CA 125 II (cutoff 160 U/mL), Cancer Associated Serum Antigen (CASA) (cutoff 6.5 U/mL) and CYFRA 21-1 (cutoff 2.4 μg/L) gave the following positivity rates /4/:

  • Overall (%) 47/47/31/44
  • In the primary diagnosis of serous and mucinous adenocarcinomas 36/50/50/33% and 43/21/21/36%, respectively.

The combination of tumor markers gave:

  • The highest increase in diagnostic sensitivity for CA 125 and CA 72-4, from 47% single to a total of 58% for establishing the primary diagnosis
  • For CA 72-4 and CASA from 36/50% to 61% and from 43/21% to 47% in serous and mucinous ovarian cancer, respectively.

Nonetheless, CA 125 is considered as the first line tumor marker for ovarian cancer, whereas CA 72-4 was thought to be inferior to CA 125 /1617/. Other gynecological cancers

Elevated CA 72-4 was found in 24% of patients with breast cancer /3/, in 14% of those with cervical cancer and in 54% of those with endometrial cancer /5/.

28.10.6 Comments and problems

Reference interval

In the serum of healthy individuals, concentrations of around 1–3 U/mL were measured using commercially available CA 72-4 tests. The upper reference interval is reported as 3–6 U/mL.


No special problems regarding stability have been reported. Temporary storage is possible for up to one week at 4 °C; otherwise freezing is necessary at a temperature below –25 °C.

28.10.7 Pathophysiology

A report was published about the production of monoclonal antibody B72.3 which was directed against a membrane enriched extract of breast cancer metastasis as the antigen /18/. This antibody reacted with a determinant (CA 72-4) on a mucin like, high molecular weight, tumor associated glycoprotein complex (with a molecular weight above 1 million) identified as TAG-72 /19/.

B72.3 reacts with the following tissues: breast cancer in up to 84% /20/, colon cancer in above 90% /21/, non small cell lung cancer in up to 96% /22/, epithelial ovarian cancer in up to 100% /23/, less with endometrial cancer /24/, pancreatic cancer /25/, gastric cancer /26/, prostate cancer /27/ as well as other cancers and fetal tissues such as colon, stomach and esophagus. In contrast, no stain uptake by normal adult tissue including the liver, spleen, heart, breast, uterus, lung, bone marrow, colon, stomach, lymph nodes and kidney was found /18/. Thus, TAG-72 is considered to be a pan carcinoma and oncofetal antigen.

The monoclonal antibody B72.3 has also been successfully employed to improve differential diagnostic evaluation of cytological findings from fine needle aspiration of breast masses /28/ as well as those from other body fluids (ascitic fluid, pleural exudate), e.g. in order to differentiate between pulmonary adenocarcinoma and malignant mesothelioma /29/.

Furthermore, the monoclonal antibodies B72.3 and CC49 were used earlier in radio immunoscintigraphic procedures following labeling with 125I, 131I and 111In /30/. They have even been used for intraoperative, antibody guided radio localization of tumors in patients with cancer of the colon, ovary and breast /31/.

Biochemically, the purification of TAG-72 from a human cancer cell line (LS-174T) produced a high molecular weight protein with resistance against chondroitinase and with similarity to blood group related oligosaccharides; thus TAG-72 is considered to be a mucin like molecule /32/. During the production of second generation monoclonal antibodies directed against TAG-72 purified by immunoaffinity techniques, 28 different MAb were obtained (CC series for colon cancer). Among these, several have higher binding constants than B72.3, especially CC49 /33/. The latter shows the same or better binding affinity to tumor tissues from an immunohistochemical point of view. Immunoaffinity purification of TAG-72 with the monoclonal antibody CC49 leads to a homogeneous, high molecular weight mucin of 200–400 kDa with a protein component presenting a molecular weight of 40 kDa and two highly purified, high molecular forms of TAG-72. The B72.3 epitope has been identified as a sialyl-Tn antigen (NeuAcα[2–6]GalNAcα-0-Ser) and the epitope of the second generation MAb CC49 as core-sialyl-oligosaccharide alditol (fraction 2c) /3435/.


1. Stieber P, Fateh-Moghadam A, Wädlich H, Nagel D, Lamerz R, Denecke D. CA 72-4: A new tumour marker for stomach cancer. In: Klapdor R (ed). Recent results in tumor diagnosis and therapy. München: Zuckschwerdt, 1990: 23–26.

2. Heptner G, Domschke S, Domschke W. Comparison of CA 72-4 with CA 19-9 and carcinoembryonic antigen in the serodiagnosis of gastrointestinal malignancies. Scand J Gastroenterol 1989; 24: 745–50.

3. Filella X, Molina R, Jo J, Bedini JL, Joseph J, Ballesta AM. Tumor associated glycoprotein-72 (TAG-72) levels in patients with non-malignant and malignant disease. Bull Cancer 1992; 79: 271–7.

4. Hasholzner U, Baumgartner L, Stieber P, Meier W, Hofmann K, Fateh-Moghadam A. Significance of the tumour markers CA 125 II, CA 72-4, CASA and CYFRA 21-1 in ovarian carcinoma. Anti-Cancer Res 1994; 14: 2743–6.

5. Guadagni F, Roselli M, Cosimelli M, Ferroni P, Spila A, Cavaliere F, et al. CA 72-4 serum marker – a new tool in the management of carcinoma patients. Cancer Invest 1995; 13: 227–38.

6. Gaspar MJ, Arribas I, Coca MC, Diez-Alonso M. Prognostic value of carcinoembryonic antigen, CA 19-9 and CA 72-4 in gastric carcinoma. Tumor Biol 2001; 22: 318–22.

7. Marrelli D, Pinto E, de Stefano A, de Manzoni G. Farnetani M, Garosi L, Roviello F. Preoperative positivity of serum tumor markers is a strong predictor of hematogenous recurrence of gastric cancer. J Surg Oncol 2001; 78: 253–8.

8. Ucar E, Semerci E, Ustun H, Ystim T, Huzmeli C, Gullu M. Prognostic value of preoperative CEA, CA 19-9, CA 72-4, and AFP-levels in gastric cancer. Adv Ther 2008; 25: 1075–84.

9. Ychou M, Duffour J, Kramar A, Gourgou S, Grenier J. Clinical significance and prognostic value of CA 72-4 compared with CEA and CA 19-9 in patients with gastric cancer. Disease Markers 2000; 16: 105–10.

10. Cidon EU, Bustamante R. Gastric cancer: tumor markers as predictive factors for preoperative staging. J Gastrointest Canc 2010. https://doi.org/10.1007/s12029-010-9161-0.

11. Louhimo J, Kikkola A, Alfthan H, Stenman UH, Haglund C. Preoperative hCGβ and CA 72-4 are prognostic factors in gastric cancer. Int J Cancer 2004; 111: 929–33.

12. Filella X, Molina R, Mengual PJ, Ballesta AM. Significance of CA 72-4 in patients with colorectal cancer. Comparison with CEA and CA 19-9. J Nucl Biol Med 1991; 35: 158–61.

13. Louhimo J, Carpelan-Holmström M, Alfthan H, Stenman UH, Järvinen HJ. Serum hCGβ, CA 72-4 and CEA are independent prognostic factors in colorectal cancer. Int J Cancer 2002; 101: 545–8.

14. Louhimo J, Alfthan H, Stenman UH, Haglund C. Serum HCGβ and CA 72-4 are stronger prognostic factors than CEA, CA 19-9 and CA 242 in pancreatic cancer. Oncology 2004; 66: 126–31.

15. Karatolios K, Maisch B, Pankuweit S. Tumormarker im Perikarderguss bei malignen und nichtmalignen Perikardergüssen. Herz 2011; 36: 290–5.

16. Einhorn N, Knapp RC, Bast RC, Zurawski VR. CA 125 assay used in conjunction with CA 15-3 and TAG-72 assays for discrimination between malignant and non-malignant diseases of the ovary. Acta Oncol 1989; 28: 655–7.

17. Jäger W, Adam R. Vergleich der Serumkonzentration des CA 72-4 mit dem CA 125 während des klinischen Verlaufs von Ovarialkarzinom-Patientinnen. Onkologie 1989; 12: 164–6.

18. Colcher D, Horan Hand P, Nuti M, Schlom J. A spectrum of monoclonal antibodies reactive with human mammary tumor cells. Proc Natl Acad Sci 1981; 78: 3199–3208.

19. Johnson VG, Schlom J, Paterson AJ, Bennett J, Magnani JL, Colcher D. Analysis of a human tumor-associated glycoprotein (TAG-72) identified by monoclonal antibody B72.3. Cancer Res 1986; 46: 850–7.

20. Lottich SC, Johnston WW, Szpak CA, Delong ER, Thor A, Schlom J. Tumor-associated antigen TAG-72: correlation of expression in primary and metastatic breast carcinoma lesions. Breast Cancer Res Treatm 1985; 6: 49–56.

21. Stramignioni D, Bowen R, Atkinson B, Schlom J. Differential reactivity of monoclonal antibodies with human colon andenocarcinomas and adenomas. Int J Cancer 1983; 31: 543–52.

22. Szpak CA, Johnston WW, Roggli V, Kolbeck J, Lottich C, Vollmer R, Thor A, Schlom J. The diagnostic distinction between malignant mesothelioma of the pleura and adenocarcinoma of the lung as defined by a monoclonal antibody (B72.3). Amer J Pathol 1986; 122: 252–60.

23. Thor A, Gorstein F, Ohuchi N, Szpak CA, Johnston WW, Schlom J. Tumor-associated glycoprotein (TAG-72) in ovarian carcinomas defined by monoclonal antibody B72.3 J Natl Cancer Inst 1986; 76: 995–1006.

24. Soisson AP, Berchuck A, Lessey BA, Soper JT, Clarke-Pearson DL, McCarty KS, Bast RC. Immunohistochemical expression of TAG-72 in normal and malignant endometrium: correlation of antigen expression with estrogen and progesterone receptor levels. Am J Obstet Gynecol 1989; 161: 1258–63.

25. Lyubsky S, Madariaga J, Lozowski M, Mishriki Y, Schuss A, Chao S, Lundy J. A tumor-associated antigen in carcinoma of the pancreas defined by monoclonal antibody B72.3. Amer J Clin Pathol 1988; 89: 160–7.

26. Ohuchi N, Thor A, Nose M, Fujita J, Kyoguku M, Schlom J. Tumor-associated glycoprotein (TAG-72) detected in adenocarcinomas and benign lesions of the stomach. Int J Cancer 1986; 38: 643–50.

27. Myers RB, Schlom J, Srivastava S, Grizzle WE. Expression of tumor-associated glycoprotein 72 in prostatic intraepithelial neoplasia and prostatic adenocarcinoma. Modern Pathol 1995; 8: 260–5.

28. Lundy J, Lozowski M, Mishriki Y. Monoclonal antibody B72.3 as a diagnostic adjunct in fine needle aspirates of breast masses. Ann Surg 1986; 203: 399–402.

29. Martin SE, Moshiri S, Thor A, Vilasi V, Chu EW, Schlom J. Identification of adenocarcinoma in cytospin preparations of effusions using monoclonal antibody B72.3 Amer J Clin Pathol 1986; 86: 10–6.

30. Esteban JM, Colcher D, Sugarbaker P, Carrasquillo JA, Bryant G, Thor A, Reynolds JC, Larson SM, Schlom J. Quantitative and qualitative aspects of radiolocalization in colon cancer patients of intravenously administered MAb B72.3. Int J Cancer 1987; 39: 50–9.

31. Tuttle SE, Jewell SD, Mojzisik CM, Hinkle GH, Colcher D, Schlom J, Martin EW. Intraoperative radioimmunolocalization of colorectal carcinoma with a handheld gamma probe and MAb B72.3: comparison of in vivo gamma probe counts with in vitro MAb radiolocalization. Int J Cancer 1988; 42: 352–8.

32. Johnson VG, Schlom J, Paterson AJ, Bennett J, Magnani JL, Colcher D. Analysis of a human tumor-associated glycoprotein (TAG-72) identified by monoclonal antibody B72.3. Cancer Res 1986; 46: 850–7.

33. Muraro R, Kuroki M, Wunderlich D, Poole DJ, Colcher D, Thor A, et al. Generation and characterization of B72.3 second generation monoclonal antibodies reactive with the tumor-associated glycoprotein 72 antigen. Cancer Res 1988; 48: 4588–96.

34. Kjeldsen T, Clausen H, Hirohashi S, Ogawa T, Iijima H, Hakomori H. Preparation and characterization of monoclonal antibodies directed to the tumor-associated O-linked sialosyl-2-6α-N-acetylgalactosaminyl (Sialosyl-Tn) epitope. Cancer Res 1988; 48: 2214–20.

35. Hanisch FG, Uhlenbruck G, Egge H, Peter-Katalinic J. A B72.3 second-generation-monoclonal antibody (CC49) defines the mucin-carried carbohydrate epitope Galβ(1–3) (NeuAcα[2–6])GalNAc. Biol Chem Hoppe-Seyler 1989; 370: 21–6.

28.11 CA 15-3

Rolf Lamerz

CA 15-3 is a useful marker for monitoring the disease course of patients with metastatic breast cancer. The test is not suited for screening or for primary diagnosis because the diagnostic sensitivity is too low for localized disease, and a high proportion of elevated levels is associated with benign breast diseases as well as cancer of other organs.

28.11.1 Indication

Breast cancer: monitoring the outcome of treatment and disease course.

28.11.2 Method of determination

Immunometric assay and enzyme immunoassay using the two monoclonal antibodies 115D8 and DF3. The capture antibody 115D8 is bound to the solid phase while the detector/tracer antibody DF3 detects CA 15-3 after binding to the capture antibody. The limit of detection is below 1 U/mL /1/. Similar antibodies (Ma 552, Ma 695) on two different fully automatic analyzer platforms showed good correlation /2/.

28.11.3 Specimen

Serum, plasma, cerebrospinal fluid, pleural exudate, ascites fluid: 1 mL

28.11.4 Reference interval

Serum/plasma 25–40 U/mL /345/

28.11.5 Clinical significance

CA 15-3 can be elevated in benign and malignant disease (Tab. 28.11-1 – Diagnostic sensitivity of CA 15-3 in benign and malignant diseases). Benign disease

Elevated serum CA 15-3 concentrations are found in patients with:

  • Dialysis-dependent renal insufficiency (20% above 30 U/mL) /6/, HIV infection (stage dependent above 50% above 18 U/mL) /9/, chronic inflammatory liver diseases (5%) /7/, bronchial diseases (15%) /7/
  • Various benign diseases (3.3% above 40 U/mL), such as hepatic, pancreatic, rheumatic diseases and tuberculosis /4/
  • Benign breast diseases (4% above 25 U/mL) /8/, myomastopathy (3–11% above 28 U/mL) /5/, fibroadenoma (7.7%) and other benign diseases of the thorax (25% above 30 U/mL)
  • A prevalence of 4.7% at levels above 50 U/mL; 8.9% of these patients have breast diseases and 12.5% have pulmonary diseases. Breast cancer

Dependence of CA 15-3 diagnostic sensitivity on the tumor stage

For breast cancer, the diagnostic sensitivity is:

  • 19–22% (cutoff 28 U/mL) /58/ for preoperative cases
  • 32% in stage M0 cases (cutoff 50 U/mL)
  • 16% (cutoff 25 U/mL) in node negative and 54% in node positive cases /9/
  • 54–91% (cutoff 25/28/50 U/mL) in metastatic breast cancer /23, 4, 5, 9, 1011/.

Following treatment, increased levels are found in patients with:

  • No evidence of residual disease (NED) in only 5.9%
  • Complete/partial response in 29%
  • Static or progressive disease in up to 100% of all such cases (cutoff 40 U/mL) /12/.

Correlation between CA 15-3 and tumor size

The positivity rate of CA 15-3 correlates with the tumor mass: in 4–16% during stage I, in 13–54% during stage II, in 65% during stage III and in 54–91% during stage IV /14/ and/or in 14–23% during stage T1/2 /3/, 27–86% during stage T3/4 /39/ and/or in 22% for node negative and 38% for node positive cases /8/.

A low diagnostic sensitivity of 21% is found in conjunction with local and regional tumor recurrence (cutoff 35 U/mL; median 45 U/mL) /13/.

In metastatic disease, the serum CA 15-3 concentration depends on the location of the metastases. Skin metastasis is associated with a low diagnostic sensitivity (median 25 U/mL /9/ and in 36.5% above 50 U/mL) and connective tissue involvement has a diagnostic sensitivity of 40% /8/ and/or 47–83% /11/. Higher levels are observed in association with bone metastasis with 32–75% of the values above 27 U/mL /11/ and 61% above 35 U/mL /13/, without a significant difference between pulmonary metastasis or visceral metastasis /8/. The highest concentrations are measured in association with liver metastasis (median 54 U/mL) /13/, with 45.4% of the values above 50 U/mL) /9/ and 64% above 35 U/mL) /13/, and in the presence of multiple metastases (median 93 U/mL) /9/.

Correlation between CA 15-3 and disease course

During a monitoring period of 13–40 months, CA 15-3 detects tumor recurrence with a diagnostic sensitivity of 45–77%, a specificity of 94–98% and with positive predictive values of 41–92% (lead time 3–18 months) /14/.

In the case of metastasis and while undergoing treatment, the probability of progressive disease or response to treatment fluctuates, in the presence of CA 15-3 concentration increases or decreases ≥ 25%, between 75% and 94% for progressive disease and between 72% and 93% for response to treatment /1415/.

Comparison of CA 15-3 to other tumor markers

Besides CA 15-3 as the first line MUC 1 gene marker, CEA is historically the first, and still important, marker in breast cancer while other markers such as cytokeratins (TPA, TPS, CYFRA 21.1) and soluble oncoproteins (c-erbB-2) are of limited significance /15/.

In breast cancer, CA 15-3 is superior to CEA /1015/. In combination with CEA, CA 15-3 results in a significant increase in diagnostic sensitivity for detecting tumor recurrence from 41% for CA 15-3 and 40% for CEA to 56% for both markers combined. The combination of CA 15-3 and CEA has a lead time of 2–18 months (mean 5.2 months) before clinical or radiological detection at a diagnostic sensitivity of 40–60%.

For the detection of metastasis, the diagnostic sensitivity is 60–80%, with higher sensitivity for liver metastasis (85–90%) than for bone metastasis (65–75%) at a specificity of up to 95% /15/.

CA 15-3 as a prognostic/predictive marker

CA 15-3 (cutoff 30 U/mL) is an independent prognostic indicator in breast cancer. Five year disease free survival and overall survival are 44% and 67%, respectively, in patients with elevated preoperative levels compared with 65% and 83%, respectively, in patients with low concentrations /16/.

Similar rates (5-year disease free survival/overall survival of 45% in patients with elevated preoperative concentrations and 86% in normal concentrations; cutoff 30 U/mL) were found in a prospective study. Tumor stage and CA 15-3 are the two most powerful predictors of survival /17/.

In a different study /18/ evaluating patients after primary tumor surgery with a median follow-up of 69 months, statistical evidence was provided for preoperative CA 15-3 (cutoff 31 U/mL) as significant prognostic marker in node negative breast cancer. Moreover, elevated CA 15-3 levels were measured in estrogen receptor positive tumors (cutoff 35 U/mL: 70% versus 46%) in advanced breast cancer patients at first disease relapse /19/.

Other investigators found a high prognostic value in the kinetics of CA 15-3 (lead time, first elevated concentration and interval between first abnormal CA 15-3 and diagnosis) before and at first metastasis in breast cancer /20/.

In a study on the combined use of serum CEA and CA 15-3 with disease free survival and death from disease in 1046 women with breast cancer without metastasis at the time of primary diagnosis, it was found that elevated preoperative serum marker values correlated with early relapse (CA 15-3) and death from disease (CEA, CA 15-3) in univariate analyses /21/. By comparing pre- and postoperative values, a decline in values post surgery was found. In those patients where marker levels of CEA decreased more than 33%, a significantly higher risk for relapse and death from disease was observed in the univariate analysis. In the multivariate analysis, this decrease in CEA proved to be an independent prognostic factor.

Among 740 patients with stages I–III breast cancer, elevated preoperative levels of CA 15-3 and CEA were identified in 12.4% and 10.7% of the cases, respectively. Tumor size above 5 cm, lymph node metastasis (≥ 4) and advanced stage were associated with higher preoperative levels. Elevated CA 15-3 and CEA levels were associated with significantly poorer disease free survival and overall survival compared to normal preoperative marker concentrations /22/. In the multivariate analysis, age below 35 years, tumor size above 2 cm, lymph node metastasis, estrogen receptor expression and elevated CA 15-3 and CEA preoperative values were independent prognostic factors for disease free survival.

In a prospective evaluation of the pretreatment concentrations of CEA (cutoff 5 μg/L) and CA 15-3 (cutoff 30 U/mL) in 2062 patients with primary loco regional tumor, increased concentrations were found in 12.7% and 19.6% of the patients, respectively, and one or both tumor markers were increased in 28%. Increases in each tumor marker correlated with larger tumor sizes and nodal involvement /23/. Tumor size, estrogen receptor and CEA were independent prognostic factors in the total group as well as in node positive and node negative patients. All patients with CEA above 7.5 μg/L had recurrence during follow-up. Use of both tumor markers allowed discrimination of the groups of risk in T1 node negative patients: 56.3% of recurrences were seen when one or both tumor markers were increased, whereas only 9.4% of recurrences were seen in T1 node negative patients without increases in either marker.

Prediction of recurrence

A study was performed on breast cancer recurrence including 3953 patients with one or more consecutive marker measurements during their relapse free survival (RFS) period. Abnormal CA 15-3 above 30 U/mL or 50% higher than the first value recorded and elevated ALP were the criteria considered in the analysis. During the RFS period, recurrence was reported in 720 cases (20%) before which 274 (35%) had abnormal CA 15-3 and 35 (4%) had elevated ALP /24/. Risk of recurrence increased in 30% of patients with abnormal CA 15-3 (hazard ratio 1.30) and 4% of those with abnormal ALP. Recurrence risk was greatest for patients with either (hazard ratio 2.40) and with both (hazard ratio 4.69) biomarkers abnormal. ALP better predicted liver recurrence and CA 15-3 better predicted breast cancer recurrence.

Tumor markers (CA 15-3, CEA) were determined in postoperative follow-up of 427 breast cancer patients after mastectomy with equivocal findings from conventional radiological procedures (chest X-ray, bone scintigraphy or liver echography). During the 35 month follow-up, 221 patients with a total of 332 equivocal findings had positive predictive values of 69% and 83% and negative predictive values of 98% and 91%, respectively, for the indication of the metastatic or benign origin of the equivocal findings. Clinical symptoms were not helpful in predicting metastatic disease (diagnostic sensitivity, specificity and accuracy of 60%, 53% and 54%, respectively) /25/.

A combined investigation using CA 15-3 monitoring and modern imaging (18F-FDG-PET/CT) was performed on 89 female patients with breast cancer who developed post therapy rising CA 15-3 but negative conventional imaging. Tumor deposits were detected in 40 of the 89 patients in chest wall, internal mammary nodes, lungs, liver and skeleton. In 23 of 40 patients, solitary small lesion was amenable to radical therapy. In 7 out of these 23 patients, complete disease remission lasting more than one year was observed /26/.

Treatment and monitoring of recurrence

A study on 68 breast cancer patients with distant metastasis receiving chemotherapy (salvage therapy) at the time of significant increase in the markers CEA or CA 15-3 (tumor marker guided treatment) or being treated only after radiological confirmation of metastasis (conventional treatment) had the following results:

  • A significantly longer mean lead time of the first group compared to the second group (17.3 versus 2.9 months)
  • Better survival curves from salvage therapy or from mastectomy (at 36 months from salvage therapy: 28% vs. 9% survivors, and at 84 months from mastectomy: 42% vs. 19% survivors) /27/.

The correlation between tumor marker kinetics like CEA and CA 15-3 and imaging concerning the effectiveness of chemotherapy was determined in 77 metastatic breast cancer patients by consecutive marker measurement. The markers were measured at the beginning of chemotherapy, after 20–30 days, after 40–60 days and at the time the effectiveness of chemotherapy was evaluated with imaging under strict definition of biochemical progression or response based on a marker increase or decrease ≥ 25% /28/. Approximately 70% of cases showed a correlation between tumor marker kinetics and imaging results during chemotherapy. After 1 month, no statement about treatment response was possible by using tumor marker kinetics. The effectiveness or ineffectiveness of chemotherapy could be determined correctly in 40% of patients after 2 months and in 70% of patients after approximately 3 months. Other malignant diseases

Elevated CA 15-3 levels are observed in 39–71% of patients with ovarian cancer /3810/, in 14–26% with endometrial cancer /38/, in 9.1% with uterine cancer, in 10–71% with lung cancer /510/, and in 10–61% with gastric, pancreatic and hepatocellular cancer /10/.

28.11.6 Comments and problems

Method of determination

The results of test kits from different manufacturers show differences although the same antibodies and similar methodology is used. Therefore, the same test and specimen (serum or heparinized plasma) should always be used for monitoring, specifying this in the result report.

A comparison of tests kits from different manufacturers gave large differences in the slope of the correlation line, especially for CA 15-3 values above 50–200 U/mL, despite relatively good correlation coefficients (r > 0.93); higher variation is observed with the use of serum instead of heparinized plasma /1/.

Reference interval

Mean levels of around 10–17 U/mL were measured in the serum of healthy individuals, independent of gender /34/; the upper reference interval differs with values of studies between 25 /89/, 27 /11/, 28 /5/ or 40 U/mL /412/. Serum CA 15-3 concentrations are above 25 U/mL in 4–7% of breast feeding women /3/ and above 30 U/mL in 8% of pregnant women without any elevations in the amniotic fluid.

It should be noted that subcutaneous administration of G-CSF for treatment of leukopenia following chemoradiotherapy in breast cancer patients can induce false positive elevation of CA 15-3 /29/.

The MUC1 568 A/G genotype polymorphism strongly influences CA 15-3 concentrations in genotypes AA, AG and GG of healthy women and women with either benign or malignant diseases /30/.


Stable for 24 h at 4 °C; increase in marker by 3% after 72 h. If stored in a sampling tube containing separator gel at 4 °C, an increase in marker by 2% is noted within 24 h and by 18% after 72 h /31/.

28.11.7 Pathophysiology

CA 15-3 is a high molecular weight carbohydrate antigen of 300 kDa belonging to the milk fat globule mucin family. It can be measured using two monoclonal antibodies, in particular MAb 115D8 directed against the milk fat globule antigen MAM-6a, which is attached to the solid phase, and MAb DF3 as a tracer directed against the membrane fraction of human breast cancer cells.

The 115D8 antibody was selected from a group of monoclonal antibodies directed against mammary gland differentiation antigens. Its target is an epitope on a glycoprotein (MAM-6a on the MAM-6 antigen) originating from human milk fat globule membranes /32/. The heavily glycosylated antigen was initially isolated from milk (molecular weight > 400 kDa); the MAM-6a epitope is thought to be localized on the carbohydrate chain of the antigen /32/.

Immunohistochemically, MAM-6 has been identified as an epithelial membrane marker on the apical pole of epithelial, normal ductal and alveolar structures and an antigen which is often homogeneously distributed in the cytoplasm of breast cancer cells /32/.

MAb DF3 is a monoclonal antibody directed against a membrane antigen called DF3 which is localized on human breast cancer cells and has a molecular weight of 300–400 kDa /33/. This antigen is also considered to be a differentiation antigen of malignant epithelial breast cells, being one of the high molecular weight glycoproteins of the mucin family of milk fat globule membrane antigens /34/.

Immunohistochemically, DF3 could be detected on the surface and in the cytoplasm of human breast lesion cells as well as fibroadenomatous cells /35/. Its expression in breast cancer correlates with the nuclear grade, histological grade and estrogen receptor status /36/. In addition, DF3 could be detected immunohistochemically in 95% of benign, borderline and malignant tumors of the ovary and on the surface of ovarian cancer cell lines /37/.

Numerous investigations have documented the occurrence and significance of the MUC 1 molecule as a high molecular weight type I transmembrane glycoprotein produced by the MUC 1 gene in chromosome 1q21–24 /38/ (more information in /39/). The molecule includes a cytoplasmic portion of 69 amino acids and a heavily glycosylated extracellular domain that contains numerous peptide repeats of highly conserved sequences of 20 amino acids with five potential sites of O-linked glycosylation. The freely circulating, shed MUC 1 glycoprotein contains short carbohydrate side chains and exposed repetitive epitopes on its peptide core, coming in contact with the immune system and capable of triggering humoral or cellular immune response. Anti-MUC-1 antibodies are directed against epitopes of the repetitive domain, primarily against the PDTRP sequence. Many of the monoclonal anti-MUC1 antibodies that are commercially produced and/or developed in scientific laboratories were subjected to a comparative analysis in an international workshop /40/.


The European Group on Tumor Markers (EGTM) recommends that CA 15-3 in patients with metastatic breast cancer treated with chemotherapy be determined before every chemotherapy course. In patients treated with hormone therapy, it should be measured at least every 3 months. An increase in tumor marker concentration of at least 25% of the previous value is considered to be significant, recommending that such an increase be confirmed in a second specimen obtained within a month. Similarly, confirmed decreases in serum levels of more than 50% are considered to be consistent with tumor response /41/.

The National Academy of Clinical Biochemistry (NACB) Panel states that CA 15-3 (similarly to CEA) in combination with imaging and clinical examination may be used to monitor chemotherapy in patients with advanced breast cancer. This marker may be particularly helpful in patients with none valuable disease. In such patients, two successive increases in CA 15-3 of more than 30%, each, are likely to indicate progressive disease and may result in cessation of therapy, change in therapy or entry of the patient into clinical trials /42/.


1. Huber P, Bischof P, Kretschmer R, Truschnig M, Halwachs G, Schmidt M. CA 15-3: a multicentre evaluation of automated and manual tests. Eur J Clin Chem Clin Biochem 1996; 28: 77–84.

2. Molina R, Gion M, Gressner A, Troalen F, Auge JM, Holdenrieder S, et al. Alternative antibody for the detection of CA 15-3 antigen: a European multicenter study for the evaluation of the analytical and clinical performance of the Access BR Monitor assay on the UniCel Dxl800 Immunoassay. Clin Chem Lab Med 2008; 46: 612–22.

3. Kreienberg R. Allgemeine und spezielle Laborparameter im Rahmen der Tumornachsorge bei gynäkologischen Malignomen und beim Mammakarzinom. Gynäkologe 1989; 22: 55–62.

4. Ruibal A, Genolla J, Rosell M, Gris JM, Colomer R. Serum CA 15-3 levels in patients with non-tumoral diseases, and establishment of a threshold for tumoral activity. Results in 1219 patients. Int J Biol Markers 1986; 1: 159–60.

5. Stieber P, Diergarten K, Eiermann W, Albiez A, Fateh-Moghadam A. CA 15-3: Evaluation and clinical value in breast carcinomas compared with CEA and TPA. In: Klapdor R (ed). New tumor markers and their monoclonal antibodies. Stuttgart: Thieme, 1988: 45–50.

6. Ammon A, Eiffert H, Weber MH, Rummel J, Neimann J. Tumormarker bei dialyse pflichtiger Niereninsuffizienz. Onkologie 1988; 11: 260–3.

7. Bauer R, Oehr P, Scholtes H, Kohlhas K, Böhm I, Niedecken HW. Thymidinkinase, CA 15-3 und β2-Mikroglobulin im Serum HIV-infizierter Patienten. NucCompact 1989; 20: 78–84.

8. Paulick R, Caffier H, Kaesemann H. Erste Erfahrungen mit dem monoklonalen Markersystem CA 15-3 bei Mammakarzinompatientinnen. TumorDiagn Therapie 1986; 7: 85–7.

9. Pons-Anicet DMF, Krebs BP, Mira R, Namer M. Value of CA 15-3 in the follow-up of breast cancer patients. Br J Cancer 1987; 55: 567–9.

10. Tondini C, Hayes DF, Kufe DW. Circulating tumor markers in breast cancer. Hematol Oncol Clin North Am 1989; 3: 653–74.

11. Bieglmayer C, Szepesi T, Neunteufel W. Follow-up of metastatic breast cancer patients with a mucin-like carcinoma-associated antigen: comparison to CA 15-3 and carcinoembryonic antigen. Cancer Letters 1988; 42: 199–206.

12. Sole LA, Colomer R, Navarro A, Encabo G. CA 15-3: early results of a new breast cancer marker. Anticancer Res 1986; 6: 683–4.

13. Langhammer HR, Ellgas W, Laubenbacher Ch, Erbas B, Busch R, Riesinger U, Senekowitsch R. CA 15-3 und CEA in der Rezidivdiagnostik des Mammakarzinoms unter Berücksichtigung von Lokalisation und Ausmass der Metastasierung. Tumordiagn Ther 1994; 15: 96–103.

14. Hayes DF. Tumor markers for breast cancer. Hematol Oncol Clinics North Am 1994; 8: 485–506.

15. Molina R. Tumor markers in breast cancer. In Diamandis EP, Fritsche HA, Chan DW, Schwartz MH (eds) Tumor markers, physiology, pathobiology, and clinical applications. Washington DC; AACC PRESS 2002: 165–79.

16. Shering SG. Preoperative CA 15-3 concentrations predict outcome of patients with breast carcinoma. Cancer 1998; 83: 2521–7.

17. Kumpulainen EJ, Keskikuru RJ, Johansson RT. Serum tumor marker CA 15.3 and stage are the two most powerful predictors of survival in primary breast cancer. Breast Cancer Research and Treatment 2002; 76: 95–102.

18. Gion M, Boracchi P, Dittadi R, Biganzoli E, Peloso L, Mione R, et al. Prognostic role of serum CA15.3 in 362 node-negative breast cancers. An old player for a new game. Eur J Cancer 2002; 38: 1181–8.

19. Tampellini M, Berruti A, Gorzegno G, Bitossi R, Bottini A, Durando A, et al. Independent factors predict supranormal CA 15-3 serum levels in advanced breast cancer patients at first disease relapse. Tumor Biol 2001; 22: 367–73.

20. de la Lande B, Hacene K, Floiras JL, Alatrakchi N, Pichon MF. Prognostic value of CA 15.3 kinetics for metastatic breast cancer. Int J Biol Markers 2002; 17: 231–8.

21. Ebeling FG, Stieber P, Untsch M, Nagel D, Konecny GE, Schmitt UM, et al. Serum CEA and CA 15-3 as prognostic factors in primary breast cancer. Br J Cancer 2002; 86: 1217–22.

22. Park BW, Oh JW, Kim JH, Park SH, Kim KS, Kim JH, et al. Preoperative CA 15-3 and CEA serum levels as predictor for breast cancer outcomes. Ann Oncol 2008; 19: 675–81.

23. Molina R, Auge JM, Farrus B, Zanon G, Pahisa J, Munoz M, et al. Prospective evaluation of carcinoembryonic antigen (CEA) and carbohydrate antigen 15.3 (CA 15.3) in patients with primary locoregional breast cancer. Clin Chem 2010; 56: 1148–57.

24. Keshaviah A, Dellapasqua S, Rotmensz N, Lindtner J, Crivellari D, Collins J, et al. CA 15-3 and alkaline phosphatase as predictors for breast cancer recurrence: a combined analysis of seven international breast cancer study group trials. Ann Oncol 2007; 18: 701–8.

25. Nicolini A, Carpi A, Ferrari P, Pieri L. Utility of a serum tumour marker panel in the post-operative follow-up of breast cancer patients with equivocal conventional radiological examinations. Tumor Biol 2003; 24: 275–80.

26. Grassetto G, Fornasiero A, Otello D, Bonciarelli G, Rossi E, Nashimben O, et al. 18F-FDG-PET/CT in patients with breast cancer and rising CA 15-3 with negative conventional imaging: a multicentre study. Eur J Radiol 2011; 80: 545–8.

27. Nicolini A, Carpi A, Michelassi C, Spinelli C, Conte M, Miccoli P, et al. “Tumour marker guided” salvage treatment prolongs survival of breast cancer patients: final report of a 7-year study. Biomedicine & Pharmacotherapy 2003; 57: 452–9.

28. Di Gioa D, Heinemann V, Nagel D, Untch M, Kahlert S, Bauerfeind I, et al. Kinetics of CEA and CA 15-3 correlate with treatment response in patients undergoing chemotherapy for metastatic breast cancer (MBC). Tumor Biol 2011; 32: 777–85.

29. Briasoulis E, Andreopoulou E, Tolis CF, Bairaktari E, Katsaraki A, Dimopoulos MA, et al. G-CSF induces elevation of circulating CA 15-3 in breast carcinoma patients treated in an adjuvant setting. Cancer 2001; 91: 909–17.

30. Kruit A, Tilanus-Linthorst MM, Boonstra JG, van Schaik RHN, Grutters JC, van den Bosch JMM, et al. MUC1 568 A/G genotype-dependent cancer antigen 15-3 levels in breast cancer patients. Clin Biochem 2009; 42: 662–5.

31. Banfi G, Parma P, Pontillo M. Stability of tumor markers CA 19.9, CA 125 and CA 15.3 in serum obtained from plain tubes and tubes containing thixotropic gel separator. Clin Chem 1997; 43: 2430–1.

32. Hilkens J, Kroezen V, Buijs F, Hilgers J, van Vliet M, de Voogd W, et al. MAM-6, a carcinoma associated marker: preliminary characterization and detection in sera of breast cancer patients. In: Ceriani RL, ed. Monoclonal antibodies and breast cancer. The Hague: Martinus Nijhoff, 1985: 28–42.

33. Hayes DF, Sekine H, Ohno T, Abe M, Keefe K, Kufe DW. Use of a murine monoclonal antibody for detection of circulating plasma DF3 antigen levels in breast cancer patients. J Clin Invest 1985; 75: 1671–8.

34. Abe M, Kufe D. Structural analysis of the DF3 human breast carcinoma-associated protein. Cancer Res 1989; 49: 2828–9.

35. Sekine H, Ohno T, Kufe DW. Purification and characterization of a high molecular weight glycoprotein detectable in human milk and breast carcinoma. J Immunol 1985; 135: 3610–5.

36. Lundy J, Thor A, Maenza J. Monoclonal antibody DF3 correlates with tumor differentiation and hormone receptor status in breast cancer patients. Breast Cancer Res Treatment 1985; 5: 269–76.

37. Friedman EL, Hayes DF, Kufe DW. Reactivity of monoclonal antibody DF3 with a high molecular weight antigen expressed in human ovarian carcinomas. Cancer Res 1986; 46: 5189–94.

38. von Mensdorff-Pouilly S, Snijdewint FGM, Verstraeten AA, Verheijen RHM, Kenemans P. Human MUC1 mucin: a multifaceted glycoprotein. Int J Biol Markers 2000; 15: 283–56.

39. Duffy MJ, Evoy D, McDermott EW. CA 15-3: Uses and limitation as a biomarker for breast cancer. Clin Chim Acta 2010; 411: 1869–74.

40. Rye PD, Price MR. ISOBM TD-4. International Workshop on Monoclonal Antibodies against MUC1. Tumor Biol 1998; 19, Suppl 1: 1–152.

41. Molina R, Barak V, van Dalen A, Duffy MJ, Einarsson R, Gion M, et al. Tumor markers in breast cancer – European group on tumor markers recommendations. Tumor Biol 2005; 26: 281–93.

42. Duffy MJ, Esteva SJ, Harbeck N, Hayes DF, Molina R (breast cancer). National academy of clinical biochemistry laboratory medicine practice guidelines for use of tumor markers in testicular, prostate, colorectal, breast, and ovarian cancers. Clin Chem 2008; 54: e11–e79.

28.12 Calcitonin (CT)

Lothar Thomas

CT is a peptide hormone secreted by the para follicular C cells of the thyroid gland. It serves as a specific and sensitive tumor marker for sporadic and familial forms of medullary thyroid carcinoma. Besides parathyroid hormone (PTH) and calcitriol (1,25-dihydroxycholecalciferol), CT is the third Ca2+-regulating hormone. PTH and calcitriol raise the serum Ca2+ concentration, whereas CT has a decreasing effect.

28.12.1 Indication

Determination of basal CT /1/:

  • Although CT is not a specific marker for medullary thyroid carcinoma, its routine measurement represents a useful tool in the preoperative evaluation in patients with nodular thyroid disease
  • Positive familial history of the hereditary form of medullary thyroid carcinoma or of multiple endocrine neoplasia
  • Postoperative monitoring and follow-up of patients with confirmed medullary thyroid carcinoma.

Pentagastrin stimulation test

  • In case of CT levels outside the reference interval but not really pathologic (gray area) and suspected sporadic medullary thyroid carcinoma
  • In case of CT concentrations within the reference interval and suspected hereditary medullary thyroid carcinoma.

28.12.2 Method of determination

Immunoassays based on enzyme or luminescence labeling /2/, one-step sandwich assays based on the streptavidin-biotin technology /3/. All tests detect human CT. Commercial tests are calibrated against the 2nd International WHO calibrator 89/260.

Pentagastrin stimulation test

Principle: after pentagastrin administration, patients with MTC or C-cell hyperplasia show a markedly more pronounced increase in CT than normal individuals whose CT concentration increases up to three times the baseline value /4/.

Procedure: an indwelling catheder is used to collect 5 mL of blood (basal concentration), 0.5 μg of pentagastrin per kg of body weight are administered as an intravenous bolus injection, additional blood samples (5 mL) are taken at 2 min. and 5 min. after pentagastrin injection.

28.12.3 Specimen

Serum or plasma (heparin or EDTA): 1 mL

28.12.4 Reference interval

Refer to Tab. 28.12-1 – Threshold values below which the likelihood of MTC is small.

28.12.5 Clinical significance

Medullary thyroid carcinoma is a malignant neoplasia of the C-cells accounting for 5–10% of thyroid tumors. The medullary thyroid carcinoma is of the sporadic type in 70–80% of cases and has a prevalence of approximately 0.6% in thyroid nodular disease. The remaining 20–30% are distributed over three familial types:

  • Multiple endocrine neoplasia type 2a (MEN 2a)
  • Multiple endocrine neoplasia type 2b (MEN 2b)
  • A hereditary type not associated with MEN
  • C-cell hyperplasia, which is controversially discussed as to its premalignant nature, is also associated with elevated CT level.

Medullary thyroid carcinoma releases CT and, occasionally, other proteins such as CEA, NSE, serotonin, chromogranin, somatostatin, substance P, preopiomelanocortin derived products, and gastrin releasing peptide /4/. However, the specificity of CT as a marker for medullar thyroid cancer only applies to concentrations above 100 ng/L because, besides in C-cell hyperplasia, elevated serum CT concentrations are also observed in pulmonary and pancreatic endocrine tumors, autoimmune thyroid disorder, hypergastrinemia pseudo hypoparathyroidism type 1A, and renal failure.

Influence factors also play a significant role (Tab. 28.12-2 – Elevated calcitonin without the presence of medullary thyroid cancer). Men have higher levels than women. Depending on the test used, chronic kidney disease leads to considerable variation in increased basal CT values /2/; the lower cutoff value in the pentagastrin stimulation test indicating the presence of MTC has been raised from 100 ng/L to 400 ng/L by some authors /8/.

CT levels in diagnostics, post surgery and during follow-up as well as statements regarding molecular genetic analysis are shown in Tab. 28.12-3 – Serum calcitonin in medullary thyroid carcinoma.

Endocrine active tumors of the foregut (esophagus, stomach, duodenum, pancreas) can also secrete CT. The range of CT concentrations among five patients with foregut tumors was 42–7460 ng/L; all five patients had liver metastases and died within a period of 1.2–27.2 months /9/.

28.12.6 Comments and problems

Method of determination

The analytical quality of CT immunoassays varies between different manufacturers. Some tests also determine procalcitonin in patients with sepsis. CT cutoff concentrations for pentagastrin stimulation testing can also vary considerably /2/.

Reference interval

Since it is not possible to specify a concrete cutoff value applicable to all assays, the manufacturers’ cutoff values are to be used. A good test will define separate cutoff values for men and women.


Decreased CT concentrations are found already after 2–3 h of storage at 20 °C and after ≥ 6 h of storage at 4–8 °C; after 12 h of storage, CT concentration is reduced by 23%. No influence is seen after several days of storage at –40 °C /2/.

28.12.7 Pathophysiology

CT is a polypeptide with a molecular weight of 3.5 kDa and is composed of 32 amino acids with a disulfide bridge between positions 1 and 7 plus a carboxy-terminal prolinamide. CT is the amino acid sequence 60–91 of procalcitonin which is cleaved into N-ProCT, CT and katacalcin (Fig. 19.5-4 – Procalcitonin (amino acids 1–116)).

The gene for calcitonin encodes yet another peptide referred to as calcitonin gene-related peptide (CGRP). The expression of CT and CGRP is tissue specific; whereas CT is expressed predominantly in the C-cell, CGRP is almost exclusively found in the peripheral nervous system.

The physiological relevance of CT is the short term regulation of Ca2+. High plasma increases in CT lead to a pronounced decrease in Ca2+ because of the inhibitory effect of CT on osteoclasts which have specific CT receptors. Besides Ca2+, gastrointestinal hormones (e.g., gastrin and catecholamines) also have a stimulating effect on the secretion of CT. In all cases, these effects last for brief periods of time only; in contrast, continuous plasma Ca2+ elevation does not result in an increase in CT. Elevated CT levels in patients with medullary thyroid carcinoma have no impact on plasma Ca2+ or bone metabolism.

MTC develops in parafollicular C-cells derived from the crest of the neural tube.


1. Dietlein M, Wieler H, Schmidt M, Schwab R, Goretzki PE, Schicha H. Routine measurement of serum calcitonin in patients with nodular thyroid disorders. Nuklearmedizin 2008; 47: 65–72.

2. Bieglmayer C, Vierhapper H, Dudczak R, Niederle B. Measurement of calcitonin by immunoassay analyzers. Clin Chem Lab Med 2007; 45: 662–6.

3 Kahaly GJ, Schimnich AA, Davis TE, Feldkamp J, Karger S, König J, et al. United States and European multicenter prospective sudy for the analytical performance and clinical validation of a novel sensitive fully automated immunoassay for calcitonin. Clin Chem 2017; 63: 1469–96.

4. Vitale G, Ciccarelli A, Craglia M, Galderisi M, Rossi R, Del Prete S, Abbruzzese A, Lupoli G. Comparison of two provocative tests for calcitonin in medullary thyroid carcinoma: omeprazole vs pentagastrin. Clin Chem 2002; 48: 1505–10.

5. Rink T, Truong PN, Schroth HJ, Diener J, Zimny M, Grünwald F. Calculation and validation of a plasma calcitonin limit for early detection of medullary thyroid carcinoma in nodular thyroid disease. Thyroid 2009; 19: 327–31.

6. D’Herbomez M, Caron P, Bauters C, Cao CD, Schlienger JL, Sapin R, et al. Reference range of serum calcitonin levels in humans: influence of calcitonin assays, sex, age, and cigarette smoking. Eur J Endocrinol 2007; 157: 749–55.

7. Basuyau JP, Mallet E, Levroy M, Brunelle P. Reference intervals for serum calcitonin in men, women and children. Clin Chem 2004; 50: 1828–30.

8. Birchardt KA, Heinzl H, Gessl A, Hörl WH, Kasserer K, Sunder-Plassmann G. Calcitonin concentrations in patients with chronic kidney disease and medullary thyroid carcinoma or C-cell hyperplasia. Kidney Int 2006; 70: 2014–20.

9. Wuilmet L, Jovenin N, Larbre H, Levy-Bobot N, Diebold MD, Jolly D, et al. Digestive calcitonin-secreting tumors of the foregut: comparison with non-calcitonin-secreting tumors. Eur J Gastroenterol Hepatol 2006; 18: 951–5.

10. Dora KM, da Silva Canali MHB, Capp C, Punales MK, Vieira JGH, Maia AL. Normal perioperative serum calcitonin levels in patients with advanced medullary thyroid carcinoma: case report and review of the literature. Thyroid 2008; 18: 895–9.

11. Constante G, Meringolo D, Durante C, et al. Predictive value of serum calcitonin levels for preoperative diagnosis of medullary thyroid carcinoma in a cohort of 5817 consecutive patients with thyroid nodules. J Endocrinol Metab 2007; 92: 450–5.

12. Karges W, Dralle H, Raue F, et al. Calcitonin measurement to detect medullary thyroid carcinoma in nodular goiter: German evidence based consensus recommendation. Exp Clin Endocrinol Diabetes 2004; 112: 52–8.

13. Gibelin H, Essique D, Jones C, et al. Increased calcitonin level in thyroid nodules without medullary carcinoma. Br J Surg 2005; 92: 574–8.

14. Vierhapper H, Niederle B, Bieglmayer C, et al. Early diagnosis and curative therapy of medullary thyroid carcinoma by routine measurement of serum calcitonin in patients with thyroid disorders. Thyroid 2005; 15: 1267–72.

15. Frank-Raue K, Raue F, Buhr HJ, Baldauf G, Lorenz D, Ziegler R. Localization of occult persisting medullary thyroid carcinoma before microsurgical reoperation: high sensitivity of selective venous catheterization. Thyroid 1992; 2: 113–7.

16. Wohlik N, Cote GJ, Evans DB, Goepfert H, Ordonez NG, Gagel RF. Application of genetic screening information to the management of medullary thyroid carcinoma and multiple endocrine neoplasia type 2. Endocrinol Metab Clin North Am 1996; 25: 1–25.

17. Machens A, Niccoli Sire P, Hoehl J, et al. Early malignant progression of hereditary medullary thyroid cancer. N Engl J Med 2003; 349: 1517–25.

18. Raue F, Grauer A. Humanes Calcitonin. In Thomas L, ed. Labor und Diagnose. Frankfurt 2008, TH-Books, S. 1316–9.

19. Pacini, F, Fontanelli M, Fugazzola L, Elisei R, Romei C, di Coscio G, Miccoli P, Pinchera A. Routine measurement of serum calcitonin in nodular thyroid diseases allows the preoperative diagnosis of unsuspected sporadic medullary thyroid carcinoma. J Clin Endocrin Metab 1994; 78: 826–9.

28.13 Carcinoembryonic antigen (CEA)

Peter Nollau, Christoph Wagener, Rolf Lamerz

28.13.1 Indication

  • Detection of tumor progression or recurrence in the postoperative monitoring of colorectal carcinomas
  • Differential diagnosis of liver tumors.

28.13.2 Method of determination

Immunoassays (ELISA). Calibration: First International reference preparation (73/601), 100 IU/Amp.

28.13.3 Specimen

Serum or plasma: 1 mL

28.13.4 Reference interval

Serum/plasma* 1.5–3.0 μg/L

* See manufacturers’ instructions

28.13.5 Clinical significance

CEA can be elevated in benign and malignant diseases (Tab. 28.13-1 – Serum CEA concentration in non malignant and malignant disease). Benign disease

The median serum CEA concentration is higher in elderly individuals and smokers than in younger individuals and non smokers /1/. Among non malignant conditions, elevations of CEA are primarily found in the following disorders: inflammatory liver disease, alcohol induced liver cirrhosis, pancreatitis, inflammatory gastrointestinal disease such as ulcerative colitis and diverticulitis, inflammatory pulmonary disease /23/. CEA in cancer patients

In contrast to non malignant conditions, serum CEA concentrations increase steadily in malignant disease because of the progressive growth of the tumor. Thus, in addition to referring to a cutoff level (horizontal assessment), test results should be longitudinally assessed based on the differences observed in serial determinations.

The diagnostic sensitivity of CEA depends on tumor stage and spread. The rate of CEA elevation and the levels increase with tumor burden.

Among malignant tumors, the diagnostic sensitivity is highest in colorectal cancer and in medullary thyroid cancer /23/. In colorectal cancer, the rates of CEA elevation according to tumor stage are as follows: Dukes A: 0–20%; Dukes B: 40–60%; Dukes C: 60–80%; Dukes D: 80–85% /4/.

In non metastatic breast cancer, serum CEA levels are elevated in only 10% of cases and, in general, they do not exceed a concentration corresponding to 5-fold the upper reference interval value. In metastatic breast cancer, the diagnostic sensitivity is in the range of 50–60%; in 25% of patients, CEA concentrations are higher than 5-fold the upper reference interval value /5/. In gastric, pancreatic, lung, ovarian and cervical cancers, CEA concentrations are elevated in advanced stages only, the rate of CEA elevations is in the range of 50–70% /6/. Moreover, elevated serum CEA levels are seen at varying frequencies in bladder, liver and kidney cancers and in melanomas and lymphomas. CEA in differentiation of cancers

For differential diagnosis of primary gastrointestinal cancers, CEA is of limited value because of overlapping of results of different cancers.

In pancreatic cancer, which is usually detected at later stages, CA 19-9 is superior to CEA.

In the differential diagnosis of liver tumors, CEA can be used as an adjunct to imaging techniques, especially when serial determinations are considered. Serum CEA levels above 8–10-fold the upper reference interval value are found sporadically in benign liver disease and in approximately 6% of patients with hepatocellular cancer. In contrast, 50–60% of the CEA levels in gastrointestinal and pancreatic cancer patients with liver metastasis are within the above mentioned range /78/.

In pancreatic cancer, which is usually detected at later stages, CA 19-9 is superior to CEA (diagnostic sensitivity 33–77%, specificity 64–100%, cutoff value 2.5–5.0 μg/L) /9/. Prognosis for colorectal cancer

In colorectal cancer, the determination of serum CEA can be used for prognostic purposes and for the diagnosis of residual tumor after resection. Preoperative CEA concentrations are of prognostic value also within defined tumor stages (TNM classification) /9/. It has been shown in various studies that tumors with high preoperative serum CEA levels are associated with a poor prognosis. Based on the information available to date, it remains to be clarified whether preoperative CEA can provide prognostic data in patients with colorectal cancer (especially those with stage Dukes B). CEA may thus be able to help identify the subset of patients with aggressive disease who might benefit from adjuvant chemotherapy /10/. However, it is important to point out that there currently are no reports showing a benefit from the use of adjuvant chemotherapy based solely on an increased preoperative CEA level.

The question of whether elevated CEA levels are caused by the primary tumor only and/or by distant metastasis can be answered by serial CEA determinations in the postoperative follow-up (6–8 weekly determinations after tumor resection). CEA concentrations that do not fall into the reference interval and increase subsequently indicate residual tumor with high certainty. CEA determination is thought to be of prognostic value in metastatic colorectal cancer after resection of liver metastases and it has been shown in various studies that high preoperative CEA concentrations are associated with a poor prognosis /11/. Monitoring post surgery in colorectal cancer

In colorectal cancer, serial determinations of CEA are the most sensitive non invasive method for the diagnosis of local or distant tumor recurrence after resection of the primary tumor. For each patient, an individual baseline CEA concentration is defined. When CEA increases steadily for at least 2 months, tumor recurrence is probable.

Kinetics (slope analysis) can help to differentiate local recurrence from distant metastasis. Slope analysis showed a median CEA increase of 0.24 μg/L per 10 days for local recurrence and of 1.7 μg/L for liver metastasis. In general a CEA increase > 1 μg/L per 10 days is indicative of distant metastasis /12/.

Depending on the definition of the increase and on the relevant reference, the positive predictive value of a CEA increase in the postoperative monitoring of colorectal cancer is in the range of 65–84%. The negative predictive value is in the range of 85–95%. Thus, stable CEA concentrations exclude recurrence with a relatively high probability /131415/.

According to meta analyses comprising retrospective and prospective studies, intensive follow-up post-surgery after curative operation of the primary tumor was associated with a significantly prolonged (up to 9%) 5-year survival rate. Regular CEA determination was a crucial diagnostic component in this context. Most likely, the increase in the 5-year survival rate at intensive follow-up taking into account the CEA concentration is due to early detection and improvements in technique in liver metastasis surgery.

Liver metastasis develops in more than three quarters of patients, who underwent incomplete operation of the primary tumor, and is resectable in approximately one quarter of these patients where successful resection can achieve 5-year survival rates of 21–48% /11/.

Based on this situation, various oncological associations recommend in colorectal cancer to perform CEA monitoring post-surgery every 2–3 months for at least 3 years after diagnosis. In particular, intensive follow-up is recommended for patients with stage Dukes B and C disease with the objective of early detection and surgery of liver metastasis and solitary lung metastasis. An increase in CEA by 30% compared to the previously measured value is considered to be significant /1617/. It should be pointed out, however, that the actual benefit of postoperative CEA monitoring is not unequivocally confirmed until it has been verified by relevant randomized prospective studies. The diagnostic benefit of CEA determination will increase if more efficient therapeutic options are available.

CEA monitoring post-surgery in colorectal cancer patients with preoperative CEA concentrations below 5 μg/L has been discussed controversially and/or rejected.

In a study /18/ involving 186 colorectal cancer stage I–III patients with 146 initial non secretors (initially normal preoperative levels of CEA) and 40 secretors (elevated levels at presentation), 22 patients with recurrent colorectal cancer (16 non secretors and 6 secretors) were detected corresponding to 50% and 66%, respectively. Likewise, recurrent disease was detected in 272 of 954 of colorectal cancer patients. Normal preoperative CEA values were present in 63% of patients with recurrent disease. 60% of the patients with a normal preoperative CEA and recurrent disease had elevated CEA values during follow-up when the last measurement was done within 3 months before recurrent disease was diagnosed /19/.

In a retrospective analysis /20/ on 533 colorectal cancer patients who underwent resection with a curative intent, the 5-year survival rate for preoperative CEA above 5 μg/L with respect to a post-operative CEA level drop rate of 60% was 83%. Besides the depth of invasion (hazard ratio 2.6) and lymph node metastasis (hazard ratio 2.2), the CEA drop rate (hazard ratio 3.0) was an independent prognostic factor for poor survival.

In an analysis of the postoperative surveillance of patients with early (stages I and IIA) and late (stages IIB and III) colorectal cancer stages, salvage rates were the same (36–37%). Median survival after second surgery after recurrence was 51.2 and 35.8 months for early and late stage patients, respectively. Single sites of first recurrence did not significantly differ between early and late stages, but multiple sites of recurrence occurred less often in early stage patients (3.6% vs. 28.6%, for early vs. late, respectively). Methods of first detection of recurrence were not significantly different, showing superiority for CEA (29.1% vs. 37.4%), followed by CT (23.6% vs. 26.4%), chest X-ray (7.3% vs. 12.1%) and colonoscopy (12.7% vs. 8.8%), for early versus late stage disease, respectively /21/.

In patients with advanced colorectal cancer undergoing chemotherapy with serial CEA monitoring, CEA flare occurred in 78 of 670 cases besides frequent CEA elevations. Compared with patients with increasing CEA, patients with CEA flare had a significantly better radiological objective response rate (11% vs. 73%), median progression free survival (3.1 vs. 8.3 months) and overall survival (10.9 vs. 17.7 months). CEA flare is an independent predictive and prognostic factor for tumor response and survival /22/.

In the development of metachronous liver metastasis (20–40%) during the follow-up of 1099 patients after curative surgery, a preoperative CEA concentration above 5 μg/L (odds ratio 1.591), depth of invasion (odds ratio 2.3), lymph nodes metastasis (odds ratio 2.0) and vascular invasion (odds ratio 1.9) were found to be independent prognostic factors of a high risk group and indicators for intensive follow-up /23/. Monitoring in non small cell lung cancer (NSCLC)

The positivity rates for preoperative CEA above 5 μg/L and CYFRA 21-1 above 2.8 μg/L in 193 stage I NSCLC patients with adenocarcinoma were 27.8% and 7.8%, respectively, with an overall 5-year survival rate of 79.3% and a median follow-up of 35.5 months /24/. Patients with preoperatively elevated CEA had shorter recurrence free survival and early recurrence and were considered to be good candidates for adjuvant chemotherapy.

A prospective study on the predictive and prognostic role of decline in serum CEA concentrations (cutoff 5 μg/L) and CYFRA 21-1 (cutoff 3.2 μg/L) in 107 patients with advanced NSCLC undergoing conventional chemotherapy had the following results after the second course of chemotherapy:

  • Radiologic objective response rate 44%
  • CEA and CYFRA 21-1 response rate (more than 20% reduction compared to baseline level) of 38% and 61%, respectively (median survival was 9 months and, in particular, was 13 months for patients who had a CEA response and 11 months for patients who had a CYFRA 21-1 response compared with 8 months and 6 months for patients who did not respond, respectively) /25/
  • In the multivariate analysis, performance status, LD enzyme activity and CEA and CYFRA 21-1 response were confirmed as independent prognostic factors for survival and CEA and CYFRA 21-1 appeared to be reliable surrogate markers of chemotherapy efficacy.

In 105 NSCLC patients treated with gefitinib with a response rate of 27.8% and a median survival time of 9.3 months, those with no history of smoking or a CEA baseline above 5 μg/L were more likely to be sensitive to gefitinib. Multivariate analysis showed that a good performance status and elevated CEA were independent prognostic and predictive factors for gefitinib efficacy /26/. Monitoring in breast cancer

In a prospective evaluation of baseline CEA (cutoff 5 μg/L) and CA 15-3 (cutoff 30 U/mL) in 2062 breast cancer patients with primary loco regional tumor, increased concentrations were found in 12.7% and 19.6% of patients, respectively, and one or both tumor markers were increased in 28%. Increases in each tumor marker correlated with larger tumor sizes and nodal involvement /27/. Tumor size, estrogen receptor presence and CEA were independent prognostic factors in the total group. All patients with CEA concentrations above 7.5 μg/L had recurrence during follow-up. Use of both tumor markers allowed discrimination of the groups of risk in T1 node negative patients: 56.3% of recurrences were seen when one or both tumor markers were increased, whereas only 9.4% of recurrences were seen in node negative patients without increases in either marker.

According to the tumor marker recommendations of the European Group on Tumor Markers (EGTM), CA 15-3 and CEA are not suited for screening or early detection of breast cancer but are helpful in the early detection of distant metastasis. Moreover, preoperative elevations of both markers in combination with other prognostic factors are associated with poor outcome in breast cancer /28/.

Serial determinations of both markers is recommended for the early detection of recurrence in breast cancer patients showing no evidence of disease if the detection of metastatic spread is followed by therapeutic action.

Moreover, CEA and CA 15-3 should be measured before each course of chemotherapy and every 3 months for therapy monitoring. An increase in the markers by more than 25% of the previously measured value and a second control value above the reference interval are considered to be significant and should be verified within a month.

A steady, verified marker increase indicates progression of the disease and a verified marker increase by more than 50% indicates tumor response.

28.13.6 Comments and problems


In some assays, CEA concentrations determined in serum and plasma are different. The manufacturers’ instructions should be followed regarding the selection and treatment of specimens.

Reference interval

The median and distribution of CEA levels in serum or plasma depend on age and smoking habits. Since CEA concentrations do not follow normal distribution, the 95th percentile is defined as the upper limit of the reference interval.

Method of determination

CEA carries specific and cross reactive antigenic determinants. Cross reacting determinants are found on antigens that are present, for example, in normal plasma /29/. CEA assays should not be affected by cross reacting antigens. Despite an overall good correlation between different CEA assays, divergent results may be observed in certain samples. This must be taken into account when serial CEA determinations are performed.

In sera of patients who have received mouse immunoglobulins for therapeutic or diagnostic purposes and also in healthy persons, anti-mouse Ig antibodies may be present which can interfere in assays based on murine monoclonal antibodies /30/.


At least 24 h at 4 °C. Since particular assays may be affected by the protein content of the sample, the recommendations of manufacturers vary significantly with respect to sample stability. Therefore, the manufacturers’ instructions should be followed.

28.13.7 Pathophysiology

CEA is a glycoprotein with a carbohydrate content of about 50%. The molecular weight of CEA isolated from colorectal carcinoma or cultured human colonic carcinoma cells is approximately 180 kDa /31/.

According to the redefined nomenclature for members of the CEA family, CEA is classified as CEACAM5 (carcino-embryonic antigen related cellular adhesion molecule) /32/. CEA consists of an extracellular N-terminal immunoglobulin variable-region-like domain and another six extracellular immunoglobulin constant-region-like domains.

CEA is anchored to the cell membrane via phosphatidylinositol and can be converted from the membrane-bound form to a soluble form by phospholipases. CEA degradation takes place mainly in the liver which may cause elevated serum CEA concentrations in non malignant liver disease. The half life of CEA in serum is 2–8 days.

CEA is a normal component of the colorectal mucosa and, in addition, is found in other epithelia such as the vaginal epithelium and glandular tissues such as gastric foveolae and sweat glands /31/. The highest CEA tissue concentrations are found in primary colorectal cancers and their liver metastases; concentrations can exceed those in normal colonic mucosa by a factor of 500.

CEA is also expressed in other carcinomas such as gastric, breast and lung cancers. In the latter tissues, concentrations are significantly lower than in colorectal cancer /33/.

The gene coding for CEA is the member of a gene family which includes at least 17 transcriptionally active genes of high structural homology /31/. Because of the structural homology of different gene products, monoclonal and polyclonal antibodies which have been raised against CEA may cross react with other members of the CEA family. Granulocytes, macrophages and bile canaliculi contain cross reacting antigens, but no CEA. Thus, monoclonal antibodies used for the specific detection of CEA should not cross react with products from these cells or tissues /34/.

By using transfected cells expressing CEA and other members of the CEA family, it was shown that CEA binds to itself (homophilic binding) and to other family members (heterophilic binding) /31/. The intermolecular association of CEA monomers is mediated by different domains /35/. Since in the colon and other epithelia, members of the CEA family are located at the apical cell poles, it is improbable that the glycoproteins function as cell adhesion molecules. Possibly, CEA and other members of the CEA family are involved in structuring of the glycocalyx or the regulation of bacterial colonization and play a role in preventing entry of pathogens /36/.


1. Reynoso G, Keane M. Carcinoembryonic antigen in prognosis and monitoring of patients with cancer. In: Herberman R, McIntire KR (eds). Immunodiagnosis of cancer, Vol 1. New York: Marcel Dekker, 1979: 239–51.

2. Hansen HJ, Snyder JJ, Miller E, et al. Carcinoembryonic antigen (CEA) assay. A laboratory adjunct in the diagnosis and management of cancer. Hum Pathol 1974; 5: 139–47.

3. Hirai H. A collaborative clinical study of carcinoembryonic antigen in Japan. Cancer Res 1977; 37: 2267–74.

4. Lamerz R, Reithmeier A, Stieber P, Eiermann W, Fateh-Moghadam A. Role of blood markers in the detection of metastases from primary breast cancer. Diagn Oncol 1991; 1: 88–97.

5. Lamerz R. Tumormarker. Prinzipien und Klinik. Dtsch Aerzteblatt 1989; 86: 771–7.

6. Cooper MJ, Mackie CR, Skinner DB, Moossa AR. A reappraisal of the value of carcinoembryonic antigen in the management of patients with various neoplasms. Br J Surg 1979; 66: 120–3.

7. Lamerz P, Stieber P, Borlinghaus P, Fateh-Moghadam A. Tumor markers in cancer of the liver. Diagn Oncol 1991; 1: 363–72.

8. Lamerz R, Borlinghaus P. Differentialdiagnose maligner Lebererkrankungen. Diagnose & Labor 1986; 36: 56–69.

9. Bünger S, Laubert T, Roblick UJ, Habermann JK. Serum biomarkers for improved diagnostic of pancreatic cancer: a current overview. J Cancer. Res Clin Oncol 2011; 137: 375–89.

10. Duffy MJ. Carcinoembryonic antigen as a marker for colorectal cancer: is it clinically useful? Clin Chem 2001; 47: 624–30.

11. Cromheecke M, de Jong KP, Hoekstra HJ. Current treatment for colorectal cancer metastatic to the liver. Eur J Surg Oncol 1999; 25: 451–63.

12. Brommendorf T, Anderer FA, Staab HJ, Hornung A, Kieninger G. Carcinoembryonales Antigen: Diagnose der Tumorprogression bei gastrointestinalen Tumoren. Dtsch Med Wschr 1985; 110: 1963–8.

13. Hall NR, Finan PJ, Stephenson BM, Purves DA, Cooper EH. The role of CA-242 and CEA in surveillance following curative resection for colorectal cancer. Br J Cancer 1994; 70: 549–53.

14. Steele jr G, Ellenberg S, Ramming K, et al. CEA monitoring among patients in multi-institutional adjuvant G.I. therapy protocols. Ann Surg 1982; 196: 162–9.

15. Tate H. Plasma CEA in the post-surgical monitoring of colorectal carcinoma. Br J Cancer 1982; 46: 323–30.

16. Duffy MJ, van Dalen A, Haglund C, et al. Clinical utility of biochemical markers in colorectal cancer: European Group on Tumour Markers (EGTM) guidelines. Eur J Cancer 2003; 39: 718–27.

17. Sturgeon CM, Duffy MJ, Stenman UH, Lilja H, Brünner N, Chan DW, et al. NACB laboratory medicine practice guidelines for use of tumor markers in testicular, prostate, colorectal, breast, and ovarian cancers. Clin Chem 2008; 54: e11–e79.

18. Holt A, Nelson RA, Lai L. Surveillance with serial serum carcinoembryonic levels detect colorectal recurrences in patients who are initial nonsecretors. Am Surgeon 2010; 76: 1100–3.

19. Grossmann I, de Bock GH, Meershoek-Klein Kranenbarg WM, van de Velde CJH, Wiggers T. Carcinoembryonic antigen (CEA) measurement during follow-up for rectal carcinoma is useful even if normal levels exist before surgery. A retrospective study of CEA values in the TME trial. Eur J Surgery 2007; 33: 183–7.

20. Lee WS, Baek JH, Kim KK, Park YH. The prognostic significant of percentage drop in serum CEA post curative resection for colon cancer. Surg Oncol 2010. https://doi.org/10.1016/j.suronc.2010.10.003.

21. Tsikitis VL, Malireddy K, Green EA, Christensen B, Whelan R, Hyder J, et al. Postoperative surveillance recommendations for early stage colon cancer based on results from the clinical outcomes of surgical therapy trial. J Clin Oncol 2008; 27: 3671–6.

22. Strimpakos AS, Cunningham D, Mikropoulos C, Petkar I, Barbachano Y, Chau I. The impact of carcinoembryonic antigen flare in patients with colorectal cancer receiving first-line chemotherapy. Ann Oncol 2010; 21: 1013–9.

23. Chuang SC, Su YC, Lu CY, Hsu HAT, Sun LC, Shih YL, et al. Risk factors for the development of metachronous liver metastasis in colorectal cancer patients after curative resection. World J Surg 2011; 35: 424–9.

24. Matsuoka K, Sumitomo S, Kakashima N, Nakajima D, Misaki N. Prognostic value of carcinoembryonic antigen and CYFRA 21-1 in patients with pathological stage I non-small cell lung cancer. Eur J Cardio-thoracic Surg 2007; 32: 435–9.

25. Ardizzoni ’A, Cafferata MA, Tiseo M, Filiberti R, Marroni P, Grossi, et al. Decline in serum carcinoembryonic antigen and cytokeratin 19 fragment during chemotherapy predicts objective response and survival in patients with advanced nonsmall cell lung cancer. Cancer 2006; 107: 2842–9.

26. Okamoto T, Nakamura T, Ikeda J, Maruyama R, Shoji F, Miyake T, et al. Serum carcinoembryonic antigen as a predictive marker for sensitivity to gefitinib in advanced non-small cell lung cancer. Eur J Cancer 2005; 41: 1286–90.

27. Molina R, Auge JM, Farrus B, Zanon G, Pahlisa J, Munoz M, et al. Prospective evaluation of carcinoembryonic antigen (CEA) and carbohydrate antigen 15.3 (CA 15.3) in patients with primary locoregional breast cancer. Clin Chem 2010; 56: 1148–57.

28. Molina R, Barak V, van Dalen A, Duffy MJ, Einarsson R, Gion M, et al. Tumor markers in breast cancer – European group on tumor markers recommendations. Tumor Biology 2005; 26: 281–93.

29. Neumaier M, Fenger U, Wagener C. Delineation of four carcinoembryonic antigen (CEA) related antigens in normal plasma by transblot studies using monoclonal anti-CEA antibodies with different epitope specificities. Mol Immunol 1985; 22: 1273–7.

30. Wagener C, Wickert L, Meyers W. Limited improvement of tumour diagnosis by the simultaneous determination of carcinoembryonic antigen (CEA) and of a tumour-associated CEA-related antigen of Mr 128,000 in serum. J Clin Chem Clin Biochem 1989; 27: 643–52.

31. Hammarström S. The carcinoembryonic antigen (CEA) family: structures, suggested functions and expression in normal and malignant tissues. Semin Cancer Biol 1999; 9: 67–81.

32. Nomenclature announcement. Redefined nomenclature for members of the carcinoembryonic antigen family. Exp Cell Res 1999; 252: 243–9.

33. Wagener C, Müller-Wallraf R, Nisson S, Gröner J, Breuer H. Localization and concentration of carcinoembryonic antigen (CEA) in gastrointestinal tumors: correlation with CEA levels in plasma. J Natl Cancer Inst 1981; 67: 539–47.

34. Nap M, Hammarström ML, Børmer O, et al. Specificity and affinity of monoclonal antibodies against carcinoembryonic antigen. Cancer Res 1992; 52: 2329–39.

35. Gaida FJ, Pieper D, Roder UW, Shively JE, Wagener C, Neumaier M. Molecular characterization of a cloned idiotypic cascade containing a network antigenic determinant specific for the human carcinoembryonic antigen (CEA). J Biol Chem 1993; 268: 14138–45.

36. Frängsmyr L, Baranov V, Prall F, Yeung MM-W, Wagener C, Hammarström S. Cell- and region-specific expression of biliary glycoprotein and its messenger RNA in normal human colonic mucosa. Cancer Res 1995; 55: 2963–7.

28.14 Cyfra 21-1

Lothar Thomas

The epithelial cytokeratins constitute a family of 20 polypeptides distinguishable by their molecular weight and isoelectric point. The cytokeratin family is expressed by all epithelial cells and constitutes a useful marker of epithelial differentiation. During malignant transformation of normal epithelial cells the cytokeratin patterns are usually maintained. Thus, the most abundant cytokeratins in carcinomas , the cytokeratins 7, 8, 18, 19, are also found in simple epithelia /1/.

Cytokeratin 19 is an acidic subunit with a molecular weight of 40 kDa. The keratin is expressed in the cytoplasm of several epithelial tumors including lung cancer. Following proteolytic degradation of the N-terminal and C-terminal regions, the cytokeratin 19 fragment is soluble and released in the blood because of cell lysis and tumor necrosis.

Cytokeratin 19 fragments are present throughout the body, but are primarily produced in the lung. They are mainly excreted via the kidneys. Therefore, renal insufficiency can lead to elevated serum concentrations. The cytokeratin 19 fragment is referred to as Cyfra 21-1 and is a new cytoskeleton marker /1/.

28.14.1 Indication

  • Differential diagnosis, prognosis, monitoring post-surgery, follow-up and detection of recurrence in lung cancer [recommendations of the European Group of Tumor Markers (ETG) and the National Academy of Clinical Biochemistry (NACB)] /2/
  • Differential diagnostic discrimination of lung cancer from pulmonary round lesions of uncertain etiology /3/
  • Use in combination with other tumor markers in lung cancer (Tab. 28.14-1 – Use of CYFRA in combination with other tumor markers in lung cancer/3/.
  • Monitoring of bladder cancer /4/.

28.14.2 Method of determination

Enzyme immunoassay, immunoradiometric assay. All CYFRA 21-1 assays use the two mouse monoclonal antibodies BM 21-1 and KS 19-1 which are directed specifically against the cytokeratin fragment 19 /5/. Nevertheless, the results obtained with test kits from different manufacturers show a moderate correlation.

28.14.3 Specimen

Serum, pleural exudate: 1 mL

28.14.4 Reference interval

95% of healthy controls have levels < 2.0 μg/L /6/

95% of patients with benign pulmonary diseases have levels < 3.3 μg/L /6/

28.14.5 Clinical significance

Cytokeratin 19 is not organ specific. Therefore, Cyfra 21-1 is not only elevated in lung cancer, but also in other solid tumors (Tab. 28.14-2 – Diagnostic sensitivity of Cyfra 21-1 in malignant disease). Elevated concentrations may also be found in benign diseases. Therefore, such patients have different cutoffs (Tab. 28.14-3 – Cyfra 21-1 cutoffs in healthy individuals and in benign disease). Benign diseases

In 80% of cases, healthy individuals have serum Cyfra 21-1 levels below 1.5 μg/L. Cyfra 21-1 concentrations above 10 μg/L are associated with benign disease only in rare cases (< 1%). Cyfra 21-1 concentrations in benign diseases are shown in Tab. 28.14-4 – Cyfra 21-1 in benign diseases. Lung cancer

Cyfra 21-1 is the most sensitive tumor marker in NSCLC, especially in squamous cell carcinoma (SCCA) (Tab. 28.14-5 – Cyfra 21-1 in non small cell lung cancer (NSCLC)). Although no meta-analysis is available, the diagnostic value of this marker has been demonstrated by numerous studies /9/. For instance, the marker is recommended for diagnosis of NSCLC and for differential diagnosis of lung cancer if biopsy is not possible. Moreover, Cyfra 21-1 is determined for prognosis in early and late stages of NSCLC /10/. Concentrations above 3.3 μg/L have a diagnostic sensitivity of 59% and a specificity of 94% in NSCLC patients /11/. The diagnostic sensitivity of Cyfra 21-1 for small cell lung cancer (SCLC) is reported as 46–61% given a diagnostic specificity of 95% for benign pulmonary diseases /12/. Fig. 28.14-1 – Comparison of tumor markers for the differentiation between SCLC and benign lung disease shows a comparison of various tumor markers for the differentiation between SCLC and benign lung disease. Bladder cancer

Bladder cancer is the second most common malignancy in the urogenital tract, with urothelial carcinoma accounting for almost 90% of cases. At clinical presentation, 70–80% of patients have superficial cancers (Ta, T1), while the rest have bladder muscle invasive (T2–T4) or metastatic cancers. Early diagnosis of a muscle invasive tumor or metastasis is significant for prognosis.

Taking into account all stages of urinary bladder cancer, the diagnostic sensitivity of Cyfra 21-1 and other tumor markers is not satisfactory. Considering only the muscle invasive carcinomas, the profile will be different: the diagnostic sensitivity of Cyfra 21-1 is higher and increases to 52–56% (TPA 39–42%, TPS 31%). The values and the positive rates of a study /15/ are shown in Tab. 28.14-6 – Cyfra 21-1 in the diagnosis of urinary bladder cancer.

Diagnostic sensitivities of Cyfra 21-1 are 4–16% in stage 0 and 71–73% in stage IV /16/. Because of its high diagnostic sensitivity, Cyfra 21-1 is useful for the detection of recurrent muscle invasive cancers. Cyfra 21-1 in pleural exudate

The Cyfra 21-1 concentration is higher in pleural exudate of malignant origin than in that of benign origin. No distinction is possible between primary lung cancer and pulmonary metastasis of other primary cancers /17/. Given a cutoff of 20.9 μg/L, the diagnostic specificity for lung cancer is 71% with a positive predictive value of 82%.

28.14.6 Comments and problems

Method of determination

Although currently available assays employ antibodies (BM 21-1 and KS 19-1) identical to those first described and tested, it should not always be assumed that CYFRA 21-1 assays will give comparable test results.

Influencing factors

The concentration of Cyfra 21-1 is not dependent on biological factors such as age, gender and smoking habits /18/. It also does not show dependence on the menstrual cycle /8/.

During pregnancy, the median Cyfra 21-1 level is about 1.4 μg/L and thus within the reference interval (< 2.0 μg/L) up to the 39th week of gestation. During the 39th–40th week of gestation, significant increases in Cyfra 21-1 occur (median 3.4 μg/L), possibly caused by uterine contractions as part of the normal course of pregnancy or premature labor /19/.

Markedly elevated Cyfra 21-1 levels can occur immediately after endotracheal intubation and in conjunction with long term positive pressure ventilation as well as due to any injury involving cytokeratin rich tissue (massive trauma, surgery).


Cyfra 21-1 is stable in whole blood for up to one week at room temperature; nonetheless early separation of the serum is generally recommended. Stability in serum at –20 °C or –80 °C is for several years.

Interference factors

Contamination of the sample by saliva may cause falsely elevated Cyfra 21-1 levels /8/. Hemolysis, jaundice and hyperlipidemia do not interfere with Cyfra 21-1 determination.

Biological half life

Biological half life is 2–5 h. In general, a postoperative control determination should be performed after 2–3 days.

28.14.7 Pathophysiology

The cytoskeleton is a network of filaments with influence on the morphology of cells in their body environment /3/. It maintains the structural integrity of the cells, anchoring intra cytoplasmic organelles to the cellular membranes. The following structures compose the cytoskeleton of the cells /1/:

  • Filaments with a diameter of 60 Angstrom
  • Filaments with a diameter of 150 Angstrom
  • Micro tubules of myosin (250 Angstrom)
  • An intermediate system of fibrous filaments composed of chemically heterogenous subunits.

The subunit structure defines five major classes of intermediate filaments /1/:

  • Keratin filaments, found in epithelial cells and in cells of epithelial origin
  • Desmin filaments, found predominantly in smooth skeletal and cardiac muscle cells
  • Vimetin filaments, found in mesenchymal cells and cells of mesenchymal origin
  • Neurofilaments, found in neurons
  • Glial filaments, found in all types of glial cells.

Cytokeratins are proteins which, together with actin filaments and micro tubuli, form the cytoskeleton of epithelial cells /20/. Like vimentin and desmin, they have long been established in histopathology as intermediate filaments for the differentiation between physiological and abnormal tissue.

In contrast to non soluble cytokeratins, fragments of cytokeratins are soluble and, therefore, can be detected in serum.

Although cytokeratin 19 is not considered to be organ specific or tumor specific, it is characterized by a more restrictive distribution pattern within the body than other cytokeratins. It occurs primarily in pulmonary tissue and, in particular, in malignant lung cancers. This provides the basis for the clinical relevance of Cyfra 21-1 and its high significance as a so called pan marker in lung cancer.

The occurrence or a certain combination of individual cytokeratins is tissue characteristic /20/:

  • Multilayered epithelium, such as squamous cell epithelium, is characterized by cytokeratins 1–6 and 9–17
  • Single layer columnar epithelium is defined by cytokeratins 8 and 18
  • Glandular epithelium is characterized by cytokeratins 7, 19 and 20.

An individual, differentiation independent combination of two or more cytokeratin polypeptides is expressed by each epithelial cell. The polypeptides of the acid type I keratins (cytokeratins 9–20) and the alkaline type II keratins (cytokeratins 1–8) always occur in dimers in this combination at a ratio of 1 : 1.

Two dimers form a tetramer complex, tetramers polymerize into protofilaments. Eight protofilaments then form an intermediate filament. The secondary structure of cytokeratins is composed of an α-helical mid-section, the so called rod domain, which has an important function in the heterodimerization of type I and type II cytokeratins, as well as the amino terminal head and the carboxyl terminal tail.

Cytokeratin 19 with a molecular weight of 36 kDa is the smallest cytokeratin. Elevated cytokeratin fragment 19 concentrations are present in malignant diseases and, in particular, in NSCLC.


1. Buccheri G, Ferrigno D. Lung tumor markers of cytokeratin origin. Lung Cancer 2001; 34: S65–S69.

2. Stieber P, Holdenrieder S. Lung cancer biomarkers – where we are and what we need. Cancer Biomarkers 2010; 6: 221–4.

3. Stieber P. CYFRA 21-1 (Cytokeratin-19-Fragmente). In Thomas L, ed. Labor und Diagnose. TH-Books, Frankfurt 2008: S. 1323–8.

4. Andreadis C, Touloupidis S, Galaktidou G, Kortasaris AH, Boutis A, Mouratidou D. Serum Cyfra 21-1 in patients with invasive bladder cancer and its relevance as a tumor marker during chemotherapy. J Urol 2005; 174: 1771–5.

5. Bodenmüller H. The biochemistry of Cyfra 21-1 and other cytokeratin tests. Clin Lab Invest 1995; 55, suppl 221: 60–6.

6. Stieber P, Hasholzner U, Bodenmüller H, Nagel D, Sunder-Plassmann L, Dienemann H, Meier W, Fateh-Moghadam A. CYFRA 21-1: a new marker in lung cancer. Cancer 1993; 72: 707–13.

7. Molina R, Agusti C, Filella X, Jo J, Joseph J, Gimenez N, Ballesta A. Study of a new tumor marker, CYFRA 21-1, in malignant and nonmalignant diseases. Tumor Biol 1994; 15: 318–25.

8. Stieber P, Poley S. Tumormarker: Gibt es Befunde ohne Krankheitswert? Der Bay Int 1996; 16: 22–31.

9. Molina R, Holdenrieder S, Auge JM, Schalhorn A, Hatz R, Stieber S. Diagnostic relevance of circulating biomarkers in patients with lung cancer. Cancer Biomarkers 2009/2010; 6: 163–78.

10. Pujol JL, Molinier O, Ebert W, Daures JP, Barlesi F, Buccheri G, et al. CYFRA 21-1 is a prognostic determinant in non-small-cell lung cancer: results of a metaanalysis in 2063 patients. Br J Cancer 2004; 90: 2097–2105.

11. Wieskopf B, Demangeat C, Purohit A, Stenger R, Gries P, Kreisman H, Quoix E. CYFRA 21-1 as a biological marker of non-small cell lung cancer. Chest 1995; 108: 163–9.

12. Pujol JL, Grenier J, Daures JP, Daver A, Pujol H, Michel FB. Serum fragment of cytokeratin subunit 19 measured by CYFRA 21-1 immunoradiometric assay as a marker of lung cancer. Cancer Res 1993; 53: 61–6.

13. Stieber P, Dienemann H, Nagel D, Hasholzner U, Reinmiedl J, Zimmermann A, Hofmann K, Fateh-Mog­hadam A. Prognostic relevance of CYFRA 21-1 in lung cancer – a multivariate analysis. Anticancer Res 1995; 15 (6A): 2390.

14. Dienemann H, Stieber P, Zimmermann A, Hofmann H, Müller C, Banauch D. Tumor-marker CYFRA 21-1 in non small cell lung cancer (NSCLC): role for detection of recurrence. Lung Cancer 1994; 11 Suppl 1: 46.

15. Stieber P, Schmeller N, Schambeck Ch, Hofmann K, Reiter W, Hasholzner U, Fateh-Moghadam A. Clinical relevance of CYFRA 21-1, TPA-IRMA and TPA-LIA-mat in urinary bladder cancer. Anticancer Res 1996; 16: 3793–8.

16. Washino S, Hirai M, Matsuzaki A, Kobayashi Y. Clinical usefulness of CEA, CA 19-9, and CYFRA 21-1 as tumor markers for urothelial bladder carcinoma. Urol Int 2011; 87: 420–8.

17. Satoh H, Sumi M, Yagyu H, Ishikawa H, Suyama T, Naitoh T, Saizoh T, Hasegawa S. Clinical evaluation of CYFRA 21-1 in malignant pleural fluids. Oncology 1995; 52: 211–4.

18. Ebert W, Dienemann H, Fateh-Moghadam A, Scheulen M, Konietzko N, Schleich T, Bombardieri E. Cytokeratin 19-fragment CYFRA 21-1 compared with carcinoembryonic antigen, squamous cell carcinoma antigen and neuron specific enolase in lung cancer. Results of an international multicenter study. Eur J Clin Chem Clin Biochem 1994; 32: 189–99.

19. Fiebig M, Stieber P, Knitza R, Hofmann K, Nagel D, Fateh-Moghadam A. Tumorassoziierte Antigene in der Schwangerschaft. Der Bay Int 1994; 14: 52–6.

20. Broers JL, Ramaekers FC, Rot MK, Oostendorp T, Huysmans A, van Muijen GN, Wagenaar SS, Vooijs GP. Cytokeratins in different types of human lung cancer as monitored by chain-specific monoclonal antibodies. Cancer Res 1988; 48: 3221–9.

28.15 Des-gamma carboxy prothrombin (DCP)

Rolf Lamerz

DCP, also known as protein induced by vitamin K absence/antagonist II (PIVKA II), is an abnormal, nonfunctional prothrombin resulting from a lack of carboxylation in the liver. In hepatocellular carcinoma (HCC), DCP induces cell proliferation and migration and stimulates angiogenic factors. DCP is a tumor marker in patients with chronic hepatitis and liver cirrhosis and suspected HCC or in HCC patients for therapy monitoring and follow-up /123/.

28.15.1 Indication

Suspected hepatocellular carcinoma (e.g., in patients with liver cirrhosis).

Therapy monitoring and follow-up of patients with primary hepatocellular carcinoma (e.g., post-surgery) especially after curative ablative therapy or trans arterial chemoembolization (TACE).

28.15.2 Method of determination

Immunoassays such as PIVKA-II kit /4/ and electrochemiluminescence immunoassay /5/ and liquid-phase binding assay /6/ were evaluated /7/. Some assays use arbitrary units (AU) and others measure the concentration in ng/mL. 1 mAU corresponding to 0.019 ng is the conversion factor applicable to the liquid-phase binding assay. Asserachrom PIVKA-II (monoAb P1-2-B9) is another EIA measuring the DCP in ng/mL.

28.15.3 Specimen

Plasma, serum, pleural exudate, ascitic fluid: 1 mL

28.15.4 Reference interval

Up to 40 mAU/mL (0.8 ng/mL)

According to commercially available kits (Eitest PIVKA-II, Picolumi PIVKA-II) 40 mAU/mL or (LiBASys) in ng/mL /456/.

28.15.5 Clinical significance

DCP determination is suited for the screening/surveillance of risk groups (chronic hepatitis B and C and liver cirrhosis) with suspected development of HHC and for postoperative monitoring to enable early recognition of recurrence, in particular, after ablative therapy or TACE /89, 1011/.

DCP sensitivities and specificities

In a review /12/ of a summary compilation of 11 publications between 1983 and 1991 using various methods of DCP determination on 750 HCC patients, a diagnostic sensitivity of 66% was found, but in tumors smaller than 3 cm the sensitivity was only 18%. The cutoff in the most sensitive immunoassay was 0.1 AU/mL, corresponding to 100 ng/mL.

A comparison of DCP and AFP in patients with HCC and in patients with liver cirrhosis showed diagnostic sensitivities of 28–89% and 47–68%, respectively, and specificities of 87–96% and 82–97%, respectively /1314, 15, 1617/.

Moreover, in follow-up studies involving 78–734 cases of cirrhosis patients during a follow-up period of 13–48 months, early HCC was detected in 14 to 35 cases with a diagnostic sensitivity of 23–57% for DCP and 14–54% for AFP with a diagnostic specificity (for the largest study /18/) of 90% for DCP and 62% for AFP /181920/.

Histopathological assessment

In a histopathological analysis of HCC resection samples for serum AFP-L3 and DCP concentrations in 111 HCC patients, positivity for AFP-L3 (cutoff > 10 μg/L) was found in 38 and for DCP in 63 patients.

AFP-L3-positive HCC was characterized by an infiltrative growth type and a poorly differentiated type compared to AFP-L3-negative HCC.

The frequencies of infiltrative growth type, vascular invasion and intrahepatic metastasis in DCP-positive HCC were significantly higher than in DCP-negative HCC.

In AFP-L3-positive and DCP-positive HCC, the frequency of a poorly differentiated growth type was higher than in HCC positive for either AFP-L3 or DCP and HCC negative for both AFP-L3 and DCP.

In all, AFP-L3 was related to progression from moderately differentiated to poorly differentiated HCC, whereas DCP was a useful indicator of vascular invasion /21/.

The tissue DCP expression and the serum DCP concentration reflect the malignant potential of HCC and thus may be useful indications for the prognosis of small HCC /22/. Serum DCP can also originate from other sources than from HCC tissue /23/. The origin of elevated serum DCP may lie not only in HCC tissue but also in non-cancer tissues. The HCC lesion itself appears to influence the production of DCP in surrounding non-cancer tissues /24/.

Serum and tissue DCP concentrations correlated with the DCP expression levels in HCC and non-HCC tissue /25/. Both elevated tissue and serum DCP concentrations indicated poor outcome. In the multi variant analysis, DCP expression in the entire liver tissue in combination with intrahepatic metastasis were significant prognostic factors. Hepatocellular carcinoma screening

Numerous clinical studies have been performed to determine the usefulness of DCP in HCC screening /18, 14, 15, 17, 1920/. In patients with chronic liver disease, serum DCP concentrations were not elevated compared to AFP and are therefore highly specific to HCC with varying diagnostic sensitivity. In particular, the sensitivity was rather low in small HCC, for example it was 35% and 39.3% in HCC smaller than 2 cm and 3 cm, respectively, with a 60% diagnostic sensitivity for all HCC /26/. Therefore, other complementary tumor markers such as AFP and AFP-L3 were used /151719/.

According to a newer study /27/, DCP (cutoff 150 mAU/mL) was more sensitive (diagnostic sensitivity 74%) for the detection of HCC than AFP (cutoff 20 μg/L; sensitivity 59%) and AFP-L3 (cutoff 10%; sensitivity 42%) with specificities of 70%, 90% and 97%, respectively.

Among 1,031 patients randomized in the HALT-C trial, a case-control study with 39 HCC cases (24 early stage) and 77 matched controls was conducted on sera from 12 months prior (month –12) to the time of HCC diagnosis (month 0) to analyze the serum AFP (cutoff 150 μg/L) and DCP (cutoff 40 mAU/mL) concentrations, with the following results /28/:

Diagnostic sensitivity of 74% and specificity of 86% for DCP (cutoff 40 mAU/mL) at month 0

  • Diagnostic sensitivity of 43% and specificity of 100% for DCP (cutoff 150 mAU/mL) at month 0
  • Diagnostic sensitivity of 61% and specificity of 81% for AFP (cutoff value 20 μg/L) at month 0
  • Diagnostic sensitivity of 43% and specificity of 94% for AFP (cutoff 200 μg/L) at month 0
  • At month –12, the sensitivity and specificity at the low cutoffs were 43% and 94%, respectively, and for DCP and 47% and 75%, respectively, for AFP
  • Combining both markers increased the diagnostic sensitivity to 91% at month 0 and 73% at month –12, but the specificity decreased to 74% and 71%, respectively.

The Japanese consensus-based clinical practice guideline recommends for HCC screening to perform simultaneous, combined determinations of DCP, AFP and AFP-L3 at regular intervals, especially in small HCC tumors /29/. DCP in the selection of liver transplant recipients

DCP is recommended as a variable for extended recipient selection criteria in living donor liver transplantation (LDLT) in addition to established criteria, especially in Japanese groups. Elevated DCP and AFP concentrations correlate significantly with the incidence of positive vascular invasion and advanced tumor stage and are strong independent factors for poor prognosis after transplantation in HCC patients. For instance, disease-free survival was significantly improved in patients with DCP ≤ 300 mAU/mL and a tumor diameter ≤ 5 cm compared to higher values /30/. Besides the Milan criteria, DCP above 100 mAU/mL and AFP above 200 μg/L proved to be reliable predictors for poor prognosis following LDLT /31/. According to a different multivariate analysis, only a high DCP concentration was found to be an independent risk factor for the recurrence of HCC after LDLT /32/. Role of DCP in various therapies

In a multivariate analysis performed within the scope of a prospective study on preoperative AFP and DCP doubling times compared to 19 clinical factors involving HCC patients before hepatectomy, patients with doubling times ≤ 30 days for AFP and ≤ 16 days for DCP showed a significantly worse disease-free and overall survival. Thus, AFP and DCP proved to be useful tools to predict early postoperative recurrence and poor prognosis /33/.

In patients who were diagnosed as initial HCC and underwent different types of curative treatment (hepatectomy or loco regional thermal ablation), pretreatment elevation of AFP-L3 and, in particular, DCP significantly affected decreased survival rate and, thus, had high prognostic values only in patients treated with LTA /34/.

In a retrospective analysis on HCC patients diagnosed with stage Child-Pugh A liver cirrhosis who underwent hepatic surgery or radio frequency ablation (RFA) with a maximum tumor diameter ≤ 3 cm and tumors numbering ≤ 3, the 3- and 5-year survival rates were 90.3%/79.0% and 87.4%/74.8%, respectively, for both treatment procedures. However, the 1- and 3-year tumor recurrence-free survival rates of the resection group (83.1%/51.0%) were higher than in the RFA group (82.7%/41.8%).

Multivariate analysis identified prothrombin time ≥ 80% (hazard ratio 2.72) as an independent prognostic factor for survival in the resection group and DCP < 100 AU/mL (hazard ratio 5.49) and a thrombocyte count ≥ 100 × 109/l (hazard ratio 1.26) for the RFA group.

Since high DCP levels reflect the biologic aggressiveness and progression of HCC tumors, surgical resection rather than RFA was recommended for such patients /35/. Function, correlation and prognosis

DCP, a protein induced by vitamin K absence/antagonist II, is an abnormal carboxylation product during prothrombin formation. DCP acts as an autologous mitogen for HCC cell lines /36/. This leads to the formation of abnormal prothrombin without carboxylation of the 10 glutamic acid residues at the N terminus and no coagulation activity.

However, DCP has a different biological effect in HCC and acts as an autocrine/paracrine growth stimulation factor /36/, by stimulating cell proliferation in HCC cell lines along the Met-Janus kinase 1 signal transducers and activators of the transcription 3 (STAT3) signaling pathway. DCP is thought to induce human umbilical vein endothelial cell (HUVEC) proliferation and migration, as indicated by the correlation between the cell proliferation marker PCNA and DCP over expression in HCC tissue /3637/. Hence, besides acting as a growth stimulation factor, DCP also increases the gene expression of angiogenic factors such as EGF-R, VEGF and MMP-2.

DCP is considered to be a prognostic biomarker and predictor of fast tumor progression and poor prognosis /838/. It is used as a pathological and prognostic indicator for HCC patients and appears to reflect the invasive role of HCC better than AFP. The concurrent determination of serum and tissue DCP concentrations is even more helpful for predicting the prognosis in HCC patients than the determination of serum DCP alone.

Patients who are seropositive for DCP and seronegative for AFP show a higher incidence of HCC with distinct margin, large tumors more than 3 cm in diameter, fewer nodules or a higher frequency of moderately to poorly differentiated HCC /39/. DCP correlates with the HCC stage better than AFP.

The frequency of intrahepatic metastasis, portal vein tumor invasion, hepatic vein thrombosis and capsular infiltration is higher in patients with positive DCP /89, 1011/ and survival is poorer than in those with negative DCP /40/. Therefore, DCP is suggested as a marker in portal vein invasion and intrahepatic metastasis /41/.

A study involving HCC patients with simultaneous elevations of DCP and AFP-L3 showed higher frequencies of infiltrative growth type, vascular invasion and intrahepatic metastasis than in patients negative for DCP /42/. Thus, the combination of DCP and AFP-L3 is likely to be an especially effective marker for the diagnosis and differential diagnosis of HCC.

Other studies investigated the usefulness of DCP as an indicator of recurrent HCC following curative treatment. Positive serum DCP levels correlated significantly with clinicopathological factors such as vascular invasion, intrahepatic metastasis, tumor size and TNM stage, tumor recurrence, and frequently with tumor recurrence with a poor overall survival rate and disease-free survival rate /24/.

In a prospective study, elevated DCP correlated significantly more often with portal vein invasion (PVI) compared to negative findings and was the most useful predisposing clinical marker for the development of PVI /10/. DCP can also be used as a prognostic indicator in small HCC < 3 cm. High DCP concentrations are associated with a higher risk of HCC recurrence and poorer overall survival /43/. Monitoring using DCP and other tumor markers

DCP, like AFP is a good marker for monitoring hepatocellular carcinoma (HCC ) clearance following curative treatment and for detecting HCC recurrence. However, there are differences between DCP and AFP:

  • DCP and AFP correlate neither positively nor negatively; 30% of patients who are negative for AFP are positive for DCP
  • DCP is a specific HCC marker compared to AFP because of the lower incidence of elevated concentrations in other liver diseases
  • The plasma half-life of DCP (40–72 h) is shorter than that of AFP (5–7 days). Hence, DCP reflects the therapeutic efficacy in HCC within a shorter period of time.

The combined determination of DCP and AFP was not superior to the determination of DCP alone in one study /12/, but was considered to be more useful in another study /40/. Moreover, the simultaneous determination of DCP and AFP-L3 has also been recommended as being more efficient for the early diagnosis of HCC /19/.

In a study, DCP (cutoff 84 mAU/mL), AFP (cutoff 25 ng/mL) and AFP-L3 (cutoff 10%) were determined in 144 patients with HCC, 47 cases with chronic hepatitis B and C and 49 cases with liver cirrhosis. The concentration of all three markers was significantly higher in HCC with the following diagnostic sensitivities, specificities and positive predictive values:

  • 87%, 85% and 86.8% for DCP
  • 69%, 87% and 69.8% for AFP
  • 56%, 90% and 56.1% for AFP-L3.

DCP proved to be the best HCC biomarker because of better marker rates, direct correlation with tumor size and normal concentrations in non HCC disease /43/.

In HCC patients with increases in AFP to 20–200 μg/L, AFP-L3 and DCP were highly specific biomarkers (diagnostic specificity 86.6% and 90.2%, respectively): of 29 HCC patients with AFP levels < 20 μg/L, 13 had increased AFP-L3 or DCP concentrations. Both AFP-L3- and DCP positive patients showed significant differences in lower cumulative 1- and 2-year HCC free rates compared with the overall group /44/.

A prospective study /45/ on AFP-L3 (cutoff 10%), DCP (cutoff 200 mAU/mL) and AFP (cutoff 25 μg/L) in patients with histologically proven HCC showed diagnostic sensitivities of 61.6%, 72.7%, 67.7%, respectively, and 85.9% for a combination of all three markers. Significant differences were observed for portal vein invasion in AFP-L3 and AFP concentrations, whereas DCP was significantly associated with metastasis.

A retrospective study /46/ on HCC patients with and without microvascular invasion (MVI) who underwent hepatectomy had the following results in univariate and multivariate analyses: age under 65 years, DCP ≥ 200 mAU/mL, preoperative tumor size ≥ 5 cm and poorly differentiated carcinoma were independent predictors of MVI. When age, DCP and tumor size were scored as a preoperative combined index, the total score demonstrated significant correlation with the extent of MVI and with survival after hepatic resection.

28.15.6 Comments and problems

DCP concentrations can be affected by prolonged jaundice, intrahepatic cholestasis with vitamin K deficiency and the administration of warfarin (dicumarol) or antibiotics /47/.

28.15.7 Pathophysiology

The prothrombin molecule is synthesized in the liver depending on the presence of vitamin K dependent γ-glutamyl carboxylase. The prothrombin precursor has 10 glutamine residues in the N terminus that are converted into γ-carboxy glutamic acid (Gla) residues by enzymatic activity of γ-glutamyl carboxylase. All of these glutamine residues must be converted into Gla before thrombin can obtain coagulation activity. This function is assumed post translationally.

In DCP, not all of the 10 glutamic acid residues are transformed to Gla residues. Instead, some remain as glutamic acid residues.

Reportedly, vitamin K and γ-glutamyl carboxylase are decreased significantly in HCC tissue /36/.

Similarly to the hepatocyte growth factor, DCP was found to bind cell surface receptor Met also causing HCC cell proliferation /36/. The Met-Janus kinase 1-STAT3 signaling pathway may be a major signaling pathway for DCP-induced cell proliferation.

The addition of vitamin K causes a significant decrease in DCP secretion in HCC cell lines, similarly to the dose-dependent inhibiting effect of vitamin K on the growth and, thus, malignity, of HCC cell lines /48/.

Another biological effect of DCP on HCC malignity based on its potential of stimulating angiogenesis around HCC tissue was investigated using human umbilical vein endothelial cells (HUVEC) /37/. According to these investigations, DCP stimulates the DNA synthesis and the migration activity of HUVEC, but does not stimulate normal prothrombin. DCP was found to bind with the kinase domain receptor (KDR), alternatively referred to as vascular endothelial growth factor receptor (VEGF-R), and stimulates the KDR phospholipase C-(PLC-γ) and mitogen activated protein kinase (MAPK) signaling pathway, causing enhanced DNA synthesis and cell migration /44/.

The potential of DCP to stimulate proliferation and invasive activity of HUVEC cells and induce over expression and secretion of EGFR, VEGF, TGF-α and bFGF in HUVEC cells and HCC cells was also demonstrated in vitro /49/.


1. Inagaki Y, Tang W, Masatoshi M, Hasegawa K, Sugawara Y, Kokudo N. Clinical and molecular insights into the hepatocellular carcinoma tumour marker des-γ-carboxyprothrombin. Liver International 2011; 31: 22–35.

2. Masuda T, Miyoshi E. Cancer biomarkers for hepatocellular carcinomas: from traditional markers to recent topics. Clin Chem Lab Med 2011; 49: 959–66.

3. Malaguarnera G, Giordano M, Paladina I, Berretta M, Cappelani A, Malaguernera M. Serum Markers of hepatocellular carcinoma. Dig Dis Sci 2010; 55: 2744–55.

4. Suzuki H, Akahane Y, Tanaka M, ’Tankikawa K, Okuda H, Saito A, et al. Clinical evaluation of PIVKA-II kit (ED-036). Kann Tan Sui 1996; 33: 1069–76.

5. Takatsu K, Nakanishi T, Watanabe K, Okuda H, Saito A, Tanaka M, et al. Development and performance of an assay kit for PIVKA-II (ED-038) by ECL technique. Jpn J Clin Exp Med 1996; 73: 2656–64.

6. Yamaguchi I, Nakamura K, Hiromichi K, Masuda Y, Kanke F, Kobatake S, et al. Development of des-γ-carboxy prothrombin (DCP) measuring reagent using the LiBASys clinical analyzer. Clin Chem Lab Med 2008; 46: 411–6.

7. Owen WE, Roberts RF. Letter. Performance characteristics of the LiBASys des-γ-carboxy prothrombin assay. Clin Chim Acta 2008; 389: 183–5.

8. Suehiro T, Sugimachi K, Matsumata T, Itasaka H, Taketomi A, Maeda T. Protein induced by vitamin K absences or antagonist II as a prognostic marker in hepatocellular carcinoma. Comparison with alpha-fetoprotein. Cancer 1994; 73: 2464–71.

9. Gotoh M, Nakatani T, Masuda T, et al. Prediction of invasive activities in hepatocellular carcinomas with special reference to α-fetoprotein and des-gamma-γ-carboxy-prothrombin. Jpn J Clin Oncol 2003; 33: 522–6.

10. Kioke Y, Shiratori Y, Sato S, et al. Des-γ-carboxy prothrombin as a useful predisposing factor for the development of portal venous invasion in patients with hepatocellular carcinoma. Cancer 2001; 91: 561–9.

11. Sakon M, Onden M, Gotoh M, et al. Relationship between pathologic prognostic factors and abnormal levels of des-gamma-carboxy prothrombin and alpha-fetoprotein in hepatocellular carcinoma. Am J Surg 1992; 163: 251–6.

12. Weitz IC, Liebman HA. Des-gamma-carboxy (abnormal) prothrombin and hepatocellular carcinoma. Hepatology 1993; 18: 990–7.

13. Marrero JA, Lok ASF. Newer markers for hepatocellular carcinoma. Gastroentetology 2004; 127: S113–S119.

14. Marrero JA, Su GL, Wei W, Emick D, Conjeevaram HS, Fontana RJ, et al. Des-gamma carboxyprothrombin can differentiate hepatocellular carcinoma from non-malignant chronic liver disease in American patients. Hepatology 2003; 37: 1114–21.

15. Nomura F, Ishijima M, Kuwa K, Tanaka T, Ohnishi K. Serum des-gamma-carboxy prothrombin levels determined by a new generation of sensitive immunoassays in patients with small-sized hepatocellular carcinoma. Am J Gastroenterol 1999; 94: 650–4.

16. Aoyagi M, Yanagi M, Suda T, Suzuki Y, Asakura H. The usefulness of determining des-gamma-carboxyprothrombin by sensitive enzyme immunoassay in the early diagnosis of patients with hepatocellular carcinoma. Cancer 1998; 82: 1643–8.

17. Lamerz R, Runge M, Stieber P, Meissner E. Use of serum PIVKA-II determination for differentiation between benign and malignant liver diseases. Anticancer Res 1999; 19: 2489–93.

18. Ishii M, Gama H, Chida N, Ueno Y, Shinzawa H, Takagi T, et al. Simultaneous measurements of serum alpha-fetoprotein and protein induced by vitamin K absence for detecting hepatocellular carcinoma. South Tohoku District Study Group. Am J Gastroenterol 2000; 95: 1036–40.

19. Shimauchi Y, Tanaka M, Kuromatsu R, Ogata R, Tateishi Y, Itano S, et al. A simultaneous monitoring of Lens culinaris agglutinin A-reactive alpha-fetoprotein and des-gamma-carboxy prothrombin as an early diagnosis of hepatocellular carcinoma in the follow-up of cirrhotic patients. Oncol Rep 2000; 7: 249–6.

20. Ikoma J, Kaito M, Ishihara T, Nakagawa N, Kamei A, Fujita N, et al. Early diagnosis of hepatocellular carcinoma using a sensitive assay for serum des-gamma carboxyprothrombin: a prospective study. Hepatogastroenterology 2002; 49: 235–8.

21. Miyaaki H, Nakashima O, Kurogi M, Eguchi K, Kojiro M. Lens culinaris agglutinin-reactive alpha-fetoprotein and protein induced by vitamin K absence II are potential indicators of a poor prognosis: a histopathological study of surgically resected hepatocellular carcinoma. J Gastroenterol 2007; 42: 962–8.

22. Tamano M, Sugaya H, Oguma M, et al. Serum and tissue PIVKA-II expression reflect the biological malignant potential of small hepatocellular carcinoma. Hepatol Res 2002; 22: 261–9.

23. Shimada M, Yamashita Y, Hamatsu T, et al. The role of des-gamma-carboxy prothrombin levels in hepatocellular carcinoma and liver tissues. Cancer Lett 2000; 159: 87–94.

24. Tang W, Miki K, Kokudo N, et al. Des-gamma-carboxyprothrombin in cancer and non-cancer liver tissue of patients with hepatocellular carcinoma. Int J Oncol 2003; 22: 969–75.

25. Xiang CH, Zhang W, Inagaki Y, et al. Measurement of serum and tissue des-gamma-carboxyprothrombin in resectable hepatocellular carcinoma. Anticancer Res 2008; 28: 2219–24.

26. Okuda H, Nakanishi T, Takatsu K, et al. Measurement of serum levels of des-gamma-carboxy prothrombin in patients with hepatocellular carcinoma by a revised enzyme immunoassay kit with increased sensitivity. Cancer 1999; 85: 812–8.

27. Marrero JA, Feng Z, Wang Y, Nguyen MH, Befeler AS, Roberts LR, et al. α-fetoprotein, des-gamma carboxyprothrombin, and lectin-bound α-fetoprotein in early hepatocellular carcinoma. Gastroenterol 2009; 137: 110–8.

28. Lok AS, Sterling RK, Everhart JE, Wright EC, Hoefs JC, Di Bisceglie AM, et al. Des-γ-carboxy prothrombin and α-fetoprotein as biomarkers for the early detection of hepatocellular carcinoma. Gastroenterology 2010; 138: 493–502.

29. Izumi N. Diagnostic and treatment algorithm of the Japanese Society of Hepatology: a consensus-based practice guideline. Oncology 2010; 78S1: 78–86.

30. Shimada M, Yonemura Y, Ijichi H, et al. Living donor liver transplantation for hepatocellular carcinoma: a special reference to a preoperative des-gamma-carboxyprothrombin value. Transplant Proc 2005; 37: 1177–9.

31. Todo S, Furukawa H, Tada M. Japanese Liver Transplantation Study Group Extending indication: role of living donor liver transplantation for hepatocellular carcinoma. Liver Transpl 2007; 13: S48–54.

32. Taketomi A, Sanefuji K, Soejima Y, et al. Impact of desgamma-carboxy prothrombin and tumor size on the recurrence of hepatocellular carcinoma after living donor liver transplantation. Transplantation 2009; 87: 531–7.

33. Masuda T, Beppu T, Horino K, Komori H, Hayashi H, Okabe H, et al. Preoperative tumor marker doubling time is a useful predictor of recurrence and prognosis after hepatic resection of hepatocellular carcinoma. J Surg Oncology 2010; 102: 490–6.

34. Toyoda H, Kumada T, Kaneoka Y, Osaki Y, Kimura T, Arimoto A, Oka H, et al. Prognostic value of pretreatment levels of tumor markers for hepatocellular carcinoma on survival after curative treatment of patients with HCC. J Hepatol 2008; 49: 223–32.

35. Kobayashi M, Ikeda K, Kawamura Y, Yatsuji H, Hosaka T, Sezaki H, et al. High serum des-gamma-carboxyprothrombin level predicts poor prognosis after radiofrequency ablation of hepatocellular carcinoma. Cancer 2009; 115: 571–80.

36. Suzuki M, Shiraha H, Fujikawa T, Takaoka N, Ueda N, Nakanishi Y, et al. Des-gamma-carboxy prothrombin is a potential autologous growth factor for hepatocellular carcinoma. J Biol Chem 2005; 280: 6409–15.

37. Fujikawa T, Shiraha H, Ueda N, Takaoka N, Nakanishi Y, Matsuo N, et al. Des-gamma-carboxy prothrombin-promoted vascular endothelial cell proliferation and migration. J Biol Chem 2007; 282: 8741–8.

38. Liebman HA, Furie BC, Tong MJ, Blanchard RA, Lo KJ, Lee SD, et al. Des-gamma-carboxy (abnormal) prothrombin as a serum marker of primary hepatocellular carcinoma. N Engl J Med 1984; 310: 1427–31.

39. Okuda H, Nakanishi T, Takatsu K, Saito A, Hayashi N, Yamamoto M, et al. Comparison of clinicopathological features of patients with hepatocellular carcinoma seropositive for alpha-fetoprotein alone and those seropositive for des-gamma-carboxy prothrombin alone. J Gastroenterol Hepatol 2001; 16: 1290–6.

40. Fujiyma S, Tanaka M, Maeda S, et al. Tumor markers in early diagnosis, follow-up and management of patients with hepatocellular carcinoma. Oncology 2002; 62: 57–63.

41. Toyosaka A, Okamoto E, Mitsunobu M, Oriyama T, Nakao N, Miura K. Intrahepatic metastases in hepatocellular carcinoma: evidence for spread via the portal vein as an efferent vessel. Am J Gastroenterol 1996; 91: 1610–5.

42. Miyaaki H, Nakashima O, Kurogi M, Eguchi K, Kojiro M. Lens culinaris agglutinin-reactive alpha-fetoprotein and protein induced by vitamin K absence II are potential indicators of a poor prognosis: a histopathological study of surgically resected hepatocellular carcinoma. J Gastroenterol 2007; 42: 962–8.

43. Khan KN, Yatsuhashi H, Yamasaki K, et al. Prospective analysis of risk factors for early intrahepatic recurrence of hepatocellular carcinoma following ethanol injection. J Hepatol 2000; 32: 269–78.

44. Sterling RK, Jeffers L, Gordon F, Venook AP, Reddy KR, Satomura S, et al. Utility of lens culinaris agglutinin-reactive fraction of α-fetoprotein and des-gamma-carboxyprothrombin, alone or in combination, as biomarkers for hepatocellular carcinoma. Clin Gastroenterol Hepatol 2009; 7: 104–13.

45. Carr BI, Kanke F, Wise M, Satomura S. Clinical evaluation of lens culinaris agglutinin-reactive α-fetoprotein and des-gamma-carboxy prothrombin in histologically proven hepatocellular carcinoma in the United States. Dig Dis Sci 2007; 52: 776–82.

46. Kairori M, Ishizaki M, Matsui K,Kwon AH. Predictors of microvascular invasion before hepatectomy for hepatocellular carcinoma. J Surg Oncology 2010; 102: 462–8.

47. Kanazami N, Takeda S, Inoue S, Ohshima K, Sugimoto H, Kaneko T, et al. PIVKA-II during perioperative period in patients with hepato-biliary-pancreatic diseases. Hepato-Gastroenterology 2000; 47: 1695–9.

48. Ma M, Qu XJ, Mu GY, et al. Vitamin K2 inhibits the growth of hepatocellular carcinoma via decrease of desgamma-carboxy prothrombin. Chemotherapy 2009; 55: 28–35.

49. Wang SB, Cheng YN, Cui SX, et al. Des-gamma-carboxy prothrombin stimulates human vascular endothelial cell growth and migration. Clin Exp Metastasis 2009; 26: 469–77.

28.16 Human chorionic gonadotropin (hCG)

Lothar Thomas

hCG is a heterodimer glycoprotein hormone of multiple forms, composed of a non covalently bound α- and β-subunit. hCG is produced by placental trophoblast cells as well as by non trophoblastic tissue such as the pituitary gland and neoplastic cells. In addition, hCG is glycosylated in various tissues to a varying extent, resulting in a wide spectrum of hyperglycosylated to hypoglycosylated variants.

Intact hCG, hyperglycosylated hCG and the free β subunit are the forms of hCG produced by healthy tissues and tumors (Fig. 28.16-1 – Structure and nomenclature for hCG/1/.

Intact hCG

Intact hCG is the predominant form in serum and plasma in pregnancy and trophoblastic disease.

Hyperglycosylated hCG

Hyperglycosylated hCG is primarily formed in choriocarcinoma and testicular cancer. The determination of hyperglycosylated hCG is of low significance in routine clinical diagnostics and is determined to a varying extent depending on the commercial test used for intact hCG determination.

Free β subunit

Secretion of free β subunit is primarily associated with non gestational malignancy (germ cell tumors).

Proteases present in macrophages associated with tumors or present in serum degrade intact hCG and hyperglycosylated hCG (Fig. 28.16-1 – Structure and nomenclature for hCG), resulting initially in metabolites such as nicked (cleaved) hCG and nicked hyperglycosylated hCG (hCGn). Nicked hCG rapidly dissociates into free nicked β subunit (hCGβn) and free α subunit. The C-terminal peptide is cleaved from the free nicked β subunit, and further degradation results in the hCGβ core fragment (hCGβcf). All of these variants may be present in serum and urine, and an ideal hCG tumor marker assay would detect all of these forms. Tumors that produce only free β subunits of hCG (hCGβ) can be most effectively detected by assays that specifically measure this form.

Tab. 28.16-1 – Diseases and conditions predominantly associated with specific hCG forms shows the individual diseases and conditions predominantly associated with specific hCG forms /2/.

28.16.1 Indication

The tumor markers intact hCG and hCGβ are used for diagnosis, follow-up, and monitoring of therapy.

Absolute indication

Germ cell tumors:

  • Hydatidiform mole and choriocarcinoma in women
  • Testicular cancer in men
  • Extra gonadal germ cell tumors.

Relative indication

Patients with increased risk for germ cell tumor:

  • Cryptorchidism
  • Healthy, monozygotic twin of a patient with testicular cancer
  • Patients in complete remission after therapy for testicular cancer; due to increased risk of development of a contralateral secondary tumor
  • Non trophoblastic solid tumors
  • Detection in ascitic fluid and pleural exudate in primary tumor of unknown origin.

Indication during pregnancy (see Section 38.3 – Human chorionic gonadotropin)

  • Early diagnosis of pregnancy
  • Diagnosis of spontaneous abortion
  • Prenatal diagnosis of chromosomal aneuploidies (e.g. trisomy 21).

28.16.2 Method of determination

Three different types of immunoassays are generally available for the determination of hCG:

  • Only the intact dimeric hCG molecule, a heterodimer composed of the α and β subunits (hCG intact)
  • Only the free β subunit (hCGβ)
  • Both intact hCG plus hCGβ (total hCG, also called β-hCG). These immunoassays in addition detect nicked hCG, nicked free hCGβ and β core fragment to a varying extent. For instance, the cross reactivity of a frequently used commercial β-hCG immunoassay is 96% for nicked hCG, 120% for hCGβ, 92% for nicked hCGβ and 35% for the β core fragment /3/.

Quantitative assay in serum

Immunoassays based on the principle of competitive or immunometric methodology. Commercially available assays are used to determine /4/:

  • Total hCG: these assays use a monoclonal antibody directed against α-subunit epitopes of the hCG molecule (e.g., antibody 2119) as capture antibody. A labeled antibody directed against β-subunit epitopes, such as antibody 4001, is used as a tracer. Intact hCG, nicked hCG, hyperglycosylated hCG and the free β- subunit are measured.
  • Free β-subunit (hCGβ): in this assay, capture and tracer antibodies are only directed against epitopes of the β-chain.

For the use of hCG as a tumor marker, the test system must recognize both intact hCG and hCGβ because certain subtypes of testicular cancer, especially seminomas and rarely choriocarcinomas, may only secrete free β-chain but not intact hCG. Hence, either a β-hCG assay must be used or alternatively, if the assay only detects intact hCG, two separate assays may be used for the specific detection of hCG and hCGβ. In many cases, β-hCG assays will detect hCGβ but not always with satisfactory detection limit.

28.16.3 Specimen

28.16.4 Reference interval

See Tab. 28.16-2 – Reference interval for hCG.

28.16.5 Clinical significance

If pregnancy has been ruled out, levels of β-hCG or hCGβ assays above the cutoff value suggest, with high certainty, the presence of a malignant tumor. This includes non trophoblastic tumors, which may also be capable of secreting intact hCG and hCGβ although to a smaller extent trophoblastic tumors. The prevalence of increased hCG values in malignant diseases is shown in Tab. 28.16-3 – Prevalence of increased β-hCG and hCGβ in malignant disease /7/. Testicular germ cell tumors

Testicular cancers are the most common malignant tumors in men between 20 and 40 years of age. Incidence is 4–8 new cases in 100,000 men per year. Clinical symptoms of testicular cancer may include circumscribed hardness or major swelling and minor pain or dull ache in the groin. About 95% of all testicular cancers are germ cell tumors which mostly contain several different histological types (i.e., seminomas, non seminomatous tumors and mixed tumors). Mixed tumors consist of seminoma components and non seminomatous components and are summarized under non seminomatous cancers due to their biological behavior. They are also referred to as non seminomatous testicular germ cell cancer (NSGCT). The prevalence of seminoma and non seminomatous cancer is approximately the same (Tab. 28.16-4 – Prevalence of testicular tumor subtypes/9/. Non seminomatous cancers also include trophoblastic neoplasms such as choriocarcinoma, placental trophoblastic carcinoma and epithelioid trophoblastic carcinoma. Extra gonadal germ cell tumors

Extra gonadal germ cell tumors account for 2–6% of the germ cell tumors and are located in the mediastinum, lung, sacroiliac region, retroperitoneum and pituitary gland. In an analysis, 65% were found to be in the mediastinum /10/. Markers of germ cell tumors

Seminomas and non seminomatous cancers express specific markers that can be determined in serum. The most important markers are AFP, hCG and the enzyme LD. Pure seminomas do not express AFP. In non seminomatous germ cell tumors, about 50% of cases have elevated serum levels of hCG and 60% of AFP, while either marker is elevated in 90% of cases /6/.

The markers are significant for differential diagnosis (Tab. 28.16-5 – Histological classification of germ cell tumors):

  • hCG is formed in seminomas (20–30%) and non seminomatous cancers (choriocarcinoma more than 90%). Elevated concentrations can also be found in mixed tumors containing choriocarcinomatous components and syncytiotrophoblastic giant cells /12/. Trophoblastically differentiated teratomas (WHO: choriocarcinoma +/– teratoma or other non seminomatous cancers) invariably produce hCG, whereas differentiated teratomas (WHO: dermoid cyst, teratoma) and yolk sac tumors never synthesize hCG. Non trophoblastic tumors primarily secrete hCGβ.
  • AFP is not expressed by pure seminomas, but by 90–95% of yolk sac tumors, 20% of teratomas and 10% of embryonic cancers.
  • LD; the enzyme is elevated in half of patients with seminoma and non seminomatous cancers.

In rapidly growing tumors, doubling of hCG values may occur within a few days. After complete removal of an hCG secreting tumor, serum concentration declines with a half life of 1–3 days.

Guidelines for the use of tumor markers in testicular cancer have been published by:

The pattern of hCG in testicular cancers is described in Tab. 28.16-7 – Behavior of serum markers in testicular cancer.


The criteria regarding the prognostic staging of testicular cancers according to the International Germ Cell Consensus Group are described in Tab. 28.16.8 – Diagnosis and treatment germ cell cancer: classification of prognostic groups.


Post surgery: persistent or continuous elevation of hCG and/or AFP during the postoperative course following orchiectomy suggests that the tumor was not limited to the testis and following retroperitoneal lymphadenectomy not restricted to the surgical area.

Chemotherapy: during chemotherapy, transient increases in serum hCG concentrations are common and may occur when tumor masses undergo necrosis and release tumor marker. Placental trophoblastic tumors

Histological differentiation is between complete moles, partial moles in the absence of an embryo, invasive moles and chorioepitheliomas. Complete moles are found in 1 in 2000 pregnancies and invasive moles in 1 in 200,000 pregnancies. The majority of complete moles mostly become clinically apparent as delayed abortions during the second trimester whereas partial moles may manifest during the first trimester in the form of spontaneous abortions. The risk of malignant transformation of a hydatidiform mole is estimated to be 3–15% /22/. Approximately 40–50% of chorioepitheliomas originate from hydatidiform moles.

The FIGO staging and classification for trophoblastic tumors is shown in Tab. 28.16-9 – FIGO staging and classification for trophoblastic tumors. hCG synthesis in trophoblastic tumors

Almost all trophoblastic tumors synthesize hCG /24/. While patients with delayed or spontaneous abortions or with ectopic pregnancies have significantly lower serum hCG values, hCG concentrations are elevated in trophoblastic tumors. However, discrimination is difficult and only the lack of a decline in hCG values after delayed or spontaneous abortions will let suspect an underlying hydatidiform mole or trophoblastic tumor /25/. The best use is assessment of the hCG level in combination with ultrasound examination /8/.

The mean hCGβ to intact hCG ratio is highest for choriocarcinoma /26/, and complete hydatidiform moles have higher hCG concentrations (up to 1 million IU/L) than partial moles. In hydatidiform moles, hCG values decline to normal within 12 weeks; half life is prolonged to approximately 4 days (e.g., due to residual tissue within the myometrium, the uterine blood vessels and the lungs). Plateau or increasing hCG levels indicate proliferating tissue or underlying malignant transformation; this development can occur weeks to months prior to the onset of clinical manifestations /27/.

Postoperatively, hCG must be determined weekly until 3 weeks after normalization and then monthly for up to 6 months. After this period of time, approximately 80% of patients are free of disease. In about 20% of patients, among whom 16% have local residual tumor and 4% have metastasis, chemotherapy becomes necessary. Transient increases in hCG may be frequently observed during chemotherapy /28/. In the case of complete remission, hCG monitoring should be performed at 3–6 month intervals for a period of 5 years.

28.16.6 Comments and problems

hCG standardization

The various hCG molecules have been internationally standardized for more than 25 years /29/:

  • hCG [3rd WHO International Standard (IS) 75/537 and 4th WHO IS 75/589]
  • hCGβ [1st International Reference Preparation (IRP) 75/537]
  • hCGα [1st IRP 75/569].

Assay results are specified in biological units (IU/L).

The new WHO reference reagents /1/ include 6 ampouled preparations of intact hCG, hCGn, hCGα, hCGβ, hCGβn, and hCGβcf. Results are specified in concentration units (nmol/L). The conversion of IU to pmol and μg is shown in Tab. 28.16-10 – hCG conversion. Twelve commercially available β-hCG assays recognizing intact hCG and hCGβ were tested with the WHO reference reagents. Recognition of hCGβ varied markedly (CV 37%) /30/. Most assays overestimated hCGβ.

hCG determination in germ cell tumors

In the presence of germ cell tumors, hCG should not only be determined with an assay that only recognizes intact hCG or total hCG (β-hCG), but an assay with a high detection limit to hCGβ should also be used. For the β-hCG assay as the only test used (measured in IU/L), the upper reference interval value is 5.05 IU/L (17 pmol/L). However, if a cancer primarily expresses hCGβ, as is the case, for example, in early stages of seminoma, many cancers will not be detected based on this high cutoff. Theoretically, according to a study /17/ 42% of marker-positive seminomas and 8% of non seminomatous testicular cancers would have been missed with an β-hCG assay.

Because hCG is used as a tumor marker, all hCG assays should have equimolar recognition of intact hCG and ß-hCG as recommended by the American Society of Clinical Oncology /36/.

Some immunoassays only recognize intact hCG and hence should never be used as tumor marker /37/.

Reference interval

hCG, in a similar manner to LH, increases during menopause:

High dose hook effect

In immunometric assays, high dose hook effect may result in falsely low hCG levels. Manufacturers have specified the value above which a high dose hook effect may occur as 400,000 to 2 million IU/L. According to a study /31/, the value can almost double to 800,000 to 3.6 million IU/L.

Renal insufficiency

In some postmenopausal women with renal insufficiency requiring dialysis, serum hCG levels can be elevated up to 10-fold of normal in the absence of an underlying tumor /32/.


β-hCG in serum at 21 °C or 40 °C, recovery rate after 6 days 94 ± 3.1% and 94 ± 8.3%, respectively /33/.

28.16.7 Pathophysiology

hCG, like the glycoprotein hormones LH, FSH and TSH, consists of an α- and a β-chain. The α-subunits of LH, FSH, TSH and hCG are almost identical. The β-chains resemble each other only in part of their sequences. The hCG metabolites shown in Fig. 28.16-1 – Structure and nomenclature for hCG have no biological activity. The β-chains of hCG and LH are 80% homologous in the region of the first 115 amino terminal amino acids. This explains the similar biological properties of the two hormones. The hCG molecule has a very high carbohydrate content accounting for 30% of its molecular weight.

Whereas the α-subunit is encoded by a single gene, six encoding genes exist for the β-subunit. Trophoblast cells of the placenta and tumor tissue in hydatidiform moles, choriocarcinomas and non seminomatous testicular tumors predominantly secrete intact hCG together with a small proportion of hCGβ. Numerous non trophoblastic malignant tumors are capable of secreting hCGβ commonly, due to the ubiquitous distribution of the hCG genes, and rarely release the intact hCG molecule /34/.

Alterations in the carbohydrate moiety of hCG and its metabolites have an impact on hepatic degradation and, as a result, on the half life. The biological activity of hCG also depends on the degree of glycosylation. Desialylated hCG has been shown to exert an antagonistic effect at the human TSH receptor /35/.

It is generally believed that testicular germ cell tumors (TGCT) arise from cells of the germline that are inhibited in maturation /9/. TGCT in adults is thought to be initiated during fetal development caused by alterations in primordial germ cells during their migration to the embryonic genital ridges or after reaching the gonads. This process is presumably caused by a mutation in the KIT gene before further cell division and before the cells reach the gonads. The KIT encodes the tyrosine kinase receptor. A non invasive stage also referred to as intratubular germ cell neoplasia unclassified (IGCNU) or carcinoma in situ precedes progression to TGCT. The IGCNU can develop into seminoma or non seminomatous cancer during puberty.


1. Bristow A, Berger P, Bidart JM, Birken S, Norman R, Stenman UH, et al. Establishment, value assignment, and characterization of new WHO reference reagents for six molecular forms of human chorionic gonadotropin. Clin Chem 2005; 51: 177–82.

2. Gronowski AM. Clinical assays for human chorionic gonadotropin: what should we measure and how? Clin Chem 2009; 55: 1900–4.

3. Du Toi A, Bouwhuis J, Matson M, Musaad S, Davidson JS. Comparison of 2 human chorionic gonadotropin assays as tumor markers. Clin Chem 2010; 56: 1502–6.

4. Ferraro S, Trevisiol C, Gion M, Panthegini M. Human chorionic gonadotropin assays for testicular tumors: closing the gap between clinical and laboratory practice. Clin Chem 2018; 64: 270–8.

5. Alfthan H, Haglund C, Dabek J, Stenman UH. Concentration of human choriongonadotropin, its beta-subunit, and the core fragment of the beta-subunit in serum and urine of men and nonpregnant women. Clin Chem 1992; 38: 1981–7.

6. Stenman UH. Testicular cancer: the perfect paradigm for marker combinations. Scand J Clin Lab Invest 2005; 65: 181–8.

7. Kuida CA, Braunstein GD, Shintaku P, Said JW. Human chorionic gonadotropin expression in lung, breast, and renal carcinomas. Arch Pathol Lab Med 1988; 112: 282–5.

8. Mann K, Hörmann R. Humanes Choriongonadotropin. In: Thomas L, ed. Labor und Diagnose. Frankfurt 2008; TH Books, S. 1328–34.

9. Horwich A, Shipley J, Huddart R. Testicular germ cell cancer. Lancet 2006; 367: 754–65.

10. Moran CA, Schuster S. Primary germ cell tumors of the mediastinum. Analysis of 322 cases with special emphasis on teratomous lesions and a proposal of histopathological classification and clinical staging. Cancer 1997; 80: 681–90.

11. Mostofi FK, Sesterhenn IA, Sobin LH. World Health Organisation International Histological Classification of tumours: histological typing of testis tumours. Second Edition. Berlin; Springer: 1998.

12. Nochomovitz LE, Rosa J. Current concepts of the histogenesis, pathology, and immunochemistry of germ cell tumors of the testis. Pathol Annu 1978; 13: 327–62.

13. International germ cell cancer collaborative group. International germ cell cancer consensus classification: a prognostic factor-based staging system for metastatic germ cell cancers. J Clin Oncol 1997; 15: 594–603.

14. Schmoll HJ, Souchon R, Krege S, Albers P, Beyer J, Kollmannsberger C, et al. European consensus on diagnosis and treatment of germ cell cancer: a report of the European germ cell cancer consensus group (EGCCCG). Ann Oncol 2004; 15: 1377–99.

15. Gilligan TD, Seidenfeld J, Basch EM, Einhorn LH, Fancher T, Smith DC, et al. American society of clinical oncology clinical practice guideline on uses of serum tumor markers in adult males with germ cell tumors. J Clin Oncol 2010; 20: 3338–3404.

16. Saller B, Clara R, Spoettl G, Siddle K, Mann K. Testicular cancer secretes intact human choriogonadotropin (hCG) and its free β-subunit: evidence that hCG (+ hCG-β) assays are the most reliable in diagnosis and follow-up. Clin Chem 1990; 36: 234–9.

17. Lempiäinen A, Stenman UH, Blomquist C, Hotakainen K. Free β-subunit of human chorionic gonadotropin in serum is a diagnostically sensitive marker of seminomatous testicular cancer. Clin Chem 2008; 54: 1840–3.

18. Raiss GG, Andaloussi MMB, Raissouni SS, Mrabti HH, Errihani HH. Spermatocytic seminoma at the national institute of oncology in Morocco. BMC Res Notes 2011; 4: 393.

19. Neumann A, Keller T, Jocham D, Doehn C. Die humane plazentare alkalische Phosphatase (hPLAP) ist der am häufigsten erhöhte Serummarker beim Hodentumor. Aktuel Urol 2011; 42: 311–5.

20. Bartlett NL, Freiha FS, Torti FM. Serum markers in germ cell neoplasms. Haematol Oncol Clin North Am 1991; 5: 1245–60.

21. Gonzalez-Sanchez V, Moreno-Perez O, Sanchez-Pellicer P, Sanchez-Ortiga R, Guerra R, Dot MM, et al. Validation of the human chorionic gonadotropin immunoassay in cerebrospinal fluid for the diagnostic work-up of neurohypophyseal germinomas. Ann Clin Biochem 2011; 48: 433–7.

22. Gille J. Chorionkarzinom. Med Klin 1975; 70: 532–4.

23. Ngan H, Benedet J L, Bender HG, Jones H W, Percorelli S, Montrucolli G. FIGO staging for gestational trophoblastic neoplasia. Int J Obstet Gynecol 2002; 77: 285–7.

24. Skinner MS, Seckinger D. Evaluation of beta-subunit chorionic gonadotropin as an aid in diagnosis of trophoblastic disease. Ann Clin Lab Sci 1979; 9: 347–52.

25. Bagshawe KD. Choriocarcinoma. A model for tumor markers. Acta Oncol 1992; 31: 99–106.

26. Chen F, Goto S, Furuhashi Y, Tomoda Y. Radioimmunoassay of the serum free β-subunit of human chorionic gonadotropin in trophoblastic disease. J Clin Endocrinol Metab 1987; 64: 313–8.

27. Miller DS, Lurain JR. Classification and staging of gestational trophoblastic tumors. Obstet Gynecol Clin North Am 1988; 15: 477–90.

28. Tomoda Y, Asai Y, Arii Y, Kaseki S, Hideo N, Miwa T, Saiki N, Ishizuka N. Criteria of complete remission from trophoblastic neoplasia with the use of human chorionic gonadotropin (hCG) excretion pattern as a parameter. Cancer 1977; 40: 1016–25.

29. Storing PL, Gaines-Das RE, Bangham DR. International reference preparation of human chorionic gonadotrophin for immunoassay: potency estimates in various bioassay and protein binding assay systems; and international reference preparations of the α and β subunits of human chorionic gonadotrophin for immunoassay. J Endocrinol 1980; 84: 295–310.

30. Sturgeon C, Berger P, Bidart JM, Birken S, Burns C, Norman RJ, et al. Differences in recognition of the 1st WHO international reference reagents for hCG-related isoforms by diagnostic immunoassays for human chorion gonadotropin. Clin Chem 2009; 55: 1484–91.

31. Mahdill HA, Jones GRD. High-dose hook effect in six automated human chorionic gonadotropin assays. Ann Clin Biochem 2010; 47: 383–5.

32. Hubinont C, Dountrelepont JM, van Herweghem JL, Gervey C, Schwers J. Comparison of human chorionic gonadotropin and pregnancy-specific β1-glycoprotein in nonpregnant patients undergoing haemodialysis. ­Nephron 1986; 43: 149–50.

33. Kardana A, Cole LA. The stability of hCG and free β in serum samples. Prenat Diagn 1997; 17: 141–7.

34. Hoermann R, Gerbes AL, Spoettl G, Jüngst D, Mann K. Immunoreactive human chorionic gonadotropin and its free β-subunit in serum and ascites of patients with malignant tumors. Cancer Res 1992; 52: 1520–4.

35. Mann K, Hoermann R. Thyroid stimulation by placental factors. J Endocrinol Invest 1993; 16: 378–84.

36. Grenache DG.Current practices when reporting quantitative human chronic gonadotropin test. J Appl Lab Med 2020; 5: 850–7.

37. Gronowski AM. Why is it so hard to report quantitative human chorionic goanadotropin results? JALM 2020; September: 847–9.

28.17 HER-2/neu

Lothar Thomas

The human epidermal growth factor receptor 2 proto oncogene (HER-2, neu, ErbB-2) is a transmembrane glycoprotein receptor that has intracellular tyrosine kinase activity. The receptor is over expressed in 20–25% of invasive primary breast cancers, and in gastric cancer. HER-2 is primarily analyzed in tissue and the corresponding glycoprotein is assayed in tissue.

The ERBB2 gene [v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, neuro/glioblastoma derived oncogene homolog (avian)], generally referred to as HER-2, encodes a protein that can be determined using the assay HER-2/neu. HER-2 is located on chromosome 17q21. The gene product is a transmembrane glycoprotein belonging to the tyrosine kinase receptor family and plays an important role in the regulation of cell growth, differentiation and survival. Tumor cells of certain solid cancers (breast cancer, gastric cancer) express more receptors on the cell surface than normal cells or other tumors. HER-2/neu over expression plays an important role in the treatment and prognosis of breast and gastric cancers. The over expression of HER-2 is primarily analyzed in tissue and the corresponding glycoprotein is assayed in tissue [by immunohistochemistry or amplified by fluorescence in situ hybridization (FISH) analysis] and in serum.

HER-2/neu positivity alone or in association with a positive nodal status, expansive tumor or the presence of metastasis in breast cancer are associated with a higher rate of disease recurrence and mortality.

Determination of HER-2/neu in tumor tissue

Determination is performed /1/:

  • Primarily by immunohistochemistry. The HER-2 glycoprotein in the cell membrane is determined using antibodies. Depending on the degree of staining, 0/1+ identifies a negative status and 3+ a positive status.
  • In an intermediate status (2+), fluorescence in situ hybridization (FISH) measurement of gene amplification in tissue is performed. Over expression or normal status of the gene HER-2 are measured with labeled DNA probes. The gene is considered to be over expressed if the ratio of the number of HER-2 copies per tumor cell in tumor tissue and in chromosome 17 is ≥ 2.

Determination of HER-2/neu in serum

The extracellular domain (ECD) of HER-2/neu is shed physiologically from the cell membrane and released in the circulation. Soluble ECD, also referred to as HER-2/neu-shed antigen, is detectable in serum and, contrary to tissue analysis which is a one-off event, can be measured any time in the course of the disease. Although only the ECD is measured, the diagnostic designation of the parameter is HER-2/neu.

As in breast cancer, the HER-2 receptor is also over expressed in gastric cancer but, contrary to breast cancer, is only focally expressed. The serum HER-2/neu concentrations are lower in metastatic gastric cancer than in breast cancer. The determination of serum HER-2/neu has not been found to play a role in gastric cancer to date.

This section only deals with serum HER-2/neu.

28.17.1 Indication

HER-2/neu in serum should be determined in patients with metastatic breast cancer.

28.17.2 Method of determination


Principle: two step sandwich immunoassay using the direct chemiluminescence method. This method uses the acridinium ester labeled mouse antibody TA-1 and the fluorescein labeled monoclonal mouse antibody NB-3. Both antibodies react with the corresponding epitopes on the extracellular domain of HER-2/neu. The anti-fluorescein capture antibodies bind to paramagnetic particles. There is a proportionate relationship between the amount of HER-2/neu in the sample and the relative light units measured.

28.17.3 Specimen

Serum: 1 mL

28.17.4 Reference interval

Refer to Tab. 28.17-1 – HER-2/neu in healthy controls.

28.17.5 Clinical significance

The glycoprotein HER-2/neu is neither organ-specific nor tumor-specific. Contrary to tumor markers such as CEA, however, pronounced serum elevations indicate invasive and metastatic breast cancer. HER-2/neu in healthy individuals

HER-2/neu can be detected physiologically in both genders. The concentration rises slightly with increasing age (Tab. 28.17-1); postmenopausal women (median 12.4 μg/L) have slightly higher values than premenopausal women (median 10.9 μg/L) /2/. HER-2/neu in non metastatic cancer

HER-2/neu levels in most patients with colorectal, gastric, liver, pancreatic, ovarian, bladder, kidney and lung cancers are comparable to those in healthy individuals. In breast cancer and gastrointestinal cancers, however, the 95th percentile is increased up to 30 μg/L /3/. HER-2/neu in patients with breast cancer

In the treatment of breast cancer, it is important to determine the serum HER-2/neu level besides CA 15-3 and CEA. In patients with HER-2/neu levels above 30 μg/L /2/:

  • With negative tissue HER-2/neu status, tissue status monitoring should be performed and distant metastasis should be looked for with sensitive imaging techniques
  • With positive tissue HER-2/neu status, distant metastasis should be looked for with sensitive imaging techniques.

Three to four weeks after completion of primary treatment (post-surgery or following adjuvant chemotherapy), HER-2/neu is determined, together with CA 15-3 and CEA:

  • As a relevant baseline parameter for follow-up. Elevated concentrations of CA 15-3 and CEA in the course of follow-up indicate metastasis and/or, more rarely, secondary cancer. The HER-2/neu status in the metastatic tissue should be determined if the presence of metastasis is confirmed by imaging, the HER-2/neu status in the primary tumor was negative and the serum HER-2/neu concentration increased by more than 50% compared to baseline after primary treatment.
  • For the indication of treatment with trastuzumab in a negative HER-2/neu primary tumor. If the evaluation of the metastatic tissue shows a HER-2/neu positive status, treatment with trastuzumab is possible. The serum concentrations of CA 15-3, CEA and HER-2/neu should be determined before start of therapy and then measured every three weeks for therapy monitoring.

In the presence of a primary tumor of unknown origin, HER-2/neu concentrations ≥ 50 μg/L with high probability point to breast cancer. Primary tumor

In breast cancer, HER-2/neu, like CA 15-3 and CEA, correlates with tumor size showing peak values at pT 3/4 , but does not correlate with the nodal status /24/. There is no correlation between histological tumor grading and HER-2/neu concentration, nor with regard to the hormone receptor status. HER-2/neu concentrations correlate, however, with the levels of CA 15-3 and CEA.

The immunohistochemically determined status of HER-2/neu is positive in 20–25% of patients with invasive breast cancer and in 30–70% of patients with metastatic cancer /56/. There is a positive correlation between the tissue HER-2/neu status and the serum HER-2/neu concentration, especially in immunohistochemistry score 3+. Considering the immunohistochemical HER-2/neu status, 50% of the HER-2/neu 3+ tumors and 42% of the HER-2/neu 0–2+ tumors released HER-2/neu > 40 μg/L /3/. HER-2/neu levels > 15 μg/L correlate to HER-2/neu positive tissue in one third of patients, but also reveals false positive findings in 15% of cases. Almost all patients with serum levels above 30 μg/L have a positive HER-2/neu status in tumor tissue /7/ and concentrations above 40 μg/L point to metastatic breast cancer /23/. The levels of patients with distant metastasis do not differ from those of healthy individuals.

Recurrence free survival

HER-2/neu concentrations in recurrence free breast cancer patients are comparable with those in women with no active tumor /2/.

However, the tissue HER-2/neu status influences recurrence free survival after 3 years. According to a study /2/, 3-year recurrence free survival was 87.1–96% in patients with an immunohistochemical score of 0, 1+ and 2+ and only 71.2% in patients with a score of 3+.

Serum HER-2/neu level is also of prognostic significance. For instance, the recurrence free survival after 3 years was 88.1% in patients with a concentration below 15 μg/L and only 71.4% in those with higher values.

Follow-up and therapy monitoring

Endocrine therapy is the recommended systemic therapy for hormone receptor positive HER-2/neu negative metastatic breast cancer. Patients with a tissue negative HER-2/neu status can develop elevated serum HER-2/neu concentrations. According to a study /8/, this was found in 29 of 69 patients with a tissue negative HER-2/neu status of the primary tumor. Patients with an initially positive status had mean concentrations of 225 μg/L (peak 14,000 μg/L), while those with a negative tissue status of the primary tumor showed a mean value of 27.3 μg/L (peak 200 μg/L).

Trastuzumab based therapy

Increasing HER-2/neu concentrations are associated with progressive metastatic disease and poor response to chemotherapy /9/. Moreover, elevated serum HER-2/neu concentrations indicate poor response to hormone therapy /10/.

Trastuzumab is a humanized monoclonal antibody that targets HER-2/neu extracellular domain. Single treatment with trastuzumab achieves a response of 12–30%, depending on the HER-2/neu status. Combined treatment with trastuzumab and other chemotherapeutic agents such as paclitaxel, docetaxel, vinorelbin and platinic salts leads to a synergistic effect, resulting in improved response rates, reduced progression of the disease and prolonged survival. According to a study /11/, decrease in serum HER-2/neu concentration is a predictor for response, duration of response and time to progression (Tab. 28.17-2 – Levels of HER-2/neu and clinical outcomes after trastuzumab therapy).

28.17.6 Comments and problems

It is assumed that inflammation, infection and liver disease, especially in the presence of liver metastasis, can be associated with mild elevations of HER-2/neu.

28.17.7 Pathophysiology

The human epidermal growth factor receptor 2 proto oncogene (HER-2/neu, Erb-2) encodes a transmembrane receptor which is involved in cellular signal transduction based on its intracellular tyrosine kinase activity. The receptor HER-2/neu mediates signals from growth factors for cell proliferation, differentiation and survival.

The receptor is a 185 kDa glycoprotein that contains an extracellular ligand binding domain (ECD), a transmembrane segment and intracellular tyrosine kinase activity. Binding of a growth factor to the ECD results in hetero dimerization of two receptors, enhanced signaling and phosphorylation of the intracellular domain.

The tumor cells in 20–30% of breast cancer patients show increased content of the extracellular domain Erb-2, causing enhanced expression of HER-2 receptors on the cell membrane and, thus, increased shedding. The cleavage of this domain is thought to enhance the phosphorylation of intracellular tyrosine kinase and, thus, signal transduction to the cell /12/. The biological effects of ERBB2 over expression include increased DNA synthesis, accelerated cell division rate and elevated metastatic potential.

Vascular endothelial growth factor (VEGF) is one of the most potent endothelial mitogens and plays an important role in angiogenesis. VEGF synthesis is stimulated by hypoxia and transforming growth factors, as well as by the inactivation of tumor suppressor genes such as ras or src. The over expression of HER-2/neu in breast cancer cells is associated with increased expression of VEGF at RNA level and the corresponding protein.

Exposure of breast cancer cells to HER-2/neu antibodies or trastuzumab reduces the increased formation of VEGF, especially in cells that over express HER-2/neu. Thus, it is concluded that VEGF synthesis depends on HER-2/neu and can be inhibited by trastuzumab. This supports the use of combination therapies for treatment of breast cancers that over express HER-2/neu /13/.


1. Xiao Y, Gao X, Maragh S, Telford WG, Tona A. Cell lines as candidate reference materials for quality control of ERBB2 amplification and expression assays in breast cancer. Clin Chem 2009; 1307–15.

2. Dresse M. Das HER-2/neu-Shed Antigen beim Mammakarzinom zum Zeitpunkt der Primärdiagnose. Dissertation, Institut für Klinische Chemie der Luwig-Maximilians-Universität. München 2009.

3. Stieber P, Roth HJ, Stemmler J, Schmidt H, Untch M, Liedl B, et al. The pattern of HER-2/neu release in benign and malignant diseases. J Clin Oncol 2005; 23, suppl June 1: 618 (abstract).

4. Jensen BV, Johansen JS, Price PA. High levels of serum HER-2/neu and YKL-40 independently reflect aggressiveness of metastatic breast cancer. Clin Cancer Res 2003; 9: 4423–34.

5. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WI. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987; 235: 177–82.

6. Carney WP, Neumann R, Lipton A, Leitzel K, Ali S, Price CP. Potential clinical utility of serum HER-2/neu oncoprotein concentrations in patients with breast cancer. Clin Chem 2003; 49: 1579–98.

7. Dresse M, Mayr D, Heinemann V, Kahlert S, Bauernfeind I, Nagel D, Seidel D, Stieber P. HER-2/neu in tissue and serum at time of primary diagnosis. J Clin Oncol 2009; 27, No15S: e11500 (abstact).

8. Sorensen PD, Jacobsen EH, Lagkjer ST, Bokmand S, Ostergaard B, Olsen DA, et al. Serum HER-2 concentrations for monitoring women with breast cancer in a routine oncology setting. Clin Chem Lab Med 2009; 47: 1117–23.

9. Colomer R, Montero S, Lluch A, et al. Circulating HER-2 extracellular domain and resistance to chemotherapy in advanced breast cancer. Clin Cancer Res 2000; 6: 2653–62.

10. Fedele P, Orlando L, Schiavone P, Calvani N, Caliolo C, Quarant A, et al. Recent advances in the treatment of hormone receptor positive HER2 negative metastatic cancer. Crit Rev Oncol Hematol 2015; 94: 291–301.

11. Ali SM, Carney WP, Esteva FJ, Fornier M, Harris L, Köstler WJ, et al. Serum HER-2/neu and relative resistance to trastuzumab-based therapy in patients with metastatic breast cancer. Cancer 2008; 113: 1294–301.

12. Molina MA, Codoni-Sevrat J, Albanell J, Rojo F, Arrbas J, Baselga J, et al. Trastuzumab (Herceptin), a humanized anti-Her-2 receptor monoclonal antibody inhibits basal and activated Her-2 ectodomain cleavage in breast cancer cells. Cancer Res 2001; 61: 4744–9.

13. Konecny GE, Meng G, Untch M, Wang HJ, Bauerfeind I, Epstein M, et al. Association between HER-2/neu and vascular endothelial growth factor predicts clinical outcome in primary breast cancer patients. Clin Cancer Res 2004; 10: 1706–16.

28.18 Neuron specific enolase (NSE, γ-enolase)

Rolf Lamerz

NSE is useful for monitoring the outcome of treatment and disease course in patients with neuroendocrine tumors, in particular small cell lung cancer and neuroblastoma.

It is not suited as a screening test or adjunct to primary diagnosis because of low diagnostic sensitivity and specificity.

28.18.1 Indication

Monitoring the outcome of treatment and disease course in patients with neuroendocrine tumors and APUDomas.


Small cell lung cancer (SCLC), neuroblastoma


Medullary thyroid carcinoma.

28.18.2 Method of determination

Commercially available are:

  • A competitive double antibody test. NSE contained in the sample and labeled NSE (tracer) compete for the solid phase bound antibody (in many cases polyclonal anti-NSE). After incubation with a second, enzyme labeled antibody, immune complexes are produced which are measured. The detection limit is 2 μg/L
  • An immunometric assay in which a bead, coated with monoclonal antibodies against NSE, is incubated together with the sample and a labeled monoclonal antibody against NSE. The amount of NSE bound to the bead and labeled by the antibody is measured. The detection limit is 0.5 μg/L.

28.18.3 Specimen

Serum, cerebrospinal fluid, pleural exudate: 1 mL

28.18.4 Reference interval

Refer to Tab. 28.18-1 – Reference intervals for NSE.

28.18.5 Clinical significance

NSE can be elevated in benign and malign disease (Tab. 28.18-2 – Diagnostic sensitivity of NSE in benign and malignant diseases). Benign disease

Elevated serum NSE levels are found in patients with:

  • Benign pulmonary diseases (5% > 12 μg/L) /37/
  • Cerebral disease, in particular in the cerebrospinal fluid of patients with cerebrovascular meningitis, disseminated encephalitis, spinocerebellar degeneration, cerebral ischemia and infarction, intracerebral hematomas, subarachnoid hemorrhage, head injury, inflammatory cerebral disease, organic convulsive disorder, Guillain-Barré syndrome, schizophrenia and Creutzfeldt-Jakob disease /56/.

In a prospective cohort study /8/ on NSE differentiation in 59 patients with persistent coma following cardiac arrest and 118 cases of resuscitation from cardiac arrest, a cutoff of 80 μg/L led to a diagnostic sensitivity of 63%, specificity of 100%, positive predictive value of 100%, negative predictive value of 84% and negative likelihood ratio of 37%. NSE was recommended as an important adjunct in the context of other patient characteristics and neurologic examination findings.

The differential diagnosis of patients with seizure and syncope compared with normal controls showed a significantly increased serum NSE only in the seizure group (14.97 ± 7.57 μg/L) given a cutoff of 11.5 μg/L but normal concentrations for the other two groups /9/.

A study on salivary and serum NSE as indicators for neuronal damage involved 50 patients with ischemic stroke and 75 gender and age matched risk group patients (hypertension, type 2 diabetes and ischemic heart disease) as well as 25 gender and age matched healthy controls. Salivary and serum NSE levels were significantly higher in stroke patients than in the other two groups. The cutoff for salivary NSE of 3.7 μg/L was optimum, showing 80% accuracy /10/.

Using the same test, elevated levels above 12.5 μg/l were found in 11–14% of patients with nonmalignant diseases (among these cases, levels were above 25 μg/L in 2%) as well as in uremia and in 50% of pregnancies with fetal neural tube defects /1/. Lung cancer

Prevalence of elevated NSE values

In malignant pulmonary diseases, the diagnostic sensitivity is /12, 3, 7, 1112/:

  • 7–25% (4% above 25 μg/L) in non small cell lung cancer (NSCLC)
  • 30–38% in large cell lung cancer (LCLC)(9% above 25 μg/L)
  • 18–30% in adenocarcinoma (2% above 25 μg/L)
  • 13–30% in squamous cell cancer (3% above 25 μg/L).

Depending on the upper reference interval value, in SCLC elevated NSE levels were found in 60–81% (above 11 μg/L) /311/ and/or in 69–77% (above 12.5 μg/L) /12/. Among these cases, the concentration dependent prevalences in patients with limited disease and with extensive disease were 39–67% (above 12.5 μg/L) and 86–88%, respectively /713/.

Using the same commercially available test, elevated levels > 12.5 μg/L were found in 73% of patients with SCLC (42% above 25 μg/L), 35% of these cases have limited and 65% have extensive disease /3/. If the cutoff value was raised to 25 μg/L, only a maximum of 10% of values will be above this level in the presence of other diseases (e.g., APUDoma and LCLC). A cutoff of 25 μg/L therefore allows better differentiation of lung cancer from nonmalignant pulmonary lesions and SCLC from NSCLC /1/. Comparison of NSE with other tumor markers in lung cancer

Comparative determinations of NSE (cutoff above 11.0–12.5 μg/L) and CEA (cutoff above 5.5–10 μg/L) in benign and malignant lung diseases indicated in SCLC a diagnostic sensitivity of 60–93% for NSE in comparison to 29–69% for CEA. The diagnostic specificity for benign diseases was comparable (NSE 91–95%, CEA 82–93%) but significantly higher for NSE in NSCLC (NSE 58–93%, CEA 25–68%) /14/. According to a multicenter study /15/, NSE was the leading tumor marker in SCLC; diagnostic sensitivities: NSE 77%, Cyfra 21-1 36%, squamous cell carcinoma antigen (SCCA) 32%, CEA 28%.

Besides NSE as a marker of SCLC (cutoff 17 μg/L) with a diagnostic sensitivity of 75% and a specificity of 93% for differentiation from benign lung diseases, the following markers are available for diagnosing SCLC:

  • Chromogranin A (CgA); cutoff value 65 μg/L, the diagnostic sensitivity was 66%, with a specificity of 62%
  • ProGRP; cutoff value of 53 ng/L, the diagnostic sensitivity was 80%, with a specificity of 96% /16/.
  • ProGRP was found to be an additive marker to NSEut was more specific to SCLC than NSE.

A comparison between NSE (cutoff 7.5 μg/L) and ProGRP (cutoff 49 ng/L) in SCLC versus NSCLC and benign lung diseases, the values for the diagnostic sensitivity and specificity and the positive and negative predictive values for ProGRP were as follows /17/:

  • In SCLC versus NSCLC: 64.9%, 96.5%, 93.7%, 77.4%, respectively
  • In SCLC versus benign lung diseases: 64.9%, 93.2%, 91.4%, 79.6%, respectively.

In general, most values for ProGRP are superior to those for NSE. However, the diagnostic sensitivity of NSE (20.3% vs. 77.8%) for differentiating limited and extended disease was higher than that of ProGRP (56.5% vs. 77.8%) /17/.

According to a prospective study on the tumor markers ProGRP, CEA, SCCA, CA 125, CYFRA 21-1 and NSE involving 155 patients with unconfirmed suspicion of lung cancer and in 647 patients with lung cancer (182 squamous cell cancer, 205 adenocarcinoma, 19 LCLC, 175 SCLC and 66 NSCLC patients), elevated concentrations were found in 5.3% of the benign diseases, excluding CA 125 (21.3%) /18/. Tumor markers were related to histological type and tumor extension with significantly higher serum CEA and CA 125 in adenocarcinomas, SCCA and Cyfra 21-1 in squamous tumors and ProGRP and NSE in SCLC. Patients with SCCA levels above 2 μg/L were always NSCLC, while those with SCCA below 2 μg/L and ProGRP above 100 ng/L and NSE above 35 μg/L were all SCLC patients. In the differentiation of NSCLC and SCLC, the diagnostic sensitivity was 76.7% and 79,5%, specificity was 97.2% and 99.6%, with a positive predictive value of 98.6% and 98.6% and a negative predictive value of 60.7% and 92.9%, respectively /18/.

The diagnostic sensitivity and specificity of NSE in benign and malignant diseases are shown in Tab. 28.18-2 – Diagnostic sensitivity of NSE in benign and malignant diseases. NSE and spread of the disease

NSE does not correlate with the localization of metastasis and/or with brain metastasis /1115/, although it correlates well with the clinical stage (i.e., extent of the disease) /1115/. Pattern of NSE during chemotherapy

During the course of chemotherapy, temporary NSE increases occur, in the case of responses, 24–72 hours after the first therapy cycle as a result of tumor cytolysis /1114/. Subsequently, elevated pretreatment serum levels decrease rapidly within a week or by the end of the first treatment cycle /71112/. In contrast, failure of therapy is associated with persistently elevated or intermittently declining NSE levels which do not return to normal.

Normal levels are found in 80–96% of responsive cases whereas tumor recurrence is characterized by increasing NSE levels /711, 1214/. In some of these cases, the rise occurs with a lead time of 1–4 months, often showing an exponential increase, a doubling time of 10–94 days and correlation with the survival period.

In general clinical practice, NSE is useful as the sole prognostic factor and activity marker for monitoring the disease course and treatment of SCLC patients, being clearly superior to the enzyme LD; diagnostic sensitivity 93%, positive predictive value 92% /141519/. NSE as a prognostic and predictive marker

Besides staging, pretreatment NSE levels and treatment induced minimum NSE levels in SCLC patients under platinum based chemotherapy were independent prognostic predictors of time to progression and overall survival /20/. In SCLC, pretreatment NSE (cutoff 7.5 μg/L) was clearly superior to the enzyme LD and ProGRP (threshold 49 ng/L) as a prognostic factor /17/.

In NSCLC, pretreatment Cyfra 21-1 (cutoff 3.6 μg/L) has a high prognostic value. However, an elevated pretreatment NSE concentration (cutoff 12.5 μg/L) is a significant predictor of poor outcome probably by reflecting tumor heterogeneity and underestimated neuroendocrine differentiation in SCLC /21/. Neuroblastoma

Serum NSE levels above 30 μg/L were found in 62% of children with neuroblastoma /23/. The median levels rose, depending on stage, from 13 μg/L (stage I), 23 μg/L (II), 40 μg/L (III) to 214 μg/L (IV) and 40 μg/L (IVs) /4/.

The elevated levels were lower in children with Wilms tumors (20% above 30 μg/L), with a stage dependent elevation of median levels from 16.6 μg/L (stage I), 18 μg/L (II), 29 μg/L (III) to 47 μg/l (IV). Approximately 64% of patients had levels above 25 μg/L. Better differentiation from neuroblastoma was not possible until the concentration reaches ≥ 100 μg/L, although only 50% of neuroblastoma patients had concentrations above this level /2223/.

In neuroblastomas, significant correlation exists between the level of NSE elevation, the prevalence of abnormal NSE values and the stage, as well as an inverse correlation with disease free survival /4/.

In an analysis of 196 neuroblastoma patients with the tumor markers vanillylmandelic acid (VMA) and homovanillic acid (HVA), NSE and the enzyme LD, the clinical sensitivity at diagnosis was 75% for VMA/HVA, 90% for NSE and 81% for LD. The incidence of abnormal results was lower at relapse or progression (40% for serum VMA/HVA and/or 54% for urinary VMA/HVA, 61% for serum NSE and 48% for serum LD). The diagnostic sensitivities of all markers were higher for metastatic compared with local recurrence. NSE was the best, being able to detect 42% of the localized relapses, 77% of the combined local/metastatic relapses and 69% of the metastatic recurrences /24/. Relapse or progression in neuroblastoma cannot be detected reliably by monitoring tumor markers alone. Therefore, follow-up of neuroblastoma patients must include clinical assessment and imaging studies (ultrasound, CT, MR, MIBG scintigraphy). APUDomas

APUDomas are neuroendocrine tumors such as gastrinoma, VIPoma, insulinoma and carcinoid, originating from APUD (amine precursor uptake and decarboxylation) cells (Section 14.5 – Gastrointestinal neuroendocrine tumors). APUD cells are capable of uptake and decarboxylation of amines and/or their precursors. In a total of 34% of APUDomas, elevated serum NSE (above 12.5 μg/L) was found /325/, of which only 11–15% were seen in medullary thyroid cancer /27/ without any clinical relation to the extent of the tumor, in contrast to calcitonin. Furthermore, elevated serum concentrations were found in 39% of gastrointestinal carcinoid and in 56% of gastrointestinal non carcinoid neuroendocrine tumors /26/. Seminoma

Clinically significant elevations are reported in 68–73% of patients with metastatic seminoma, with mean serum NSE levels of 40.3 μg/L /2728/.

Metastatic non seminomatous germ cell tumors of the testis have abnormal concentrations of NSE in only 15% of cases. There is a good correlation with the disease course. Other tumors

Non pulmonary, malignant diseases can show elevated NSE levels with the following prevalence rates:

  • Cancers of all stages 22% (5% above 25 μg/L), of which 41% are associated with and 11% without distant metastasis (14% above 25 μg/L
  • Lymphomas or leukemias 8% (1% above 25 μg/L), primarily T cell leukemias /29/
  • Primary brain tumors (4% above 25 μg/L) /3/.

Brain tumors such as gliomas, meningiomas, neurofibromas and neurinomas are only occasionally associated with elevated serum values. In cerebrospinal fluid, elevated NSE may occur in conjunction with primary brain tumors or brain metastasis /5/, malignant melanoma /30/ and pheochromocytoma /31/.

Elevated NSE concentrations have been observed in 14% of organ limited and in 46% of metastatic renal carcinomas; these concentrations correlate with the tumor grading as an independent prognostic factor /32/.

Elevated NSE levels are also reported in breast cancer (29% above 10 μg/L) /2/ in epithelial (38%) and, more rarely, in non epithelial tumors (5%) /3/.

28.18.6 Comments and problems

Method of determination

NSE test kits from different manufacturers do not give comparable results. Therefore, the same test should always be used for monitoring and specified in the result report.

Reference interval

In serum from healthy individuals, NSE tests of various manufacturers yielded upper reference levels of 6–17 μg/L /23, 4, 7, 26, 2832/.

For cerebrospinal fluid (CSF), reference intervals of 0–6.8 μg/L /6/ and/or 10.8 ± 4.5 μg/L /5/, independent of gender or age, have been reported /6/. The CSF/plasma ratio is 1.04 ± 0.8, as there is no blood-CSF barrier for NSE /56/.

Interference factors

Hemolysis may result in higher values due to the release of large amounts of NSE from red blood cells /3334/. The same also applies to inadequately centrifuged plasma because NSE may be released from platelets /33/. Lipemic or icteric sera do not interfere with NSE assays.


Storage of serum and plasma samples at 2–8 °C for up to 72 h, prolonged storage requires temperatures below –18 °C.

28.18.7 Pathophysiology

Enolase (EC is one of the 11 enzymes of glycolysis and converts 2-phosphoglycerate into phosphoenolpyruvate /3536/. The enzyme, in the form of a dimer, is composed of 2 out of 3 possible non species specific subunits (α, β, γ, subunit molecular weight of 39 kDa), having different immunological, biochemical, and organ specific properties.

There are five possible combinations (αα, ββ, γγ, αγ, αβ) which are each synthesized by different cells:

  • The γ subunit by nerve cells and neuroendocrine cells (APUD cells) in the intestine, lung and endocrine organs such as thyroid gland, pancreas, pituitary gland
  • The αα enolase (also known as non neuronal enolase) by glial cells and other cells ubiquitously present within the body
  • The ββ enolase by muscle cells; heart αβ, striated muscle ββ /37/.

In the brain, γ-enolase occurs in a homologous and heterologous hybrid dimeric form /35/. Because of its occurrence mostly outside neurons and neuroectodermal tissues, for example in malignant tumors, NSE is better referred to as γ-enolase (usually representing the γγ and αγ dimeric forms).

By using various polyclonal and monoclonal antisera with different specificity or different immunohistochemical fixation methods, NSE has also been detected in non neuronal and non neuroectodermal tissues /2/.

In an international workshop, an evaluation and comparative analysis was performed on 12 monoclonal antibodies to NSE by four working groups. The results have been published in Lit. /38/.


1. Hiesche K. Informations-Broschüre Pharmacia (1987).

2. Gerbitz KD, Summer J, Schumacher I, Arnold H, Kraft A, Mross K. Enolase isoenzymes as tumour markers. J Clin Chem Clin Biochem 1986; 24: 1009–16.

3. Fischbach W, Jany B, Nelkenstock R. Bedeutung der neuronenspezifischen Enolase (NSE) in der Diagnostik von Bronchialkarzinomen und neuroendokrinen Tumoren. Dtsch Med Wschr 1986; 111: 1721–5.

4. Zeltzer PM, Marangos PJ, Evans AE, Schneider SL. Serum neuron-specific enolase in children with neuroblastoma. Relationship to stage and disease course. Cancer 1986; 57: 1230–4.

5. Jacobi C, Reiber H. Clinical relevance of increased neuron-specific enolase concentration in cerebrospinal fluid. Clin Chim Acta 1988; 176: 49–54.

6. Rodriguez-Nunez A, Cid E, Eiris J, et al. Neuron-specific enolase levels in cerebrospinal fluid of neurological healthy children. Brain and Development 1999; 21: 16–9.

7. Burghuber OC, Worofka B, Schernthaner G, Vetter N, Neumann M, Dudcak R, Kuzmits R. Serum neuron-specific enolase as a useful tumor marker for small cell lung cancer. Cancer 1990; 65: 1386–90.

8. Almarez AC, Bobrow BJ, Wingerchuk DM, Wellik KE, Demaerschalk BM. Serum neuron specific enolase to predict neurological outcome after cardiopulmonary resuscitation. A critically appraised topic. Neurologist 2009; 15: 44–8.

9. Kee SY, Choi YC, Kim JH, Kim WJ. Serum neuron-specific enolase level as a biomarker in differential diagnosis of seizure and syncope. J Neurol 2010; 257: 1708–12.

10. Al-Rawi NH, Atiyah KM. Salivary neuron-specific enolase: an indicator for neuronal damage in patients with ischemic stroke and stroke-prone patients. Clin Chem Lab Med 2009; 47: 1519–24.

11. Fischbach W, Schwarz-Wallrauch C, Jany B. Neuron-specific enolase and thymidine kinase as an aid to the diagnosis and treatment monitoring of small cell lung cancer. Cancer 1989; 63: 1143–9.

12. Gasser RW, Herold M, Müller-Holzber E, Müller LC, Salzer GM, Huber H. Neuronenspezifische Enolase als Tumormarker beim kleinzelligen Bronchuskarzinom. Dtsch Med Wschr 1988; 113: 1708–13.

13. Jorgensen LGM, Hansen HH, Cooper EH. Neuron specific enolase, carcinoembryonic antigen and lactate dehydrogenase as indicators of disease activity in small cell lung cancer. Eur J Cancer Clin Oncol 1989; 25: 123–8.

14. Fischbach W, Jany B. Neuron-specific enolase in the diagnosis and therapy monitoring of lung cancer: a comparison with CEA, TPA, ferritin and calcitonin. Int J Biol Markers 1986; 1: 129–36.

15. Ebert W, Dienemann H, Fateh-Moghadam A, Scheulen M, Konietzko N, Schleich T, Bombardieri E. Cytokeratin 19 fragment CYFRA 21-1 compared with carcinoembryonic antigen, squamous cell carcinoma antigen and neuron-specific enolase in lung cancer. Eur J Clin Biochem 1994; 32: 189–99.

16. Lamy PJ, Grenier J, Kramar A, Pujol JL. Pro-gastrin-releasing peptide, neuron specific enolase and chromogranin A as serum markers of small cell lung cancer. Lung Cancer 2000; 29: 197–203.

17. Shibayama T, Ueoka H, Nishii K, Kiura K, Tabata M, Miyatake K, Kitajima T, Harada M. Complementary roles of pro-gastrin-releasing peptide (ProGRP) and neuron specific enolase (NSE) in diagnosis and prognosis of small-cell lung cancer (SCLC). Lung Cancer 2001; 32: 61–9.

18. Molina R, Auge JM, Bosch X, Escudero JM, Vinolas N, Marrades R, et al. Usefulness of serum tumor markers, including progastrin-releasing peptide, in patients with lung cancer: correlation with histology. Tumor Biol 2009; 30: 121–9.

19. Jorgensen LGM, Osterlind K, Hansen HH, Cooper EH. Serum neuron-specific enolase (S-NSE) in progressive small-cell lung cancer (SCLC). Br J Cancer 1994; 70: 759–61.

20. Bonner JA, Sloan JA, Rowland JW, Mailliard JA, Wie­senfeld M, Krook JE, et al. Significance of neuron-specific enolase levels before and during therapy for small cell lung cancer. Clin Cancer Res 2000; 6: 597–601.

21. Pujol JL, Boher JM, Grenier J, Quantin X. Cyfra 21-1, neuron specific enolase and prognosis of non-small cell lung cancer – prospective study in 621 patients. Lung Cancer 2001; 31: 221–31.

22. Ferraro S, Braga F, Lusch R, Terenziani M, Carruso S, Panthegini M. Measurement of serum neuron-specific enolase in neuroblastoma: is there a clinical role? Clin Chem 2020; 66 (5): 667-75.

23. Cooper EH, Pritchard J, Bailey CC, Ninane J. Serum neuron-specific enolase in children’s cancer. Br J Cancer 1987; 56: 65–7.

24. Simon T, Hero B, Hunneman DH, Berthold F. Tumour markers are poor predictors for relapse or progression in neuroblastoma. Eur J Cancer 2003; 39: 1899–903.

25. Pacini F, Elisei R, Anelli S, Gasperini L, Schipani E, Pinchera A. Circulating neuron-specific enolase in medullary thyroid cancer. Int J Biol Markers 1986; 1: 85–8.

26. Cunningham RT, Johnston CF, Irvine GB, Buchanan KD. Serum neuron-specific enolase levels in patients with neuro-endocrine and carcinoid tumours. Clin Chim Acta 1992; 212: 123–31.

27. Kuzmits R, Schernthaner G, Krisch K. Serum neuron-specific enolase: a marker for response to therapy in seminoma. Cancer 1987; 60: 1017–21.

28. Fossa SD, Klepp O, Paus E. Neuron-specific enolase – a serum tumour marker in seminoma? Br J Cancer 1992; 65: 297–9.

29. Fujiwara H, Arima N, Ohtsubo H, Matsumoto T, Kukita T, Kawada H, et al. Clinical significance of serum neuron-specific enolase in patients with adult T-cell leukemia. Am J Hematol 2002; 71: 80–4.

30. Wibe E, Hannisdal E, Paus E, Aamdal S. Neuron-specific enolase as prognostic factor in metastatic malignant melanoma. Eur J Cancer 1992; 28A: 1692–5.

31. Oishi S, Sato T. Elevated serum neuron-specific enolase in patients with malignant pheochromocytoma. Cancer 1988; 61: 1167–70.

32. Rasmussen T, Grankvist K, Ljungberg B. Serum γ-enolase and prognosis of patients with renal cell carcinoma. Cancer 1993; 72: 1324–8.

33. Kato K, Asai R, Shimizu A, Suzuki F, Ariyashi Y. Immunoassay of three enolase isoenzymes in human serum and in blood cells. Clin Chim Acta 1983; 127: 353–63.

34. Pählman S, Esscher T, Bergvall P, Odelstad L. Purification and characterization of human neuron-specific enolase: radio-immunoassay development. Tumour Biol 1984; 5: 127–39.

35. Gerbitz KD. Wertigkeit der Enolase-Isoenzyme als Tumormarker. Eine Übersicht. Tumor Diagnostik & Therapie 1989; 10: 45–53.

36. Kaiser E, Kuzmits R, Pregant P, Burghuber O, Worofka W. Clinical biochemistry of neuron-specific enolase. Clin Chim Acta 1989; 183: 13–32.

37. Marangos PJ, Zis AP, Clark RL, Goddwin FK. Neuronal, non-neuronal and hybrid forms of enolase in brain: structural, immunological and functional comparisons. Brain Res 1978; 150: 117–33.

38. Paus E, Hirzel K, Lidqvist M, Höyhtyä M, Warrren DJ. TD-12 workshop report: characterization of monoclonal antibodies to neuron-specific enolase. Tumor Biol 2011, 32: 819–29.

28.19 Prostate-specific antigen (PSA)

Axel Semjonow, Lothar Thomas

Prostate cancer screening is one of the most controversial topics in public health policy, especially in the USA /1/. Although PSA (enzyme nomenclature testing, is widely performed in clinical practice, population based early prostate cancer screening solely based on this marker remains controversial. Used in combination with digital rectal examination and patients history, PSA is an important impact on the indication for prostate biopsy /2/.

Total PSA (tPSA) includes free PSA (fPSA) and complexed PSA (Fig. 28.19-1 – Total PSA consists of complex bound and free PSA).

FPSA includes the sub forms benign prostatic hyperplasia PSA (BPSA), inactive PSA (iPSA) and proenzyme PSA (proPSA). BPSA and iPSA are associated with benign tissue, while proPSA is considered a promising serum marker of prostate carcinoma. Four different proPSA isoenzymes exist in serum e.g., [-2 proPSA], [-4 proPSA], [-5 proPSA] and [-7 proPSA]. [-2 proPSA] was found to be more accurate than tPSA and fPSA in differentiating prostate carcinoma from benign prostatic hyperplasia /34/.

28.19.1 Indication

Total PSA (tPSA)

  • Screening of asymptomatic men with unsuspicious digital rectal examination for early prostate cancer after counselling them on the potential risks and benefits
  • In the presence of prostate related symptoms
  • Monitoring patients with low risk prostate cancer undergoing active surveillance
  • In patients with prostatitis for monitoring the response to treatment with antibiotics.

Free PSA/tPSA ratio

In men with unsuspicious digital rectal examination for early prostate cancer and tPSA levels between 2–10 μg/L to stratify the risk of prostate cancer.


Intended to reduce the number of unnecessary prostate biopsies in PSA tested men.

Complexed PSA (cPSA)

Indications similar to tPSA.

28.19.2 Method of determination

Refer to Ref. /56/.

Total PSA

Competitive immunoassay, sandwich immunoassay, immuno enzymometric assay. Calibration: WHO 96/670 reference preparation (First International Standard). The PSA 90 : 10 reference preparation contains 500 μg/L tPSA after reconstitution in an fPSA/tPSA ratio of 0.10.

Free PSA

Immunoassay (e.g. sandwich immunoassay) using a monoclonal antibody highly specific to free PSA and a second monoclonal antibody recognizing free and bound PSA equally.

Complexed PSA

The complexed PSA method is similar to that of tPSA, with an additional preincubation step with a specific, monoclonal anti-free PSA that blocks free PSA, and thereby impedes its detection.


Immuno enzymometric assay for the determination of an isotype of free PSA, using a monoclonal antibody.

28.19.3 Specimen

Serum, plasma (depending on the test kit; observe manufacturer’s specifications), in exceptional cases cerebrospinal fluid, pleural exudate or ascitic fluid: 1 mL

28.19.4 Reference interval

Refer to Tab. 28.19-1 – Reference intervals for PSA and derived forms.

28.19.5 Clinical significance

Elevated tPSA levels occur in benign prostatic hyperplasia, prostate cancer and acute prostatitis. Compared to prostate cancer, benign prostatic hyperplasia has a higher prevalence in men.

Population based screenings using tPSA are discussed controversially:

  • Investigations of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO) of the US Preventive Services Task Force (USPSTF) reported no benefit of prostate cancer screening using tPSA /7/.
  • The European Randomized Study of screening for Prostate Cancer (ERSPC) noted a significant reduction in mortality /8/
  • There is inconsistent definition of optimal screening intervals and tPSA cutoffs as important parameters for PSA screening for prostate cancer /9/
  • The controversy about PSA screening becomes apparent by the fact that a relevant number of men die with prostate cancer than of it /10/. However this fact does not take into account that metastasized prostate cancer does not lead to death within months. In the majority, patients with metastatic prostate cancer suffer several years before dying of the disease /11/.

Digital rectal examination alone has a poor predictive value for prostate cancer and therefore is not sufficient for the early detection.

Men wishing to undergo PSA screening should be informed of the pros and cons before their PSA is determined /12/. Prostate cancer screening

The German S3 guideline for Prostate Carcinoma does comment PSA screening as followed /2/:

  • Men aged 45 years and older with presumed life expectancy of more than 10 years should be informed about advantages and disadvantages of prostate cancer screening
  • After information the determination of tPSA should be recommended for men who wish prostate cancer screening. In addition digital rectal investigation should be recommended
  • The intervals for follow-up of screening are dependent on the t-PSA concentration: tPSA < 1 μg/L (every 4 years); tPSA 1–2 μg/L (every 2 years); tPSA > 2 μg/L (every year). Men aged > 70 years and tPSA < 1 μg/L are not recommended for early detection of prostate cancer. Screening and prostate cancer mortality

The majority of men with prostate cancer diagnosed by PSA screening have localized disease with no evidence of metastases. However, a certain number of these patients is at high risk for imminent tumor spread /13/. Two large randomized trials tested /9/ whether screening reduces prostate cancer mortality and, while the US trial (PLCO) /7/reported no benefit, the European (ERSPC) trial /8/ noted a significant reduction in mortality.

Prostate cancer screening using tPSA is limited to diagnostic uncertainty and results in a trio of drawbacks (e.g., over detection of cancer, treatment complications, and disease progression). Active surveillance has the potential to reduce side effects of over diagnosis while maintaining the quality of live in low risk prostate cancers. In case of tumor progression, there is still the possibility of curative treatment by surgery or radiation /13/.

Over diagnosis of indolent, nonlethal cancers bears the risk of over treatment associated side effects. Side effects from surgery or radiation are common (i.e., urinary, sexual and gastrointestinal complications) increases medical expenditure. The U.S. Preventive Services Task Force (USPSTF) evidence review estimated that radical prostatectomy was associated with a 20% increase in the risk of urinary incontinence and a 30% increase in erectile dysfunction after 1 to 10 years /14/.

The benefits of screening for prostate cancer at 13 years of follow-up are shown in the ERSPC trial. Men were recommended to undergo prostate biopsy for tPSA of 3.0 μg/L or greater (men with lower PSAs have lower risk features e. g, are white as opposed to African-American, younger, have no family history of prostate cancer, and a normal digital rectal examination /9/; their risk of high-grade cancer is low). Men aged 55 to 69 years were screened at 2-year intervals.

The results of this screening study were as followed /8/:

  • Benefits: PSA screening was associated with a significant absolute reduction of 0,71 prostate cancer death per 1,000 men after an average follow up of 8.8 years. This finding corresponded to a relative reduction of 20% in the rate of death from prostate cancer among men between the ages of 55 and 69 years at study entry.
  • Drawbacks: to prevent one prostate cancer death, 1.410 men would have to be screened, and an additional 48 men would have to be treated. Suspicious PSA result

Men are recommended to undergo prostate biopsy in case of significant tPSA increase (using the same tPSA assay) or if at least one of the following criteria is present /2/:

  • tPSA of ≥ 10 μg/L in asymptomatic men
  • tPSA of 4–10 μg/L and pathologic PSA ratio
  • A suspicious digital rectal examination.

Diagnostics for prostate cancer include:

  • Digital rectal examination (DRE): gold standard for prostate cancer detection in the pre-PSA era, nowadays less than 20% of clinically localized prostate cancers are detected by DRE (Fig. 28.19-2 – Comparison between digital rectal investigation and PSA). If both tPSA level and DRE are pathologic, there is a high evidence of prostate cancer. DRE requires experience and skill and, therefore, depends on the examiner. Even if performed by experienced examiners, the diagnostic sensitivity of DRE is lower than that of elevated PSA. However, if the DRE yields a cancer suspicious finding, prostate biopsy is necessary independently of the tPSA concentration.
  • Ultrasound and/or further imaging procedures. Primary diagnosis of prostate carcinoma

The diagnosis of prostate carcinoma is based on histopathological verification /2/. Staging of prostate carcinoma

The anatomical staging of prostate carcinoma is performed according the UICC TNM system. The classification is an anatomically based system that records the primary and regional nodal extent of the tumor and the absence or presence of metastases.

The Gleason grading system (Gleason score) is the strongest prognostic factor for clinical behavior of prostate cancer and treatment response.

The locally limited prostate cancer is differentiated into the following risk groups concerning a recurrence of cancer /2/:

  • Lo w risk: PSA up to 10 μg/L and Gleason score ≤ 6
  • Medium risk: PSA > 10–20 μg/L or Gleason score ≤ 7
  • High risk: PSA > 20 μg/L or Gleason score ≥ 8.

Most studies conclude that in cases with PSA values above 10–20 μg/L the probability of organ confined prostate cancer is markedly lower /1516/, while in PSA values above 50 μg/L organ confined tumors are the exception.

However, because of a substantial overlap between t-PSA concentrations in the different tumor stages, t-PSA alone has been found to be unreliable for predicting the final pathologic stage in a patient. Nevertheless, by combining various other preoperative measures such as the digital rectal examination finding and the degree of differentiation by several prostate biopsies, some investigators have determined that the predictive power of t-PSA for pathologic tumor stage can be significantly enhanced /17/. T-PSA in therapy of prostate carcinoma

Curative treatment options for patients with prostate cancer are presented in Tab. 28.19-2 – T-PSA in curative standard therapy in patients with prostate carcinoma.

For men with an initial diagnosis of metastatic prostate cancer androgen-deprivation therapy (castration)represents the standard of care. Androgen receptor signaling is altered in castration resistant prostate cancer.

The reasons are /18/:

Prostate cancer is an androgen dependent disease. Even tumors that are resistant to castration remain androgen receptor dependent. Androgen receptor antagonist therapeutics and androgen receptor gene rearrangements result in preferential expression of constitutively splice variants, the most common of which is AR-V7 /18/.T In patients with AR-V7 anti-androgen therapy in inefficient. Androgen deprivation therapy with taxanes showed an overall survival benefit. Taxanes inhibit mitosis and androgen receptor signaling by disrupting nuclear transport of the receptor /18/.The pre therapy determination of AR-V7 protein expression is a treatment specific marker for response and outcomes between androgen receptor signalling inhibitor and taxanes /19/. Principles of PSA monitoring at suspicion of prostate carcinoma

Refer to Tab. 28.19-5 – Principles of PSA monitoring in suspicion of prostate cancer.

28.19.6 Comments and problems

Method of determination

There are numerous commercially available tPSA assays and a test for complexed PSA. In addition, various assays for fPSA as an assay for [-2]pro-PSA are available. The assessing physician needs to know the assay and the corresponding reference interval from the laboratory /20/. Without this information, the PSA loses its diagnostic and prognostic significance, resulting in missed prostate tumors or unnecessary prostate biopsies.

Some commercially available tPSA assays were recalibrated using a standard preparation of the WHO resulting in a better comparability. Refer to

The lack of tPSA specificity between the different commercial assays result from the molecular differences between PSA standards. The molecular form differences cause quantitative discordances in ELISA measurements of tPSA /21/.

Detection limit: there is no generally accepted consensus for the biologically relevant tPSA level in monitoring subsequent to radical prostatectomy /22/. The assay manufacturers determine the detection limit by diluting serum PSA and determine the concentration at which the variation coefficient is higher than 20% or dilution linearity is overruled. The detection limit appears to be of great clinical significance in patients after radical prostatectomy.

High-dose hook effect: falsely low measured values (often within the normal range) can be obtained in sandwich assays and in very high PSA levels. The concentration up to which no high-dose hook effect has been observed is specified in the manufacturer’s package insert.

Ultra sensitive tPSA determination: there is no generally accepted definition of ultra sensitive assay for tPSA determination. While former test methods with a detection limit of 0.2 μg/L were termed hypersensitive, the terms ultra sensitive, supersensitive or hypersensitive are now used for methods with a detection limit < 0.1 μg/L. The results of many commercially available tests match this concentration or are lower.

The clinical relevance of ultra sensitive PSA levels after surgical removal of the prostate has not been clarified /22/. Unnecessary emotional stress of the patient following radical prostatectomy by informing him of findings regarding ultra sensitive PSA levels should be avoided. Such information should only be communicated on the patient’s request and after explaining the analytical accuracy and clinical significance of the measurement /23/.

PSA strip tests: these tests are offered for the semi-quantitative determination of tPSA based on serum or whole blood. None of these tests has met the clinical requirements regarding adequate diagnostic sensitivity and specificity and investigator independent reading accuracy /24/.

Reference interval

The reference interval depends on the commercial assay used and must be defined individually for each assay and specified in the report. The necessity of such information is emphasized in the recommendations of the interdisciplinary S3 guideline of the German Society for Urology /2/, the European Group on Tumour Markers /25/ the US-American National Academy of Clinical Biochemistry /26/ and the World Health Organization /20/. The indiscriminating application of the upper reference interval of ≤ 4.0 μg/L to any assay method leads to unnecessary prostate biopsies or missed prostate cancers.

The upper reference interval value originally defined for the method of Hybritech (Tandem-R) is 4.0 μg/L. It was defined as the concentration below which the tPSA concentration was in 97% of 207 men above 40 years of age with a clinically healthy prostate. The usefulness of this reference interval for the Hybritech method was confirmed by major screenings. However, recent results from population based screenings and other studies show that there is still a relatively high rate of detectable prostate cancer in men with lower PSA concentrations /2728/.

Age-specific reference intervals of tPSA

PSA concentration rises with increasing age. In a prospective study involving 2119 healthy men, it was found that the tPSA level correlates with age and prostatic volume /15/. It was recommended to use age-specific PSA reference intervals for improved diagnostic sensitivity in the diagnosis of cancer in younger men and increasing specificity in older men. The age-specific reference ranges determined by the Tandem-R PSA assay were 0–2.5 μg/L for men up to 50 years and 0–6.5 μg/l for men > 70 years. The results of this study were confirmed by other study groups, however, with deviating reference intervals.

Influencing factors

Biological and analytical variation: the intra individual variation for tPSA is approximately 20%. The analytical variation in immunoassays is below 5% /29/.

Circadian rhythm: absent.

Manipulation of the prostate: manipulations can induce increases in PSA concentration, especially in fPSA and to a smaller extent in tPSA /30/. An increase in tPSA induced by digital rectal examination (DRE) rarely reaches a clinically relevant extent. Just in case, however, the blood sample for tPSA determination should be taken either before a DRE or several days afterwards. Blood collection immediately after DRE leads to falsely high fPSA concentration and, as a result, prostate cancer may be missed. Elevated tPSA can occur for several weeks after acute urinary retention, prostate biopsy or other manipulations of the prostate and should only be checked after 1–2 months.

Hepatic dysfunction: acute (but not chronic) hepatic dysfunctions can presumably induce elevated PSA levels.

Renal dysfunction: renal insufficiency or dialysis do not cause any change in serum tPSA concentration.

Drugs: evidence of drug induced increase in tPSA concentrations is not available. GnRH analogs or anti-androgens lead to pronounced decreases in tPSA, 5α- reductase inhibitors cause a mean decrease in tPSA concentration by approximately 50%, however, with great variability depending on the period of intake. Statins /31/ are also thought to lower tPSA by up to 15–20%.

Human anti-mouse antibodies (HAMA): can interfere with the assay, leading to falsely high or falsely low concentrations. Serum HAMA concentrations can be determined using specific immunoassays.


tPSA and complexed PSA can be stored in whole blood without losses for at least 8 h at room temperature; under the same conditions, the fPSA concentration declines by approximately 1% per hour. For instance, in a whole blood sample containing 7 μg/L of tPSA and 2 μg/L of fPSA, the ratio of fPSA/tPSA will within 12 h decline from 29% to 26%. If a serum sample is stored at 23 °C or 4 °C, the ratio will only decline to 28% within 24 h. Therefore, samples taken for fPSA determination should be stored as serum and analyzed or frozen within 24 h. [-2]pro-PSA increases in whole blood relatively soon after blood collection and should be centrifuged within 3 h /32/.

28.19.7 Pathophysiology

Most prostate cancers originate in the peripheral zone of the prostate and are only rarely found exclusively in the transitional zone. Therefore, trans rectal ultrasound-guided biopsy of the peripheral zone is helpful in diagnosing prostate cancer (Fig. 28.19-5 – Prostate cancers originate from the peripheral zone of the prostate gland).

PSA is synthesized in the epithelial cells along the acini and in the ductal epithelium of the prostate gland and released into the prostatic ductal system (Fig. 28.19-6 – PSA is produced in the epithelial cells of the prostate gland). fPSA has a molecular weight of approximately 33 kDa and is a single-chain glycoprotein.

PSA is a kallikrein like serine protease in human prostatic fluid formally called human kallikrein 3 (hK3). The released inactive precursor form of PSA is activated extracellularly by the trypsin like activity of human kallikrein 2 (hK2) which cleaves a short fragment of the 244 amino acid precursor form of PSA resulting in the development of the mature active PSA molecule with 237 amino acids.

PSA cleaves the gel forming proteins from the seminal vesicles which is thought to induce liquefaction of the ejaculate and increase sperm motility.

Usually, PSA is present in serum at low concentration and only reaches elevated plasma levels in prostate cancer, benign prostatic hyperplasia, acute prostatitis or following prostatic biopsy.

The development of ultra sensitive PSA immunoassays and sensitive immunohistochemical techniques has proven that PSA is not specific to the prostate. For instance, PSA has been detected immunohistochemically in male and female periurethral and perianal glands, in normal endometrium, in the breast and in tumorous tissue of both genders.

PSA has also been found in low concentrations in the serum of women and the cytosol of breast cancer cells where it exists predominantly as free PSA. It has also been found in breast cancer associated with the progesterone receptor (but not with the estrogen receptor) and a favorable prognosis.

It is assumed that all tissues with steroid hormone receptors are capable of synthesizing PSA and that the enzymatically active form of PSA is involved in the growth regulation of tissue /33/.

The functions of PSA are assumed to include the inhibition of cell growth, anti carcinogenesis or anti angiogenesis as well as the induction of apoptosis. These hypotheses are supported by the observation that malignant prostatic tissue produces less PSA than benign tissue. However, these findings have no implication on the interpretation of PSA levels above 0.1 μg/L in clinical routine.


1. Roth JA, Gulati R,Gore JL,Cooperberg MR, Etzioni R. Economic analysis of prostate-specific antigen screening ans selective treatment strategies. JAMA Oncol 2016; 2: 890–9.

2. Interdisziplinäre Leitlinie der Qualität S3 zur Früherkennung, Diagnose und Therapie der verschiedenen Stadien des Prostatakarzinoms. AWMF-Register-Nummer 043/022OL

3. Le BV, Griffin CR, Loeb S, Carvalhal GF, Kan D, Baumann NA, et al. [-2] Proenzyme prostate specific antigen is more accutate than total and free prostate specific antigen in differentiating prostate cancer from benign disease in a prospective prostate cancer screening study. J Urol 2010; 183: 1355–9.

4. Osredkar J, Kumer K, Fabjan T, Hlebic G, Podnar B, Lenard G, Smrkolj T. The performance of [-2] proPSA and prostate health index tumor markers in prostate cancer diagnosis. J Lab Med 2016; 40: 419–24.

5. Bernstein Lh, Rudolph RA, Pinto MM, Viner N Zuckerman H. Medically significant concentrations of prostate-specific antigen in serum assessed. Clin Chem 1990; 36: 515–8.

6. Velonas VM, Woo HH, dos Remedios CG, Assinder SJ. Current status of of biomarkers for prostate cancer. Int J Mol Sci 2013; 14: 11034–60.

7. Andriole GL, Grawford ED, Grubb RL III, et al. Prostate cancer screening in the randomized prostate lung colorectal, and ovarian cancer screening trial: mortality results after 13 years of follow-up. J Natl Cancer Inst 2012; 104: 125–32.

8. Schröder FH, Hugosson J, Roobol MJ, Tammela TLJ, Zappa M, Nelen V ,et al. Screening and prostate cancer mortality: results of the European Randomized Study of Screening for Prostate Cancer (ERSPC) at 13 years of follow-up. Lancet 2014; 384: 2027–35.

9. Thompson IM, Tangen CM. Prostate screening comes of age. Lancet 2014; 384: 2004–6.

10. Schröder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V, et al. ERSPC Investigators. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009; 360: 1320–8.

11. Bill-Axelson A, Holmberg L, Garmo H, Taari K, Busch C, Nordling S, et al. Radical prostateectomy or watchful waiting in prostate cancer – 29-year follow -up. N Engl J Med 2018; 379: 2319–29.

12. Ito K, Schröder FH. Informed consent for prostate-specific antigen-based screening – European view. Urology 2003; 61: 20–2.

13. Pinsky PF, Prorok PC, Kramer BS. Prostate cancer screening– a perspective on the current state of evidence. N Engl J Med 2017; 376: 1285–9.

14. Chou R, Croswell JM, Dana T, Bougatsos C, Blazina I, Fu R, et al. Screening for Prostate Cancer: A Review of the Evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2011; 155: 762–71.

15. Oesterling JE, Jacobsen SJ, Chute CG, Guess HA, Girman CJ, Panser LA, Lieber MM. Serum prostate-specific antigen in a community-based population of healthy men. Establishment of age-specific reference ranges. JAMA 1993; 270: 860–4.

16. Catalona WJ, Partin AW, Slawin KM, Brawer MK, Flanigan RC, Patel A. Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA 1998; 279: 1542–7.

17. Semjonow A, Hertle L. Kann mit Hilfe des prostataspezifischen Antigens das Tumorstadium von Prostatakarzinomen beurteilt werden? Urologe [A] 1995; 34: 290–6.

18. Sartor O, deBono JS. Metastatic prostate cancer. N Engl J Med 2018; 378: 645–57.

19. Scher HI, Lu D, Schreiber NA, Louw J, Graf RP, Vargas HA, et al. Association of AR-V7 on circulating tumor cells as a treatment-specific biomarker with outcomes and survival in castration-resistant prostate cancer. JAMA Oncol 2016; 2: 1441–9.

20. Brawer MK, Benson MC, Bostwick DG, Djavan B, Lilja H, Semjonow A, Su S, Zhou Z. Prostate-specific antigen and other serum markers: current concepts from the World Health Organization Second International Consultation on Prostate Cancer. Semin Urol Oncol 1999; 17: 206–21.

21. McJimpsey WL. Molecular form differences between prostate-specific antigen (PSA) standards create quantitative discordances in PSA ELISA measurements. Scientific Reports 2016; 6: 22050 https://doi.org/10.1038/srep22050.

22. Semjonow A, de Angelis G. “Ultrasensitive” Messverfahren für das prostataspezifische Antigen (PSA): Wie tief wollen wir messen? J Lab Med 2003; 27: 16–9.

23. Semjonow A. Editorial comment on: prognostic implications of an undetectable ultrasensitive prostate-specific antigen level after radical prostatectomy. Eur Urol 2010; 57: 629–30.

24. Oberpenning F, Hetzel S, Weining C, Brandt B, de Angelis G, Heinecke A, et al. Semi-quantitative immunochromatographic test for prostate specific antigen in whole blood: tossing the coin to predict prostate cancer? Eur Urol 2003; 43: 478–4.

25. Semjonow A, Albrecht W, Bialk P, Gerl A, Lamerz R, Schmid HP, van Poppel H. Tumour markers in prostate cancer: EGTM recommendations. European Group on Tumour Markers. Anticancer Res 1999; 19: 2799–801.

26. Sturgeon CM, Duffy MJ, Stenman UH, Lilja H, Brünner N, Chan DW, Babaian R, et al. National Academy of Clinical Biochemistry. National Academy of Clinical Biochemistry laboratory medicine practice guidelines for use of tumor markers in testicular, prostate, colorectal, breast, and ovarian cancers. Clin Chem 2008: 54: e11–79.

27. Catalona WJ, Smith DS, Ornstein DK. Prostate cancer detection in men with serum PSA concentrations of 2.6 to 4.0 ng/mL and benign prostate examination. Enhancement of specificity with free PSA measurements. JAMA 1997; 277: 1452–5.

28. Thompson IM, Pauler DK, Goodman PJ, Tangen CM, Lucia MS, Parnes HL, et al. Prevalence of prostate cancer among men with a prostate-specific antigen level < or = 4.0 ng per milliliter. N Engl J Med 2004; 350: 2239–46. Erratum in: N Engl J Med 2004; 351: 147.

29. Söletormos G, Semjonow A, Sibley PE, Lamerz R, Petersen PH, Albrecht W, et al. Biological variation of total prostate-specific antigen: a survey of published estimates and consequences for clinical practice. Clin Chem 2005; 51: 1342–51.

30. Oberpenning F, Schmid HP, Fuchs-Surdel W, Hertle L, Semjonow A. The impact of intraoperative manipulation of the prostate on total and free prostate-specific antigen. Int J Biol Markers 2002; 17: 154–60.

31. Hamilton RJ, Goldberg KC, Platz EA, Freedland SJ. The influence of statin medications on prostate-specific antigen levels. J Natl Cancer Inst 2008; 100: 1511–8

32. Haese A, Huland E, Graefen M, Hammerer P, Noldus J, Huland H. Ultrasensitive detection of prostate specific antigen in the followup of 422 patients after radical prostatectomy. J Urol 1999; 161: 1206–11.

33. Rittenhouse HG, Finlay JA, Mikolajczyk SD, Partin AW. Human Kallikrein 2 (hK2) and prostate-specific antigen (PSA): two closely related, but distinct, kallikreins in the prostate. Crit Rev Clin Lab Sci 1998; 35: 275–368.

34. Semjonow A, Hamm M, Rathert P. Half-life of prostate-specific antigen after radical prostatectomy: The decisive predictor of curative treatment? Eur Urol 1992; 21: 200–5.

35. Haese A, Huland E, Graefen M, Hammerer P, Noldus J, Huland H. Ultrasensitive detection of prostate specific antigen in the followup of 422 patients after radical prostatectomy. J Urol 1999; 161: 1206–11.

36. Semjonow A. Editorial comment on: prognostic implications of an undetectable ultrasensitive prostate-specific antigen level after radical prostatectomy. Eur Urol 2010; 57: 629–30.

37. Cox JD, Gallagher MJ, Hammond EH, Kaplan RS, Schellhammer PF. Consensus statements on radiation therapy of prostate cancer: guidelines for prostate re-biopsy after radiation and for radiation therapy with rising prostate-specific antigen levels after radical prostatectomy. American Society for Therapeutic Radiology and Oncology Consensus Panel. J Clin Oncol 1999; 17: 1155.

38. Roach M 3rd, Hanks G, Thames H Jr, Schellhammer P, Shipley WU, Sokol GH, Sandler H. Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer: recommendations of the RTOG-ASTRO Phoenix Consensus Conference. Int J Radiat Oncol Biol Phys 2006; 65: 965–74.

39. Hanlon AL, Pinover WH, Horwitz EM, Hanks GE. Patterns and fate of PSA bouncing following 3D-CRT. Int J Radiat Oncol Biol Phys 2001; 50: 845–9.

40. Fowler jr JE, Pandey P, Seaver LE, Feliz TP, Braswell NT. Prostate specific antigen regression and progression after androgen deprivation for localized and metastatic prostate cancer. J Urol 1995; 153: 1860–5.

41. Robinson D, van Ellen EM, WU YM, Schulz N, Lorigo RJ, Mosquera JM, et al. Integrative clinical genomics of advanced prostate cancer. Cell 2015; 161: 1215–28.

42. Ku SY, Rosario S, Wang Y, Mu P, Seshandri M, Goodrich ZW, et al. Rb1 and Trp53 cooperate to suppress prostate cancer lineage plasticity, metastasis, and antiandrogen resistance. Science 2017; 355: 78–83.

43. Pritchard CC, Morrisey C, Kumar A, et al. Complex MSH2 and MSH6 mutations in hypermutated microsatellite unstable advanced prostate cancer. Nat Commun 2014; 5: 4988.

44. Catalona WJ, Richie JP, de Kernion JB, Ahmann FR, Ratliff TL, Dalkin BL. Comparison of prostate specific antigen concentration versus prostate specific antigen density in the early detection of prostate cancer: receiver operating characteristic curves. J Urol 1994; 152: 2031–6.

45. Bangma CH, Kranse R, Blijenberg BG, Schröder FH. The value of screening tests in the detection of prostate cancer. Part II: Retrospective analysis of free/total prostate-specific analysis ratio, age-specific reference ranges, and PSA density. Urology 1995; 46: 779–84.

46. Semjonow A, Hamm M, Rathert P, Hertle L. Prostate-specific antigen corrected for prostate volume improves differentiation of benign prostatic hyperplasia and organ-confined prostatic cancer. Br J Urol 1994; 73: 538–43.

47. Carter HB, Pearson JD, Metter EJ, Brant LJ, Chan DW, Andres R, et al. Longitudinal evaluation of prostate-specific antigen levels in men with and without prostate disease. JAMA 1992; 267: 2215–20.

48. Kadmon D, Weinberg AD, Williams RH, Pavlik VN, Cooper P, Migliore PJ. Pitfalls in interpreting prostate specific antigen velocity. J Urol 1996; 155: 1655–7.

49. Manseck A, Pilarsky C, Froschermaier S, Menschikowski M, Wirth MP. Diagnostic significance of prostate-specific antigen velocity at intermediate PSA serum levels in relation to the standard deviation of different test systems. Urol Int 1998; 60: 25–7.

50. Clinical practice guidelines in oncology: Prostate cancer early detection. V2.2007. www.nccn.org

51. Semjonow A, Brandt B, Oberpenning F, Hertle L. Unterschiedliche Bestimmungsverfahren erschweren die Interpretation des prostataspezifischen Antigens. Urologe A 1995; 34: 303–15.

52. Schmid HP, McNeal JE, Stamey TA. Observations on the doubling time of prostate cancer. The use of serial prostate-specific antigen in patients with untreated disease as a measure of increasing cancer volume. Cancer 1993; 71: 2031–40.

53. Raaijmakers R, Wildhagen MF, Ito K, Pàez A, de Vries SH, Roobol MJ, Schröder FH. Prostate-specific antigen change in the European Randomized Study of Screening for Prostate Cancer, section Rotterdam. Urology 2004; 63: 316–20.

54. Vickers AJ, Wolters T, Savage CJ, Cronin AM, O’Brien MF, Pettersson K, Roobol MJ, Aus G, Scardino PT, Hugosson J, Schröder FH, Lilja H. Prostate-specific antigen velocity for early detection of prostate cancer: result from a large, representative, population-based cohort. Eur Urol 2009; 56: 753–60.

55. Wang W, Wang M, Wang L. Adams TS, Tian Y, Xu J. Diagnostic ability of %p2PSA and prostate health index for aggressive prostate cancer: a meta-analysis. www.nature.com/scientific reports. https://dx.doi.org/10.1038%2Fsrep05012.

56. Mikolajczyk SD, Catalona WJ, Evans CL, Linton HJ, Millar LS, Marker KM, et al. Proenzyme forms of prostate-specific antigen in serum improve the detection of prostate cancer. Clin Chem 2004; 50: 1017–25.

57. Miyakubo M, Ito K, Yamamoto T, Suzuki K, Diagnostic significance of [-2]proPSA, total and transition zone prostate volume and adjusted PSA-related indices in Japanese men with total PSA in the 2.0 to 10.0 ng/ml range. Eur Urol Suppl 2011; 10: 65.

58. Catalona WJ, Parin AW, Sanda MG, Wei JT, Klee GG, Bagma CH, et al. A multicenter study of [-2]pro-prostate specific antigen combined with prostate specific antigen and free prostate specific antigen for prostate cancer detection in the 2.0 to 10.0 ng/ml prostate specific antigen range. J Urol 2011; 185: 1650–5.

59. Filella X, Gimenez N. Evaluation of [-2]proPSA and prostate health index (phi) for the detection of prostate cancer: a systematic review and meta-analysis. Clin Chem Lab Med 2013; 51: 729–39.

60. Loeb S, Catalona WJ. The prostate health index: a new test for the detection of prostate cancer. Ther Adv Urol 2014; 6: 74–7.

61. Tosoian J, Loeb S, Feng Z, Isharwal S, Landis P, Elliot D, et al. Association of [-2]proPSA with biopsy reclassification during active surveillance for prostate cancer. J Urol 2012; 188: 1131–6.

62. Stenman UH, Leinonen J, Alfthan H, Rannikko S, Tuh-kanen K, Alfthan O. A complex between prostate-specific antigen and alpha 1-antichymotrypsin is the major form of prostate-specific antigen in serum of patients with prostatic cancer: assay of the complex improves clinical sensitivity for cancer. Cancer Res 1991; 51: 222–6.

63. Christensson A, Laurell CB, Lilja H. Enzymatic activity of prostate-specific antigen and its reactions with extracellular serine proteinase inhibitors. Eur J Biochem 1990; 194: 755–63.

64. Vashi AR, Wojno KJ, Henricks W, England BA, Vessel-la RL, Lange PH. Determination of the “reflex range” and appropriate cutpoints for percent free prostate-specific antigen in 413 men referred for prostatic evaluation using the AxSYM system. Urology 1997; 49: 19–27.

65. Stephan C, Lein M, Jung K, Schnorr D, Loening SA. The influence of prostate volume on the ratio of free to total prostate specific antigen in serum of patients with prostate carcinoma and benign prostate hyperplasia. Cancer 1997; 79: 104–9.

66. Gann PH, Hennekens CH, Stampfer MJ. A prospective evaluation of plasma prostate-specific antigen for detection of prostatic cancer. JAMA 1995; 273: 289–94.

67. Schröder FH, van der Cruijsen-Koeter I, de Koning HJ, Vis AN, Hoedemaeker RF, Kranse R. Prostate cancer detection at low prostate specific antigen. J Urol 2000; 163: 806–12.

68. Woodrum D, French C, Shamel LB. Stability of free prostate-specific antigen in serum samples under a variety of sample collection and sample storage conditions. Urology 1996; 48(6A Suppl): 33–9.

69. Herrmann W, Stöckle M, Sand-Hill M, Hübner U, Herrmann M, Obeid R, et al. The measurement of complexed prostate-specific antigen has a better performance than total prostate-specific antgen. Clin Chem Lab Med 2004; 42: 1051–7.

70. Okihara K, Cheli CD, Partin AW, Fritche HA, Chan DW, Sokoll LJ, et al. Comparative analysis of complexed prostate specific antigen, free prostate specific antigen and their ratio in detecting prostate cancer. J Urol 2002; 167: 2017–23.

71. Jung K, Brux B, Lein M, Rudolph B, Kristiansen G, Hauptmann S, et al. Molecular forms of prostate-specific antigen in malignant and benign prostatic tissue: biochemical and diagnostic implications. Clin Chem 2000; 46: 47–54.

72. Lein M, Kwiatkowski M, Semjonow A, Luboldt H, Hammerer P, Stephan C, Klevecka V, Taymoorian K, Schnorr D, Recker F, Loening SA, Jung K. A multicenter clinical trial on the use of complexed-prostate specific antigen in low prostate specific antigen concentrations. J Urol 2003; 170: 1175–9.

73. Tanguay S, Begin LR, Elhilali MM, Behlouli H, Karakiewicz PI, Aprikian AG. Comparative evaluation of total PSA, free/total PSA, and complexed PSA in prostate cancer detection. Urology 2002; 59: 261–5.

74. Semjonow A, Oberpenning F, Weining C, Schmid H-P, Lein M, Fobker M, de Angelis G, Brandt B. Unterschiede zwischen PSA-Bestimmungsverfahren. In: Fornara P, Semjonow A (eds). PSA: Der Weg zum Befund – Präanalytik und Analytik des prostataspezifischen Antigens; München; Zuckschwerdt 2002: 83–113.

28.20 S100-β protein (S100B)

Lothar Thomas

S100 protein is a dimeric, low molecular weight protein of approximately 10.5 kDa (monomer). The S100 proteins belong to the multiple gene family of Ca2+-binding proteins. The proteins was given the name S100 due to their high solubility in 100% saturated ammonium sulfate solution. Various combinations of α- and β- subunits are responsible for the heterogeneity of this protein family which consists of the isoforms S100B (ββ), S100A (αα) and S100A1 (αβ). The β isoform of S100 is found in the highest concentration within the astrocytes, oligodendrocytes and peripheral Schwann cells of the central nervous system. Extra neural synthesis takes place in melanocytes, adipocytes and chondrocytes. S100B is actively secreted by these cells and released upon cell damage.

28.20.1 Indication

  • Malignant melanoma: differential diagnosis, prognosis, therapy monitoring
  • Traumatic brain injury, stroke, neurodegenerative diseases.

28.20.2 Method of determination

S100 assays detect both S100A1B (αβ) and S100B (ββ). Many assays contain monoclonal or polyclonal antibodies directed against the S100 β subunit. Determination is usually performed by electrochemiluminescence assay (ECLIA) or immunoluminometric assay (ILMA) /1/.


According to the sandwich principle, S100 in the sample in a first step binds to a biotinylated antibody and a ruthenium-labeled antibody forming an immunocomplex. In a second step, the immunocomplex binds to streptavidin-coated micro particles added to the reaction mixture. The micro particle bound immunocomplex is linked to the solid phase via interaction of biotin and streptavidin, electrochemiluminescent emission is induced and measured by a photomultiplier.


In this immunometric assay, two monoclonal antibodies to S100 bind to paramagnetic particles. A tracer antibody labeled with an isoluminal derivate is used for immunometric measurement.

28.20.3 Specimen

Serum: 1 mL

28.20.4 Reference interval



≤ 0.100 μg/L /2/

Children 3–18 years

≤ 0.16 μg/L /3/

Children < 3 years

≤ 0.20 μg/L

95% of healthy controls have concentrations below the specified reference interval if determination is performed using commercial assays. Variations may occur depending on the method used.

28.20.5 Clinical significance

Healthy individuals have a median serum S100 concentration of 0.041 μg/L; the 95th percentile is 0.096 μg/L and the 100th percentile is 0.144 μg/L. There are no gender or age related differences /4/. S100B plays an essential part in S100 determination and is therefore of diagnostic significance.

In a study /2/, apart from malignant melanoma, only 2% of patients with benign disease and only 1% of patients with malignant disease showed slightly elevated S100 values, in most cases up to 0.5 μg/L (Tab. 28.20-1 – S100 values in healthy individuals and patients/2/. The S100 levels in many benign diseases are within the range of those of healthy individuals /5/. Concentrations of up to 0.7 μg/L have been reported in liver cirrhosis and renal insufficiency.

High S100 concentrations are determined in acute traumatic brain injury, neurodegeneration and intra- and extracranial vascular injury (ischemic and hemorrhagic stroke) /6/:

Besides for brain injury, concentrations of S100 > 0.5 μg/L have high diagnostic sensitivity for malignant melanoma (Tab. 28.20-2).

28.20.6 Comments and problems


Serum should be used unless otherwise specified by the assay manufacturer.

Method of determination

Commercially available assays have different detection limits and cannot be used interchangeably unless their cutoff values are adapted. The within-person variation is 18.9% /26/.


Up to 8 h at 15–25 °C, up to two days at 4–8 °C, up to 3 months at –20 °C.

28.20.7 Pathophysiology

S100 proteins belong to the S100/calmodulin/parvalbumin/troponin C super family with a molecular weight of about 13 kDa. They are thought to be Ca2+ sensor molecules that regulate and modulate biological activities via Ca2+. Some of the S100 proteins bind to Ca2+ and Zn2+. The S100 family comprises at least 25 proteins. Binding of Ca2+ to S100B induces a large conformational change that allows hydrophobic residues to expose and interaction with other proteins in order to confer biological activity. S100B is located in the cytoplasm and nucleus of the astrocytes along with other members of the S100 family, and it regulates the cytoskeletal structure and cell proliferation /7/.

The effects of S100B depend on its concentration. At nanomolar levels, S100B has a neurotrophic activity for neuronal cells during neuronal maturation and glial cell proliferation. At micromolar levels, S100B may have deleterious effects because it stimulates the expression of pro inflammatory cytokines and induces neuronal apoptosis. Moreover, S100B interacts with the receptor for advanced glycation end products (RAGE), causing elevation in reactive oxygen species, cytochrome C release and activation of the caspase cascade. Caspases are cystein proteases with a physiological function in programmed apoptosis.

S100B can interact with the tumor suppressor gene p53, inhibiting p53 phosphorylation by protein kinase C. This induces total inhibition of p53 oligomerization, blocking the involvement of p53 in cell cycle regulation, DNA repair and the induction of apoptosis.

S100B is in high concentrations found in astroglial cells of the central nervous systems and is to a lesser degree produced by Schwann cells of the peripheral nervous system, by chondrocytes, adipocytes and Langerhans cells.

S100A is expressed by malignant melanoma cells where its concentration correlates with invasion depth and tumor size.

The half-life of S100 is approximately 30 min.


1. Hallen M, Carlhed R, Karlsson M, Hallgren T, Bergenheim M. A comparison of two different assays for determining S-100B in serum and urine. Clin Chem Lab Med 2008; 46: 1025–9.

2. Holdenrieder S, Speisberg F, Hatz R, Waidelich R, Untch M, Hofman K, Wehnl B, et al. Pattern of S100-release in benign and malignant diseases beside malignant melanoma. J Lab Med 2013; 37: 21–8.

3. Castellani C, Stojakovic T, Cichocki M, Scharnagl H, Erwa W, Gutmann A, et al. Reference ranges for neuroprotein S-100B: from infants to adolescents. Clin Chem Lab Med 2008; 46: 1296–9.

4. Wiesmann M, Missler U, Gottmann D, Gehring S. Plasma S-100β protein concentration in healthy adults is age- and sex-independent. Clin Chem 1998; 44: 1056–8.

5. Stieber P, Hatz R, Liedl B, Untch M, Heinemann V. Pattern of S100 release in benign and malignant diseases besides malignant melanoma. Anticancer Res 2003; 23: 4511.

6. Rothermundt M, Peters M, Prehn JH, Arolt V. S100B in brain damage and neurodegeneration. Microsc Res Tech 2003; 60: 614–32.

7. Yardan T, Erenler AK, Baydin A, Aydin K, Cokluk C. Usefulness of S100B protein in neurological disorders. J Pak Med Assoc 2011; 63: 276–81.

8. Molina R, Navarro J, Filella X, Castel T, Ballesta AM. S-100 protein serum levels in patients with benign and malignant diseases: false-positive results related to liver and renal function. Tumour Biol 2002; 23: 39–44.

9. Jäckel A, Deichmann M, Waldmann V, Bock M, Näher H. S-100β-Protein im Serum als Tumormarker beim malignen Melanom. Hautarzt 1999; 50: 250–6.

10. Hauschild A, Michaelsen J, Brenner W, Rudolph P, Glaser R, Henze E, Christophers E. Prognostic significance of serum S100B detection compared with routine blood parameters in advanced metastatic melanoma patients. Melanoma Res 1999; 9: 155–61.

11. Mohammed MQ, Abraha HD, Sherwood RA, MacRae K, Retsas S. Serum S100beta protein as a marker of disease activity in patients with malignant melanoma. Med Oncol 2001; 18: 109–20.

12. Seregni E, Massaron S, Martinetti A, Illeni MT, Rovini D, Belli F, et al. S100 protein serum levels in cutaneous malignant melanoma. Oncol Rep 1998; 5: 601–4.

13. Martenson ED, Hansson LO, Nilsson B, von Schoultz E, Mansson Brahme E, Ringborg U, Hansson J. Serum S-100b protein as a prognostic marker in malignant cutaneous melanoma. J Clin Oncol 2001; 19: 824–31.

14. Von Schoultz E, Hansson LO, Djureen E, et al. Prognostic value of serum analyses of S-100 beta protein in malignant melanoma. Melanoma Res 1996; 6: 133–7.

15. Jury CS, McAllister EJ, MacKie RM. Rising levels of serum S100 protein precede other evidence of disease progression in patients with malignant melanoma. Br J Dermatol 2000; 143: 269–74.

16. Guo HB, Stoffel-Wagner B, Bierwirth T, Mezger J, Klingmuller D. Clinical significance of serum S100 in metastatic malignant melanoma. Eur J Cancer 1995; 31: 924–8.

17. Banfalvi T, Gilde K, Gergye M, Boldizsar M, Kremmer T, Otto S. Use of serum 5-S-CD and S-100B protein levels to monitor the clinical course of malignant melanoma. Eur J Cancer 2003; 39: 164–9.

18. Hamberg AP, Korse CM, Bonfrer JM, de Gast GC. Serum S100B is suitable for prediction and monitoring of response to chemoimmunotherapy in metastatic malignant melanoma. Melanoma Res 2003; 13: 45–9.

19. Leidel BA, Bogner V, Zock M, Kanz KG. Das serologische Protein S100B. Unfallchirurg 2011; https://doi.org/10.1007/s00113-010-1946-x.

20. Elting JW, de Jager AE, Teelken AW, Schaaf MJ, Maurits NM, van der Naalt J, et al. Comparison of serum S-100 protein levels following stroke and traumatic brain injury. J Neurol Sci 2000; 181: 104–10.

21. Castellani C, Bimbashi P, Ruttenstock E, Sacherer P, Stojakovic T, Weinberg AM. Neuroprotein S100B – a useful parameter in paediatric patients with mild traumatic brain injury? Acta Paediatr 2009; 98: 1607–12.

22. Saenger AK, Christenson RH. Stroke biomarkers: progress and challenges for diagnosis, prognosis, differentiation, and treatment. Clin Chem 2010; 56: 21–33.

23. Missler U, Wiesmann M, Friedrich C, Kaps M. S-100 protein and neuron-specific enolase concentrations in blood as indicators of infarction volume and prognosis in acute ischemic stroke. Stroke 1997; 28: 1956–60.

24. Johnsson P, Backstrom M, Bergh C, Jonsson H, Luhrs C, Alling C. Increased S100B in blood after cardiac surgery is a powerful predictor of late mortality. Ann Thorac Surg 2003; 75: 162–8.

25. Jonsson H, Johnsson P, Birch-Iensen M, Alling C, Westaby S, Blomquist S. S100B as a predictor of size and outcome of stroke after cardiac surgery. Ann Thorac Surg 2001; 71: 1433–7.

26. Johnson PR, Gwilt SC, Neville CG. Estimates of wthin-person biological variation and reference change values of serum S100B and NSE proteins. Clin Chem 2018; 64: 866–8.

28.21 Squamous cell carcinoma antigen (SCCA)

Rolf Lamerz

The squamous cell carcinoma (SCCA) antigen is a TA-4 sub fraction which has been identified in saline extracts of cervical squamous cell cancer. SCCA levels are elevated in patients with squamous cell cancer, correspond to tumor burden, are helpful in the early detection of tumor recurrence, and in the surveillance of cancer patients.

28.21.1 Indication

Monitoring of disease course and response to therapy in patients with squamous cell carcinoma of cervix, lung, esophagus, anus and head neck region.

28.21.2 Method of determination

Double antibody radioimmunoassay (RIA)

The detection limit of the RIA is 1 μg/L.

Immunometric assay (IMA) /123/

Solid-phase sandwich assay with two monoclonal murine antibodies against different epitopes of SCCA /123/. Detection limit: 0.3 μg/L /2/. The IMA is superior to the RIA due to its better detection limit.

28.21.3 Specimen

Serum, plasma, cerebrospinal fluid, pleural exudate, ascitic fluid: 1 mL

28.21.4 Reference interval


≤ 3.0 μg/L /4567/


≤ 2.0 μg/L /2/

28.21.5 Clinical significance

SCCA can be elevated in benign and malign disease (Tab. 28.21-1 – Diagnostic sensitivity of SCCA in benign and malignant diseases). Benign disease

Elevated levels > 2–3 μg/L are observed in 6–10% in liver cirrhosis /15/, 30–64% in pancreatitis /1/, 78% in patients undergoing hemodialysis and in 44% of those without. In renal insufficiency, SCCA increase correlates with elevation of creatinine level /158/.

In benign pulmonary diseases (chronic bronchitis, COPD, tuberculosis), the prevalence of elevated SCCA is 0–40% /35, 7, 89/. In patients with idiopathic pulmonary fibrosis, SCCA is produced in many metaplastic alveolar epithelial cells and correlates positively with the pro fibrotic expression of TGFβ /10/.

There is a relationship with SCCA and allergic diseases where SCAA was found to be a predictive factor to evaluate the severity, for example, of allergic rhinitis caused by Dermatophagoides farinae /11/.

In patients with benign gynecological disease, elevated levels are found in 3–37% /812/ and specifically in 3–8% of cases with leiomyomatous uterus /1/.

Elevated levels are found in 21% of ear, nose, throat (ENT) diseases; in benign tumors (46%) higher rates are seen than in other diseases (5%) /8/.

Elevated serum SCCA concentrations are found in benign skin disease such as psoriasis (83%), correlating with the extent of the affected skin surface, pemphigus, eczema (80%) and other diseases with an inflammatory component /13/. Investigations also report humoral autoimmune responses directed against the family of SCCA proteins (SCCA, arginase 1, enolase 1 and keratin 10) in psoriasis, a T lymphocyte mediated autoimmune disease /14/. Cervical cancer Incidence of elevated SCCA

The incidence of elevated concentrations is highest in primary squamous cell carcinoma of the cervix with a rate of 45–83% and in recurrent cervical squamous cell carcinoma at a rate of 66–84% /16, 8, 1516/. During remission and in case of no evidence of disease (NED), the rate of elevated levels is 0–33%  with a mean of 7% /12/.

These data apply mostly to cervical squamous cell carcinoma (70–80%) and less to cervical adenosquamous carcinoma (56%) /1/ or cervical adenocarcinoma (0–23%) /16/. Diagnostic specificity is 96% in blood donors, 97% in patients with benign genital lesions, 97% in complete remission and 93% in cervical intra epithelial neoplasia (CIN) I–III /6/. Relationship between SCCA concentration and spread of disease

In cervical squamous cell carcinoma, serum SCCA level correlates with the extent of the disease. For instance, the SCCA positivity rates rise with advancing tumor stage as per FIGO classification from 0–25% (stage 0, CIN I–III) and 9% (stage Ia), 27–60% (Ib), 44–83% (II), 55–84% (III) to 67–100% (IV) /16, 15, 1617/.

Furthermore, serum SCCA levels correlate with the lymph node status and the clinical findings /61015/, but not with the degree of differentiation, age and other laboratory findings /12/. Higher SCCA levels are found in squamous cell cancers (100%) than in non squamous large cell (83%) or small cell (73%) cancers. Relationship between SCCA concentration and course of disease

Changes in serum SCCA concentration correlates well with the disease course /115, 1618/. Following curative surgery or radiation therapy, normalization of serum SCCA concentration occurs within 2–7 days, with a half life of < 24 h /6/.

Significant recurrent increases after return to normal levels or temporary declines of elevated concentrations indicate local or disseminated progressive disease, often with a lead time of 1–14 months (mean 2.2–3.3 months) /61618/.

Serum SCCA determination results in earlier recurrence detection in a small proportion (14%) of patients but does not contribute to better survival. Therefore, it is not recommended to carry out routine SCCA monitoring after primary cancer treatment, not even in patients in early stage SCC (stage IB–IIA) /19/. Prognostic significance of SCCA

Levels above 30 μg/L are associated with rapid tumor recurrence and short survival. Cervical cancer patients with abnormal SCCA levels persisting 2–6 weeks after treatment have the highest tumor recurrence rate (92%) /6/.

According to a study /15/, elevated pretreatment SCCA concentrations, tumor size and vascular invasive disease are independent predictors of the presence of lymph node metastasis. Furthermore, among five different markers compared, only the initial SCCA concentration is related to the survival. There is a threefold increase in the risk of tumor recurrence in node negative cervical cancer patients with elevated SCCA levels above 1.9 μg/L /15/.

According to a different study /20/ involving stage IB to IIB cervical squamous cell carcinoma patients, the combination assay of pretreatment SCCA (cutoff 1.5 μg/L) and CA 125 (cutoff 35 mU/L) concentrations seems to be significant in estimating lymph node status and prognosis. Comparison of SCCA with other tumor markers

In cervical squamous cell carcinoma, CEA is inferior to SCCA regarding diagnostic sensitivity (e.g., SCCA 60–74% CEA 31–34% for primary diagnosis and 70–73%/50–51% for the detection of tumor recurrence). This contrasts with sensitivities of 2.7%/14% for response to therapy and 56%/89% in tumors with distant metastasis /116/. For CA 125, a diagnostic sensitivity of only < 35% has been observed /12/. Other gynecological cancers

Various diagnostic sensitivities have been reported for SCCA: 0–10% in breast cancer /116/, 8–30% in endometrial cancer /616/, 30% in uterine cancer /1/, 4–20% in ovarian cancer /1616/, 19–42% in vulvar cancer /6/ and 17% in vaginal cancer /12/.

An immunohistochemical analysis of SCCA in 1360 tissue micro arrays containing breast tumor tissues showed significantly increased expression of 0.3%, 2.5% and 9.4% in grades I–III and of 2.5%, 3.1%, 8.6% in TNM stages I–III correlated to estrogen receptor/progesterone receptor double negative tumors as well as poor overall survival and recurrence free survival /21/. Lung cancer Prevalence of elevated SCCA concentration

The highest prevalence of SCCA is found in squamous cell lung cancer (39–78%) /23, 5, 7, 89/; the diagnostic sensitivity in lung cancer overall is 27% /1/. In detail, it is 33–61% in NSCLC /89/, 18% in large cell lung cancer, 4–18% in SCLC and 15–42% in adenocarcinoma /379/. Relationship between SCCA concentration and spread of disease

SCCA levels correlate with the spread of the disease showing a stage dependent increase in diagnostic sensitivity from 27–53% (I), 31–72% (II), 60–88% (III) and 71–100% (IV) /237/ or rates of 22% and 73% in limited and extensive disease, respectively /379/. In addition, there is correlation with the T and M status /3/. Relationship between SCCA concentration and course of disease

Within two days following radical tumor resection, serum SCCA concentrations decline to levels within the reference interval; however, in residual tumors, they only decline slightly and indicate tumor recurrence by a renewed increase, often with a lead time of 4–5 months /39/. Combination of SCCA with other tumor markers

According to a European multicenter study /22/ using the tumor markers Cyfra 21-1, CEA, SCCA and NSE in patients with lung cancers of various histologies and stages, patients with benign lung disease and healthy adults, Cyfra 21-1 was found to be the most sensitive single marker. Cyfra 21-1 exhibited a diagnostic specificity of 95% in benign lung disease and a sensitivity of 46% in the group of all lung cancers compared to the diagnostic sensitivities of CEA, SCCA and NSE of 32%, 25% and 28%, respectively. Taking into account the different histological types of tissue, the sensitivities of Cyfra 21-1, CEA and SCCA in SCC were 58%, 23% and 32%, respectively

The diagnostic sensitivities in adenocarcinoma were 42% and 44% for Cyfra 21-1 and CEA, but below 15% for SCCA and NSE.

NSE was the most sensitive marker in SCLC (77%) and has been complemented by ProGRP which has an even higher specificity for this type of cancer. Cancer of the head-neck region

In cancer of the head and neck region, the diagnostic sensitivity of SCCA is 34–78% /25, 8, 923/. On average, the sensitivity increases with the TNM and the clinical stage. The rates found in various tumors are 49% in maxillary sinus lesions, 34% in oral cavity lesions, 23% in lesions of the tongue, 19% in laryngeal lesions and 11–33% in pharyngeal lesions. The diagnostic sensitivity for detecting tumor recurrences is 60–75% /7/.

Given a cutoff of 1.5 μg/L, the diagnostic sensitivity of SCCA increases, depending on the stage (Tis, T1–4), from 50% to 85% /23/. The post treatment SCCA level is the most important prognostic factor for disease free and overall survival. Esophageal cancer

In esophageal cancer, the mean diagnostic sensitivity of SCCA is 30–39% with rates rising in a stage dependent manner from 0–27% (stage I), 20–40% (II), 39–61% (III) to 45–50% (IV) /27, 9, 2425/.

Following successful treatment, a decline in the SCCA concentration to within the reference interval is seen. Persistently increased levels and further increases occur in the case of residual tumor and renewed rises with tumor recurrence /9/. Normal SCCA levels also suggest the presence of localized tumor and a favorable prognosis whereas elevated SCCA levels are more likely to indicate extensive disease and a poor prognosis /24/. Genitourinary tumors

Increased SCCA concentrations are measured in patients with metastatic squamous cell carcinoma of the penis (45%) and the urethra /26/. Other tumors

Elevated SCCA levels are found in colon and pancreatic cancer at a rate of 20%.

Moreover, elevated SCCA levels with a diagnostic sensitivity of 76% and a specificity of 86% have been described in the rare anal canal cancer /4/. Elevated SCCA values were also found in cutaneous SCC /27/. SCCA, benign liver disease and HCC

Mean SCCA concentrations in patients with liver cirrhosis (0.41 μg/L) were within the normal range and were lower than in patients with hepatocellular carcinoma (HCC) and with a single nodule < 3 cm (group 1: 1,6 μg/L) and with larger or multi focal HCC (group 2: 2,2 μg/L). In contrast, immunohistochemical analysis showed inverse linear relation between SCCA and nodule size, being smallest in liver cirrhosis biopsies (264 μm2), strongest in HCC group 1 (1163 μm2) and smaller in HCC group 2 (626 μm2) and no statistically significant correlation between tissue and serum levels of SCCA /28/.

Furthermore, elevated SCCA-IgM immunocomplex values have been described in patients with benign chronic liver disease, with the highest increases determined in HCC patients /2829/. SCCA-IgM complexes were detectable in 63 of 188 (33%) patients with chronic hepatitis but in none of 100 controls /30/. A significant increase in SCCA-IgM levels over time was observed in patients with fibrosis progression (117 ± 200 U/mL/year), but not in those without histologic deterioration (8.8 ± 31 U/mL/year). In conclusion, monitoring SCCA-IgM levels appears to be a useful approach to identify patients with chronic hepatitis at higher risk for cirrhosis development.

A longitudinal evaluation of the SCCA-IgM IC in cirrhotic patients with HCV before, at the end and at 6-month and 12-month follow-up after treatment with PEG-interferon and ribavirin according to treatment outcome. The sustained virological response (SVR) versus non-response (NR) had the following results:

  • No significant difference in baseline serum levels between SVR and NR patients (451.2 AU/mL)
  • A significant decrease in median SCCA-IgM IC serum levels at the end of treatment (186.8 AU/mL) and at both 6-month (96.8 AU/mL) and 12-month follow-up (52.4 AU/mL) in SVR patients
  • No significant SCCA-IgM modifications in the NR patient group /31/.

For SCCA-1 (SERPINB3), a variant (SCCA-PD), presenting a single mutation in the reactive center (Gly351A1a), has been identified (rs3180227). This polymorphism was investigated in 45 patients with chronic hepatitis, 53 cirrhotic patients and 50 HCC patients. The SCCA-PD variant was more frequently found in cirrhotic patients (45.3%) than in patients with chronic hepatitis and healthy controls (24%, each) and intermediate values were found in HCC patients (36%).

Serum SCCA-IgM-IC concentrations were lower in patients carrying SCCA-PD than in wild type patients and the difference was statistically significant in cirrhotic patients (117.45 ± 54.45 U/mL vs. 268.52 ± 341.27 U/mL) /32/.

28.21.6 Comments and problems

Method of determination

Triglycerides ≤ 350 mg/dL (4.0 mmol/L), bilirubin ≤ 16 mg/dL (274 μmol/L) and hemoglobin ≤ 1 g/dL do not interfere with the immunometric assay.

In renal insufficiency, SCCA levels show dependence on the serum creatinine level /158/.

Because SCCA occurs in sweat, saliva and other body fluids, contamination during the pippetting process must be avoided (protective gloves, no mouth pipetting).

Positive or negative interference with the results of the immunometric assay may occur in patients with human anti-murine antibodies.

Reference interval

According to a study /3/ involving blood donors, the 95th percentile of SCCA was 3.3 μg/L in men and 5.0 μg/L in women. Pregnant women do not present elevated levels /1233/. Increased concentrations were found in amniotic fluid, ranging from 26 ± 11 μg/L (14th–16th gestational week), 91 ± 29 (17th–20th gestational week) to 670 ± 390 μg/L (30th–40th gestational week) /33/.


Temporary storage is possible for up to one week at 4 °C; otherwise freezing is necessary at a temperature below –25 °C.

28.21.7 Pathophysiology

In 1977, the first report was published on TA-4 antigen which had been identified in saline extracts of cervical squamous cell cancers /34/. Using a polyclonal rabbit antiserum directed against this antigen and following absorption, tumor cells of differentiated cervical squamous cell carcinoma could be stained cytoplasmically by means of direct immunofluorescence. SCCA which has been isolated and purified from liver metastases of cervical cancer is thought to represent one of 14 fractions of the TA-4 antigen (0.6% carbohydrate component) with a molecular weight of 42 kDa.

Immunohistochemically, TA-4 is detectable in 65% of large cell non squamous, and in 100% of large cell squamous cervical cancers but not in small cell non squamous cervical cancers /35/. Subcellularly, TA-4 is found within the cytosol, thus accounting for the fact that it is considered to be a structural protein and an index of differentiation for squamous cell carcinoma.

In addition, TA-4 is found in normal, dysplastic and malignant squamous cell tissues of the aero digestive tract with a high expression in the surface layer of normal squamous cell epithelium and in well differentiated squamous cell cancers. However, it is not detectable in dysplastic epithelial lesions of the oral cavity or in poorly differentiated squamous cell cancers /36/.

TA-4 antigen was localized by immunofluorescence in the tonofibrils of both normal buccal squamous epithelium as well as squamous cell cancers. It was identified as a 48 kDa protein and was classified as a normal cellular component associated with squamous epithelial cell differentiation.

More recent immunohistochemical investigations using a strongly reactive monoclonal antibody (MAb21) against TA-4 show stain uptake by the largest portions of the tumor nest and the intermediate layer of the non tumorous squamous epithelium of the cervix /37/.

After cloning and characterization of the SCCA antigen cDNA, it has been shown to have close amino acid homology (390 amino acids) with the serine protease inhibitor family (serpins) /38/. Cloning of the SCCA gene from normal genomic DNA led to the detection of two very similar genes, SCCA1 and SCCA2, together with other closely related serpins, at chromosome location 18q21.3. They encode for the following 45 kDa proteins with 92% identical amino acid sequences /39/:

  • SCCA1 encodes for a neutral form of protease inhibitors in normal and some malignant squamous cells
  • SCCA2 encodes for an acidic form of protease inhibitors predominantly detected in the cytosol of malignant epithelial cells and in the serum of tumor patients.

The two proteins show the following differences:

  • At their reactive site loops. SCCA1 is a potent inhibitor of the lysosomal cysteine proteinases cathepsins K, L and S, whereas SCCA2 (leupin) inhibits chymotrypsin like serine proteinases /40/.
  • SCCA1 inhibits apoptosis which may be an indication of its biological function /41/.

Elaborate proteomic, immunocytochemical and immunohistochemical analyses with Western blotting performed to analyze the expression level of SCCA1 and SCCA2 in keratinocytes and uterine squamous cell carcinoma cell lines showed that both proteins bind to the cytoplasmic carbonyl reductase /42/.

Specific discriminatory monoclonal antibodies and noncommercial ELISA with a detection limit of about 0.2 μg/L have been developed for SCCA1 and SCCA2. Compared to a previous, commercially available monoclonal ELISA, the newly developed assay allows the detection of both proteins, with a higher detection limit for SCCA2 than for SCCA1 /43/.


1. Stieber P, Fateh-Moghadam A, Knedel M. Squamous cell carcinoma (SCC)-Antigen in der Diagnostik und Verlaufsbeurteilung des Cervix-Karzinoms. GIT Lab Med 1987; 11–12: 554–8.

2. Mino-Miyagawa N, Kimura Y, Hamamoto K. An immunoradiometric assay of tumour-antigen (TA-4): a comparison with conventional radioimmunoassay. Br J Cancer 1990; 61: 520–3.

3. Ebert W, Stabrey A, Bülzebruck H, Kayser K, Merkle N. Efficiency of SCC antigen determinations for diagnosis and therapy-monitoring of squamous cell carcinoma of the lung. Tumor-Diagn & Ther 1988; 9: 87–95.

4. Petrelli NJ, Palmer M, Herrera L, Bhargava A. The utility of squamous cell carcinoma antigen for the follow-up of patients with squamous cell carcinoma of the anal canal. Cancer 1992; 70: 35–9.

5. Fischbach W, Rink C. SCC-Antigen: ein sensitiver und spezifischer Tumormarker für Plattenepithelkarzinome? Dtsch Med Wschr 1988; 113: 289–93.

6. Crombach G, Würz H, Herrmann F, Kreienberg K, Möbus V, Schmidt-Rhode P, Sturm G, Caffier H, Kaesemann H. Bedeutung des SCC-Antigens in der Diagnostik und Verlaufskontrolle des Zervixkarzinoms. Dtsch Med Wschr 1989; 114: 700–5.

7. Mino N, Iio A, Hamamoto K. Availability of tumor-antigen 4 as a marker of squamous cell carcinoma of the lung and other organs. Cancer 1988; 62: 730–4.

8. Molina R, Filella X, Torres MD, Ballesta AM, Mengual P, Cases A, Balaque A. SCC antigen measured in malignant and nonmalignant diseases. Clin Chem 1990; 36: 251–4.

9. Ebert W, Johnson JT. Tumor markers in the management of squamous cell carcinoma of the head, neck, and lung. Princeton: Excerpta Medica, 1987.

10. Calabrese F, Lunardi F, Giacometti C, Marulli G, Gnoato M, Pontisso P, et al. Overexpression of squamous cell carcinoma antigen in idiopathic pulmonary fibrosis: clinicopathological correlations. Thorax 2008; 63: 795–802.

11. Suzuki K, Inokuchi A, Miyazaki J, Kuratomi Y, Izuhara K. Relationship between squamous cell carcinoma antigen and the clinical severity of allergic rhinitis caused by dermatophagoides farinae and Japanese cedar pollen. Ann Otolo Rhinol & Laryngology 2010; 119: 22–6.

12. Kato H, de Bruijn HWA. Tumor markers in the management of squamous cell carcinoma of the cervix and vagina. Princeton: Excerpta Medica, 1987.

13. Duk JM, van Voorst Vader PC, ten Hoor KA, Hollema H, Doeglas HMG, de Bruijn HWA. Elevated levels of squamous cell carcinoma antigen in patients with a benign disease of the skin. Cancer 1989; 64: 1652–6.

14. El-Rachkidy RG, Young HS, Griffiths CEM, Camp RDR. Humoral autoimmune responses to the squamous cell carcinoma antigen protein family in psoriasis. J Invest Dermatol 2008; 128: 2219–24.

15. Duk JM, Groenier KH, de Bruijn HWA, Hollema H, ten Hoor KA, von der Zoe AGJ, Aalders JG. Pretreatment serum squamous cell carcinoma antigen: a newly identified prognostic factor in early-stage cervical carcinoma. J Clin Oncol 1996; 14: 111–8.

16. Meier W, Eiermann W, Stieber P, Schneider A, Fateh-Moghadam A, Hepp H. Experience with SCC antigen, a new tumor marker for cervical cancer. Eur J Cancer Clin Oncol 1989; 25: 1555–9.

17. Murakami A, Nakagawa T, Fukushima C, Torii M, Sueoka K, Nawata S, et al. Relationship between decreased expression of squamous cell carcinoma antigen 2 and E-cadherin in primary cervical cancer lesions and lymph node metastasis. Oncology Rep 2008; 19: 99–104.

18. de Bruijn HWA, Duk JM, van der Zee AGJ, Pras E, Willemse PHB, Boonstra H, Hollema H, et al. The clinical value of squamous cell carcinoma antigen in cancer of the uterine cervix. Tumor Biol 1998; 19: 505–16.

19. Esajas MD, Duk JM, de Bruijn HWA, Aalders JG, Willemse PHB, Sluiter W, et al. Clinical value of routine serum squamous cell carcinoma antigen in follow-up of patients with early-stage cervical cancer. J Clin Oncol 2001; 19: 3960–6.

20. Takeda M, Sakuragi N, Okamoto K, Todo Y, Minobe SI, et al. Preoperative serum SCC, CA125, and CA19-9 levels and lymph node status in squamous cell carcinoma of the uterine cervix. Acta Obstet Gynecol Scand 2002; 81: 451–7.

21. Catanzaro JM, Guerriero JL, Liu J, Ullman E, Sheshadri N, Chen JJ, et al. Elevated expression of squamous cell carcinoma antigen (SCCA) is associated with human breast carcinoma. PLos one 2011; 6: e19096.

22. Ebert W, Dienemann H, Fateh-Moghadam A, et al. Cytokeratin 19 fragment CYFRA 21-1 compared with carcinoembryonic antigen, squamous cell carcinoma antigen and neuron-specific enolase in lung cancer. Eur J Clin Chem Clin Biochem 1994; 32: 189–99.

23. Lara PC, Cuyas JM. The role of squamous cell carcinoma antigen in the management of laryngeal and hypopharyngeal cancer. Cancer 1995; 76: 758–64.

24. Sugimachi K, Kitamura M, Matsuda H, Okudaira Y. Tumor antigen TA-4: an aid in detecting post-operative recurrence of oesophageal carcinoma. Dis Markers 1987; 5: 67–73.

25. Damle SR. Usefulness of squamous cell carcinoma antigen (SCC) in carcinoma of the esophagus. Clin Chem 1988; 34: 1299–1300.

26. Wishnow KI, Johnson DE, Fritsche HA. Squamous cell carcinoma antigen in genitourinary tumors. Int J Biol Markers 1989; 4: 226–8.

27. Yagi H, Danno K, Maruguchi Y, Yamamoto M, Imamura I. Significance of squamous cell carcinoma (SCC)-related antigen in cutaneous SCC. Arch Dermatol 1987; 123: 902–6.

28. Treotoli P, Fransvea E, Angelotti U, Antonaci S, Lup L, Mazzocca A, et al. Tissue expression of squamous cellular carcinoma antigen (SCCA) is inversely correlated to tumor size in HCC. Molecular Cancer 2009; 8: 29–36.

29. Zuin J, Veggani G, Pengo P, Gallotta A, Biasiolo A, Tono N, et al. Experimental validation of specificity of the squamous cell carcinoma antigen-immunoglobulin M (SCCA-IgM) assay in patients with cirrhosis. Clin Chem Lab Med 2010; 48: 217–23.

30. Biasolo A, Chemello L, Quarta S, Cavaletto L, Bortolotti F, Caberlotto C, et al. Monitoring SCCA-IgM complexes in serum predicts liver disease progression in patients with chronic hepatitis. J Viral Hep 2008; 15: 246–9.

31. Giannini EG, Basso M, Bazzica M, Contini P, Marenco S, Savarino V. Successful antiviral therapy determines a significant decrease in squamous cell carcinoma antigen-associated (SCCA) variants’ serum levels in anti-HCV positive cirrhotic patients. J Viral Hep 2010; 17: 563–8.

32. Turato C, Ruvoletto MG, Biasolo A, Quarta S, Tono N, Bernardinello E, et al. Squamous cell carcinoma antigen-1 (SERPINB3) polymorphism in chronic liver disease. Dig Liver Dis 2009; 41: 212–6.

33. Dibbelt L, Knuppen R. Squamous cell carcinoma antigen immunoactivity is normal in maternal serum but high and increasing in amniotic fluid during pregnancy. Clin Chem 1992; 38: 2161–2.

34. Kato H, Torigoe T. Radioimmunoassay for tumor antigen of human cervical squamous cell carcinoma. Cancer 1977; 40: 1621–8.

35. Matsuta M, Kagabu T, Nishiya I. Immunohistochemical studies on tumor antigen-4 in squamous cell carcinoma of uterine cervix. Acta Histochem Cytochem 1987; 20: 75–85.

36. Kearsley JH, Stenzel DJ, Sculley TB, Cooke RA. Cellular localisation of tumour antigen (TA-4) in normal, dysplastic and neoplastic squamous epithelia of the upper aerodigestive tract. Br J Cancer 1990; 61: 631–5.

37. Kato H, Suehiro Y, Morioka M, Torigoe T, Myoga A, Sekiguchi K, Ikeda I. Heterogeneous distribution of acidic TA-4 in cervical squamous cell carcinoma: immunohistochemical demonstration with monoclonal antibodies. JPN J Cancer Res 1987; 8: 1246–50.

38. Suminami Y, Kishi F, Sekiguchi K, Kato H. Squamous cell carcinoma antigen is a new member of the serine protease inhibitors. Biochem Biophys Res Commun 1991; 181: 51–8.

39. Schneider SS, Schick C, Fish KE, Miller E, Pena JC, Treter SD; Hui SM, Silverman GA. A serine proteinase inhibitor locus at 18q21.3 contains a tandem duplication of the human squamous cell carcinoma antigen gene. Proc Natl Acad Sci USA 1995; 92: 3147–51.

40. Schick C, Pemberton PA, Shi GP, Kamachi Y, Cataltepe S, Baruski AJ, et al. Cross-class inhibition of the cysteine proteinases cathepsins K, L, and S by the serpin squamous cell carcinoma antigen 1: a kinetic analysis. Biochemistry 1998; 37: 5258–66.

41. Suminami Y, Nagashima S, Vujanovic NL, Hirabayashi K, Kator H, Whiteside TL. Inhibition of apoptosis in human tumour cells by the tumour-associated serpin, SCC antigen-1. Br J Cancer 2000; 82: 981–9.

42. Murakami A, Fukushima C, Yositomi K, Sueoka K, Nawata S, Fujimoto M, et al. Tumor-related protein, the squamous cell carcinoma antigen binds to the intracellular protein carbonyl reductase. Int J Oncol 2010; 6: 1395–1400.

43. Cataltepe S, Schick C, Luke CJ, Pak SCO, Goldfarb D, Chen P, et al. Development of specific monoclonal antibodies and a sensitive discriminatory immunoassay for the circulating tumor markers SCCA1 and SCCA2. Clin Chim Acta 2000; 295: 107–27.

28.22 Thyroglobulin (Tg)

Lothar Thomas

Tg is a large glycoprotein that is stored in the follicular colloid of the thyroid gland. It functions as a pro hormone in the synthesis of thyroid hormones (Section 30.1 – Physiological function of the thyroid). Tg is produced by normal thyrocytes and well differentiated thyroid cancer cells. The fact that Tg is only synthesized in the thyroid gland gives Tg its high diagnostic significance as an organ specific tumor marker in differentiated thyroid carcinoma (DTC).

Thyroid cancer comprises the differentiated papillary and follicular types and the less differentiated medullary and anaplastic tumor types. The differentiated types of thyroid cancer account for approximately 90% (papillary 70–80%, follicular 10–20%) and the medullary type for 2–5% of cases (Section 28.12 – Calcitonin).

There is a regional variation in the incidence of thyroid cancers in Northern Europe, with 2.3 in women and 0.9 in men and 10.8 in women and 4.4 in men per 100,000 population/year in the UK and in Austria, respectively.

28.22.1 Indication

  • Monitoring of differentiated thyroid carcinoma after total thyroid ablation by surgery and radioiodine therapy
  • Destructive thyroiditis
  • Factitious thyrotoxicosis
  • Neonatal hypothyroidism of unknown origin (suspected missing or poorly developed thyroid gland).

28.22.2 Method of determination

Radioimmunoassay (RIA) /1/, immunometric assay (IMA) /2/, ELISA /3/. Calibration is performed with CRM 457. Tg determination under stimulation

The following principles for enhancing the diagnostic sensitivity of Tg detection of residual disease after surgical or radioiodine ablation of thyroid cancer are the endogenous and exogenous stimulation test.

Endogenous stimulation test

Principle: following ablation of the thyroid gland, the patient receives thyroid hormone treatment to suppress the release of Tg by any residual thyroid tissue for 4 weeks prior to blood collection. Dis continuation of thyroid hormone therapy causes the suppressed TSH level to increase to more than 30 IU/L, thus stimulating Tg production. Highly sensitive Tg assay (functional sensitivity 0.1 μg/L) are used. These assays still detect the release of low Tg concentrations despite suppressed Tg production after thyroid hormone therapy.

Exogenous stimulation test

Principle: the Tg production is stimulated by exogenously applied TSH. Recombinant TSH (rTHS) is administered intramuscularly at a dose of 0.9 mg on two consecutive days. Blood is collected on day 5 after the first injection.

28.22.3 Specimen

Serum, plasma: 1 mL

28.22.4 Reference interval

Serum/plasma: 0.1–1 (2) μg/L /4/

Euthyroid individuals, normal TSH, non smokers, no palpable or visible thyroid gland, no family history of thyroid gland cancer, no elevated thyroglobulin and thyroid peroxidase antibodies.

28.22.5 Clinical significance

Differentiated thyroid carcinoma (DCT) is a rather indolent, low morbidity and low mortality cancer. Patients mostly present with an indolent thyroid nodule or enlarged neck lymph node. In some cases, distant metastasis is the first clinical manifestation of the disease. DTC occurs in two histological types:

  • As follicular carcinoma mainly in iodine deficient regions and with primarily hematogenic metastasis
  • As papillary carcinoma which is the most common type in regions with an adequate supply of iodine. Its metastases are predominantly lymphogenic.

The 10-year survival rate is 93% for the papillary type and 85% for the follicular type /5/. Distant metastases develop in 5–23% of patients, are the most frequent cause of mortality and reduce the 10-year survival rate to 25–48% of patients /5/.

In children, DTC accounts for 1.5–3% of all cancers, with an annual incidence of 0.5–1.5 in 1 million children. Most cases are diagnosed during puberty. At the time of diagnosis, progression of the disease is usually more advanced than in adults. For instance, 20–60% of diagnosed children have extensive tumor, 40–80% have cervical lymph nodes and 20% have pulmonary metastases /6/. Tg in differentiated thyroid carcinoma (DTC)

Fine needle aspiration is the primary diagnostic procedure for evaluating thyroid nodules or enlarged neck lymph nodes. Papillary carcinoma is easy to characterize cytologically in contrast to follicular carcinoma where reliable discrimination between benign follicular adenomas and follicular carcinomas can be difficult. The detection of vascular invasion and/or infiltration of the capsule is only possible based on a histological sample and is regarded as a prerequisite for the diagnosis of follicular carcinoma /7/.

Since Tg is produced by normal and malignant follicular cells of the thyroid gland and is also expressed in nodular goiters at elevated concentrations, it has no diagnostic significance in the detection of the small number of cancers among the large number of benign nodules in endemic regions. The determination of Tg is not useful in tumor detection when the thyroid gland is still present. According to a study /8/, in the discrimination of a malignant from a benign nodule at a preoperative Tg concentration ≥ 400 μg/L, the odds ratio for malignity is 2.36.

When fine needle aspiration biopsy (FNAB) of a thyroid nodule yields indeterminable pathology, Tg determination is currently accepted as the serum marker of choice in the detection of well differentiated thyroid cancer (DTC) recurrence. In a retrospective review /9/ of 861 consecutive thyroidectomy patients; 297 patients had indeterminate FNAB, of which 68 had serum levels of Tg measured prior to surgery. In these patients a TG ≥ 75 μg/L indicated a DTC with diagnostic sensitivity of 81% on final pathology, if medullary carcinoma, anaplastic carcinoma, or lymphoma were excluded before.

Routine measurement of serum Tg before surgical treatment of DTC is not recommended according to the American Thyroid Management Guidelines because there is no evidence that a preoperative Tg level is a significant predictive factor in the postoperative monitoring and follow-up /10/.

DTC generally has a good prognosis and a high rate of curing after treatment. Approximately 20–30% of patients with papillary thyroid carcinoma develop local or distant recurrence. Patients with this carcinoma carrying the BRAFV600E mutation have a more aggressive tumor and poorer therapy outcome /11/. It is therefore important to confirm the presence of the mutation prior to treatment and for classification into high risk and low risk patients /12/. For the BRAFV600 mutation status, see Tab. 28.22-1 – BRAFV600E mutation status. Therapy monitoring of thyroid carcinoma

The Tg serum concentration reflects:

  • The mass of thyroid tissue
  • The residual healthy tissue and the tumor tissue in the presence of carcinoma
  • The extent or TSH receptor stimulation
  • Thyroid tissue injury.

Therefore, sampling for determining the Tg level should not be performed earlier than 4–6 weeks after thyroid gland injury (fine needle aspiration, radioiodine therapy) or surgical thyroid gland ablation /13/.

After surgery as the treatment approach, normal thyroid residues may remain and produce small amounts of Tg which may increase after thyroid hormone withdrawal or stimulation with rTSH /14/. The highest diagnostic specificity regarding the absence of thyroid tissue is achieved by thyroidectomy followed by radioiodine treatment to ablate any thyroid residues. After such treatment, any detectable Tg will indicate the presence of neoplastic thyroid lesions. Ultrasound examination of the neck, 131I whole-body scintigraphy and Tg determination under rTSH stimulation or after thyroid hormone withdrawal are recommended to improve diagnostic sensitivity. Treatment outcome is monitored 3–6 months after radioiodine therapy.

Therapy monitoring of low risk patients

The risk of recurrence in these patients is below 1% and 131I whole body scintigraphy is not necessary in the following findings /15/:

  • TNM classification pT1-2 pn0 M0 and
  •  131I uptake under radioiodine treatment (after 24 h) below 2% and no iodine-containing lesions of residual thyroid tissue and
  • No suspicious storage lesions outside the thyroid location detected by 131I whole body scintigraphy 3–5 days after ablation and
  • Tg concentration below 2 μg/L after endogenous or exogenous stimulation in the absence of Tg antibodies.

Therapy monitoring of high risk patients

The risk of recurrence is low in these patients and 131I whole body scintigraphy is not necessary in the following findings /14/:

  • Clinical investigations and/or imaging techniques do not indicate the presence of loco regional recurrence or metastasis and
  • In patients receiving thyroid hormone treatment, Tg is below 2 μg/L in the absence of Tg antibodies or non interfered Tg recovery and
  • Under endogenous or exogenous stimulation, Tg is below 2 μg/L during follow-up.

The two patient groups account for ≥ 80% of the total number of thyroid gland cancer patients and can be monitored by Tg determination under stimulation and by neck ultrasound examination. Follow-up

During follow-up, the dosage of thyroid hormone should be reduced to achieve a TSH value within the reference interval. However, persistence or recurrence of the disease cannot be excluded if the Tg concentration under thyroid hormone treatment is below the detection limit. An undetectable Tg level under TSH stimulation (e.g., at the time when a 131I whole body scintigram was obtained after complete ablation of the thyroid gland) indicates complete and persistent remission /14/. Therefore, Tg determination should be based on high accuracy in the lower measuring range and a functional sensitivity of the assay below 1 μg/L.

Tg concentrations under TSH stimulation are a decisive parameter for prognostic classification. Reproducible Tg elevations below 1 μg/L in patients receiving L-thyroxin treatment should be verified by Tg determination under TSH stimulation.

Up to 20% of clinically disease free patients have baseline Tg concentrations below 1 μg/L but, under endogenous or exogenous stimulation, show values ≥ 2 μg/L. In a third of these patients, persistence or recurrence of the disease is detected by imaging as indicated by elevated Tg levels during the further course of the disease. The remaining two thirds of patients remain recurrence free and show declining Tg levels in further stimulation tests /14/. However, in patients showing an increase in Tg to 2–4 μg/L under TSH stimulation, whole body scintigraphy with a high 131I dose is helpful for diagnosis and the decision on treatment /15/.

In a study /16/, the survival of patients with DTC and distant metastasis was investigated. The 10-year and 15-year survival rates were 70.6% and 64.9%, respectively. The independent predictors of survival were younger age, surgical dissection of neck lymph nodes and low rTSH-stimulated Tg level below 400 μg/L at the discovery of metastasis.

Sensitive Tg assays

Sensitive Tg assays with a functional sensitivity of 0.1 μg/L are used to detect persistent or recurrent disease during thyroid hormone treatment. Assays with a functional sensitivity of 0.2–0.3 μg/L were reported to result in the best combination of diagnostic sensitivity and specificity /17/. In a study /18/, serum Tg was determined 6 months after thyroid ablation (i.e., thyroidectomy plus radioiodine) in a sensitive assay under thyroid hormone treatment. Undetectable Tg predicted low risk of recurrence. When Tg became detectable during follow-up, the evaluation of Tg slope in a 3–6 months period accurately discriminated patients with DTC recurrence from those without recurrence.

TENIS syndrome

Elevated or increasing Tg concentrations in DTC patients despite negative 131I whole body scintigraphy is termed the TENIS syndrome. When the serum Tg exceeds 2–10 μg/L, one should use all of the latest diagnostic means and when no operable metastases can be located, empirical high dose radioiodine therapy should be considered /19/. In a different analysis /20/, 44% of patients with negative 131I whole body scintigraphy and elevated Tg showed spontaneous decrease in Tg concentrations. If, in the remaining patients, imaging techniques did not reveal metastasis, radioiodine treatment was recommended in high risk patients with Tg above 5 μg/L under rTSH stimulation or above 10 μg/L after thyroid hormone withdrawal.

Undetectable Tg in the confirmed presence of DTC, even under stimulation

Such cases may occur if there is a reduction in iodine uptake and Tg production by the DTC, for example, in little differentiated tumors, resulting in very high TSH levels. Tg in non cancerous thyroid disease

The pattern of Tg in non-cancerous thyroid disease is shown in Tab. 28.22-2 – Serum Tg levels in non cancerous thyroid disease.

28.22.6 Comments and problems

Blood collection

Sample collection 1.5 hours after fine-needle aspiration of papillary thyroid carcinoma temporarly increases Tg concentration from 7 μg/l to 63.6 μg/l /23/.

Method of determination

IMA and ELISA are more sensitive than RIA and have shorter incubation times. These assays are calibrated against the international reference standard CRM 457. Nevertheless, international collaborative studies have shown discrepancies between commercially available assays of up to 40–60%, due to molecular heterogeneity of Tg in DTC patients, resulting in different antibody specificity and leading to biased results /4/.

Comparison of measurement of Tg by mass spectrometry (Tg-MS), two Tg RIAS and 2 immunometric Tg assays in patients with Tg autoantibodies showed the following results /24/:

  • In TgAb negative samples diagnostic sensitivities and specificities of 100% and 74–100%, respectively, were observed across all assays
  • In TgAb positive samples, all immunoassays demonstrated assay dependent Tg underestimation, ranging from 41% to 86% in comparison to Tg-MS (common cutoff 0.5 μg/L)
  • Conclusions: Tg immunoassays remain the method of choice for Tg measurement in TgAb negative patients. In TgAb positive patients with undetectable Tg by immunometric assay, the Tg-MS will detect Tg in up to 20% additional cases.

Detection limit: in IMA and ELISA, the detection limit is about 0.01–0.02 μg/L (defined as the mean plus 2-fold standard deviation of the zero calibrator).

Functional sensitivity: in IMA and ELISA, it is 0.06–0.10 μg/L (defined as the lowest Tg concentration with an inter assay variation coefficient of 20%).

High-dose hook effect: results in falsely low values in the case of TG concentrations above 1,000 μg/L, especially in IMA. Therefore, the sample should also be analyzed at a dilution of 1 : 9.

Interference factors

Heterophile antibodies: up to 3% of patient samples contain heterophile antibodies which can give falsely low Tg concentrations below 1 μg/L. They cause false negative results in IMA and false positive results in competitive assays such as RIA. Incubation of the sample in heterophilic blocking tubes can adsorb the interference by heterophile antibodies.

Thyroglobulin (Tg) antibodies: these antibodies represent the greatest interference factor in Tg analysis. They have been observed in more than 10% of the population. In particular, they interfere with IMA giving falsely low results. Therefore, samples should be analyzed for Tg antibodies before the Tg concentration is determined. As a rule, Tg antibodies disappear in the first years after total thyroid ablation. Tg antibody persistence is suspicious for persistence or recurrence of tumor. The Tg antibody level does not correlate with the degree of assay interference.

Positive Tg antibody assay is usually followed by a Tg recovery test. The procedure for detecting interference from Tg antibodies in the sample with Tg determination is as follows: a defined amount of Tg is added to the patient sample. Results outside the recovery range of 100 ± 30% suggest interference with Tg determination. In such a case, the result of Tg determination is not usable.


At 20 °C for 48 h and at 4–8 °C for 1 week.

28.22.7 Pathophysiology

Human Tg is a dimeric glycoprotein with a molecular weight of 670 kDa consisting of two identical subunits. It is produced by the follicular cells of the thyroid. Tg contains 20 asparagine linked, branched oligosaccharide chains which, like the peptide residues, are phosphorylated or sulfated and contribute to the variability of the molecule. In addition, a variable amount of iodine binds to the tyrosine residues in the protein. The iodine uptake is dependent on iodine supply to the body, iodine distribution to the follicles and the catalytic activity of thyroid peroxidase (Section 30.1 – Physiological function of the thyroid). Tg provides a matrix for thyroid hormone synthesis, is subject to post translational iodination to a varying extent and functions as a storage vehicle for thyroid hormone in the colloid of the follicles.

Thyroid carcinomas, especially those derived from follicular cells, produce a Tg with additional, complex antigenicity. Branching of the oligosaccharide chains differs from that in healthy individuals and iodination is reduced. Since reference preparation is based on healthy follicles, Tg concentrations measured in DTC patients using various commercially available assays will show discrepancies in some cases.

Since the TSH receptor exhibits transcriptional control of the Tg gene and Tg synthesis, the diagnostic sensitivity for detecting residual Tg activity after thyroid ablation will be highest under TSH stimulation.

According to the American Thyroid Association Guidelines, determination of molecular markers (BRAF, RAS, RET-PC, Pax8-PPARg or galactin 3) should only be considered if fine needle aspiration biopsy yields an indeterminate result /10/.


1. Bodlaender P, Arjonilla JR, Sweat R, Twomey SL. A practical radioimmunoassay of thyroglobulin. Clin Chem 1978; 24: 267–71.

2. Morgenthaler NG, Froehlich J, Rendl J, Willnid M, Alonso C, Bergmann A et al. Technical evaluation of a new immunoradiometric and a new immunoluminometric assay for thyroglobulin. Clin Chem 2002; 48: 1077–83.

3. Zöphel K, Wunderlich G, Franke WG. Erste Erfahrungen mit einem hochsensitiven Enzyme-linked-Immunosorbent Assay (ELISA) zur Thyreoglobulinbestimmung bei Patienten mit differenziertem Schilddrüsenkarzinom. J Lab Med 2002; 26: 425–33.

4. Giovanella L. Highly sensitive thyroglobulin measurements in differentiated thyroid carcinoma management. Clin Chem Lab Med 2008; 46: 1067–73.

5. Lee Y, Sohn EY. Differentiated thyroid carcinoma presenting with distant metastasis at initial diagnosis, clinical outcomes and prognostic factors. Ann Surg 2010; 251: 114–9.

6. Lazar L, Lebenthal J, Steinmetz A, Yackobovitch-Gavan M, Phillip M. Differentiated thyroid carcinoma in pediatric patients: comparison of presentation and course between pre-pubertal children and adolescents. J Pediatr 2009; 154: 708–14.

7. Baloch ZW, LiVolsi VA. Fine-needle aspiration of the thyroid: today and tomorrow. Best Pract Res Clin Endocrinol Metab 2008; 22: 929–39.

8. Petric R, Perhavec A, Gazic B, Besic N. Preoperative serum thyroglobulin concentration is an independent predictive factor of malignancy in follicular neoplasms of the thyroid gland. J Surg Oncol 2011; 105: 351–6.

9. Sands NB, Karls S, Rivera J, Tamila M, Hier MP, Black MJ, et al. Preoperative serum thyroglobulin as an adjunct to fine-needle aspiration in predicting well-differentiated thyroid cancer. J Otolaryngol Head Neck Surg 2010; 39: 669–73.

10. Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL, Mandel SJ, et al. Revised American thyroid association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009; 19: 1167–1214.

11. Gao WL, Wie LL, Chao YG, Wie L, Song TL. Prognostic prediction of BRAFV600E and its relationship with sodium iodide symporter in classic variant of papillary thyroid carcinomas. Clin Lab 2012; 58: 919–26.

12. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. Mutations of the BRAF gene in human cancer. Nature 2002; 417: 949–54.

13. Clark PM. Laboratory services for thyroglobulin and implications for monitoring of differentiated thyroid cancer. J Clin Pathol 2009; 62: 402–6.

14. Francis Z, Schlumberger M. Serum thyroglobulin determination in thyroid cancer patients. Best Practice & Research Clinical Endocrinology & Metabolism 2008; 22: 1039–46.

15. Dietlein M, Dressler J, Eschner W, Grünwald F, Lassmann M, Leisener B, et al. Verfahrensanweisung für die 131J-Ganzkörper-Szintigraphie beim differenzierten Schilddrüsenkarzinom (Version 3). Deutsche Gesellschaft für Nuklearmedizin e. V. Leitlinien

16. Huang IC, Chou FF, Liu RT, Tung SC, Chen JF, Kuo MC, et al. Long-term outcomes of distant metastasis from differentiated thyroid carcinoma. Clin Endocrinol 2011; 76: 439–47.

17. Schlumberger M, Hitzel A, Toubert ME, Corone C, Troalen F, Schlaeger MH, et al. Comparison of seven serum thyroglobulin assays in the follow-up of papillary and follicular thyroid cancer. J Clin Endocrinol Metab 2007; 92: 2487–95.

18. Giovanella L, Maffioli M, Ceriani L, De Palma D, Spriano G. Unstimulated high sensitive thyroglobulin measurement predicts outcome of differentiated thyroid carcinoma. Clin Chem Lab Med 2009; 47: 1001–4.

19. Silberstein EB. The problem of the patient with thyroglobulin elevation but negative iodine scintigraphy: the TENIS syndrome. Semin Nucl Med 2011; 41: 113–20.

20. Chao M. Management of differentiated thyroid cancer with rising thyroglobulin and negative diagnostic radioiodine whole body scan. Clin Oncol 2010; 6: 438–47.

21. Hüfner M. Thyreoglobulin. In Thomas L, ed. Labor und Diagnose. Frankfurt 2008, Th-Books S. 1364–6.

22. Uller RS, von Herle J. The effect of therapy on serum thyroglobulin levels in patients with Graves’ disease. J Clin Endocrinol Metab 1978; 46: 747–55.

23. Hopp AM, Ranjitkar P, Colon-Franco JM. Unexpected Change in thyroglobulin concentration. Clin Chem 2017; 63: 1775.

24. Netzel BC, Grebe SKG, Carranza Leon BG, Castro MR, Clark PM, Hoofnagle AN, et al. Thyroglobulin (Tg) testing revisited: Tg assays, TgAb assays, and correlation with results. J Clin Endocrinol Metab 2015; 100: E1074 –E1083.

28.23 Pro-gastrin releasing peptide (ProGRP)

Lothar Thomas

ProGRP is a tumor marker for small cell lung cancer (SCLC) and precursor of gastrin releasing peptide (GRP) a gut hormone which is widely distributed in the gastrointestinal tract, nervous system and pulmonary tract. GRP is also produced by small cell lung cancer (SCLC) cells and is therefore helpful in the diagnosis of SCLC. However, GRP is not useful as a tumor marker for SCLC in routine clinical diagnostics due to poor stability in specimens.

The GRP gene encodes three types of ProGRP with 115, 118, and 125 amino acids. The ProGRP types consist of a signal peptide, the native GRP (1–27), a cleavage site (residues 28–30), a constant region (residues 31–98), and a variable carboxy terminal region that results from alternative splicing. The ProGRP residue 31-98 is a more stable molecule than GRP and is used for diagnosis, prediction of prognosis, follow-up and monitoring of treatment in patients with SCLC /1/.

28.23.1 Indication

  • Suspected lung cancer
  • Differential diagnosis of lung cancer
  • Pulmonary round lesions of uncertain etiology
  • Follow-up and therapy monitoring of SCLC
  • Follow-up and therapy monitoring of medullary thyroid cancer.

28.23.2 Method of determination

Competitive ELISA on micro titer plate wells using anti-ProGRP 31–98 monoclonal antibody /2/.

Automated tests (e.g., two step sandwich assay). The analyte capture is performed with paramagnetic micro particles coated with the two monoclonal antibodies: 3G2 (amino acids of ProGRP 84–88) and 2B10 (amino acids of ProGRP 71–75) which can capture the C-terminal side of ProGRP 31–98. Detection of the ProGRP analyte micro particle complex is accomplished with an acridinium labeled mouse monoclonal antibody conjugate. The mouse monoclonal antibody 3D6-2 can capture the N-terminal side of the protein (amino acids of ProGRP 40–60). Exposing the reaction mixture to on-board trigger reagents containing peroxide at alkaline pH causes light production that is proportional to the ProGRP concentration /3/.

28.23.3 Specimen

Serum, heparinized plasma, EDTA plasma, depending on the test kit manufacturer’s specifications: 1 mL

28.23.4 Reference interval

Refer to Tab. 28.23-1 – ProGRP cutoff values.

28.23.5 Clinical significance

ProGRP can be elevated in benign and malignant disease (Tab. 28.23-2 – ProGRP in healthy individuals and in patients with malignant tumors). ProGRP in healthy individuals and in benign disease

Depending on the study, healthy individuals can have ProGRP concentrations of up to 75 ng/L /7/, while other studies did not find any concentration above 50 ng/L /5/.

Approximately 10% of patients with benign disease have ProGRP levels above 50 ng/L /5/. These patients predominantly suffer from lung disease, gastrointestinal disease and renal insufficiency (Tab. 28.23-2 – ProGRP in healthy individuals and in patients with malignant tumors).

Levels of up to 350 ng/L are measured in patients with impaired renal function /8/. ProGRP levels of up to 140 ng/L have been observed in patients with benign gastrointestinal diseases, urologic diseases and infections with markedly increased CRP. Levels up to 80 ng/L can be found in benign breast and lung diseases. ProGRP in malignant disease except the lung

Elevated ProGRP not correlated with tumor stage are found in malignant disease not involving the lung. This applies, for example, to small cell neuroendocrine tumors such as esophageal cancer or prostate cancer /9/. The elevations can be especially pronounced in medullary thyroid cancer /10/. ProGRP in small cell lung cancer

SCLC is a rapidly growing, aggressive neoplasia and, in many cases, presenting patients already have lymph node or peripheral organ metastasis. The tumor shows good response to chemotherapy and radiotherapy which is thought to be due to neuroendocrine differentiation of SCLC. Prognosis and treatment of SCLC differ from those in non small cell lung cancer (NSCLC). Therefore, the availability of tumor markers which allow differential diagnosis, assessment of the course of the disease and monitoring of therapy is important for the monitoring and follow-up of lung cancer. Sensitive tumor markers are /5/:

  • Cyfra 21-1, CEA and SCCA in NSCLC
  • NSE and ProGRP in SCLC.

Except for ProGRP, the tumor markers have the following disadvantages /5/:

  • Lack of cancer specificity because abnormal values as found in lung cancer are also seen in other malignant tumors
  • Lack of diagnostic sensitivity in follow-up and therapy monitoring requiring the use of two or more tumor markers
  • No clear correlation between the histological type of the tumor and the tumor marker. For instance, markedly elevated concentrations of CEA are found in adenocarcinomas and of Cyfra 21-1 in squamous cell cancer, but both tumor markers can also be elevated in other histological types and SCLC
  • NSE is an important tumor marker in SCLC regarding diagnosis, prognosis, follow-up and therapy monitoring, but has the disadvantage of low diagnostic sensitivity in limited disease (10–20%). ProGRP in SCLC diagnosis

At a concentration of approximately 150 ng/L, ProGRP achieves a diagnostic sensitivity of 100% for SCLC. According to a study /5/, 73% of SCLC patients have ProGRP levels above 50 ng/L, of which 65% with limited disease (LD) and 78% with extensive disease (ED). In all patients, the ProGRP concentration was 598 ± 1,448 ng/L (286 ± 423 ng/L in LD and 799 ± 1,809 ng/L in ED).

In a different study /6/, in patients with LD and a cutoff of 49 ng/L, the diagnostic sensitivity was 79.7% for ProGRP versus 57.8% for NSE. The ROC analysis (Fig. 28.23-1 – ROC analysis for differentiation of patients with SCLC from healthy persons) showed the superiority of ProGRP over the other tumor markers used in lung cancer.

According to a meta analysis /11/, the diagnostic sensitivity of ProGRP for SCLC was 72%, with a specificity of 93%. In general, ProGRP release did not correlate with tumor stage. ProGRP was detected in the LD stage with a similarly high sensitivity as in the ED stage. Differential diagnosis of pulmonary round lesions

Various incidences have been reported on elevated ProGRP in NSCLC:

  • Approximately 14% at a cutoff of 35 ng/L /12/.
  • Approximately 30% at cutoff > 50 ng/L /5/.

In discriminating between SCLC and NSCLC, the diagnostic sensitivity of ProGRP was 78.4%, with a specificity of 95%, while the diagnostic sensitivity of NSE was only 48.6% /13/.

About 80% of 233 SCLC patients and 8.1% of 421 NSCLC patients showed concentrations above a cutoff value of 46 ng/L /13/. Histologically, the NSCLC patients either had adenocarcinoma, squamous cell cancer or large cell lung cancer. It was assumed that the ProGRP release of these cancers is based on neuroendocrine differentiation.

Lung metastasis of tumors other than NSCLC can cause the release of ProGRP to a low extent, with concentrations below 100 ng/L /7/. Follow-up and therapeutic monitoring

NSE, Cyfra 21-1 and ProGRP decrease significantly during chemotherapy. ProGRP responds more sensitively to chemotherapy and decreases continuously, while the other markers show fluctuations near or within the reference interval. The percentage of patients in remission with elevated ProGRP levels was higher than those with elevated NSE /6/. In patients with recurrent SCLC, the diagnostic sensitivity of ProGRP to detect the relapse was 74% that of NSE 32% and of CEA 56% /15/. Notably, all patients with elevated ProGRP at primary tumor diagnosis also showed increased ProGRP release at relapse (12% false negative findings for NSE, 6% for CEA).

During follow-up of lung cancer patients, an increased marker value was found /16/:

  • For ProGRP in 51% and for NSE only in 25.5% of SCLC patients
  • For ProGRP in 8.6% and for Cyfra 21-1 in 55.7% of NSCLC patients.

The concurrent determination of ProGRP and NSE has an additive effect, with a diagnostic sensitivity of 79%.

28.23.6 Comments and problems


Some assays show significant variations in ProGRP depending on the sample. For instance, in the immunoassay described in Ref. /3/, the plasma concentrations were 103% (median) higher than the serum concentrations, and the relative elevation was higher in patients with SCLC than in those with benign disease /17/. It is therefore important to follow the manufacturer’s specifications regarding the specimen.

Method of determination

ProGRP is present in very low levels in serum (ng/L); hence, the intraindividual variation is high. Consequently, increases in ProGRP in levels below the 95th percentile of healthy controls should not be overestimated.


1. Yamaguchi K, Abe K, Kamoya T, Adachi I, Taguchi S, Otsubo K, Yanaihura N. Production and molecular size heterogeneity of immunoreactive gastrin-releasing pep­tide in fetal and adult lungs and primary lung tumors. Cancer Res. 1983; 43: 3932–9.

2. Aoyagi K, Miyake Y, Urakami K, Kashiwakuma T, Hasegawa A, Kodama T, Yamaguchi K. Enzyme immunoassay of immunoreactive progastrin-releasing peptide (31–98) as tumor marker for small-cell lung carcinoma: development and evaluation. Clin Chem 1995; 41: 537–43.

3. Yoshimura T, Fujita K, Kinukawa H, Matsuoka Y, Patil RD, Beligere GS, et al. Development and analytical performance evaluation of an automated chemiluminescent immunoassay for pro-gastrin releasing peptide. Clin Chem Lab Med 2009; 47: 1557–63.

4. Stieber P. ProGRP (Pro Gastrin Releasing Peptide). In Thomas L, ed. Labor und Diagnose, 7th. edition. Frankfurt 2009; TH-Books: S. 1338–41.

5. Molina R, Auge JM, Filella X, Vinolas N, Alicarte J, Domingo JM, et al. Pro-gastrin releasing peptide (proGRP) in patients with benign and malignant diseases: comparison with CEA, SCC, CYFRA 21-1 and NSE in patients with lung cancer. Anticancer Res 2005; 25: 1773–8.

6. Wojcik E, Kulpa JK, Sas-Korczynska B, Korzeniowski S, Jakuboeicz J. ProGRP and NSE in therapy monitoring in patients with small cell lung cancer. Anticancer Res 2008; 28: 3027–34.

7. Yamaguchi K, Stieber P. Diagnosis of small cell lung cancer by Pro Gastrin Releasing Peptide (ProGRP). J Lab Med 2003; 27: 26–30.

8. Nakahama H, Tanaka Y, Fuijita Y, Fujii M, Sugita M. CYFRA 21-1 and ProGRP, tumor markers of lung cancer, are elevated in chronic renal failure. Respirology 1998; 3: 207–10.

9. Nagakawa O, Furuya Y, Fujiuchi Y, Fuse H. Serum progastrin-releasing peptide (31–98) in benign prostatic hyperplasia and prostatic carcinoma. Urology 2002; 60: 527–30.

10. Inaji H, Komoike Y, Motomura K, Higashiyama M, Oht­suru M, Funai H, Kasugai T, Koyama H. Demonstration and diagnostic significance of pro-gastrin-releasing peptide in medullary thyroid carcinoma. Oncology 2000; 59: 122–5.

11. Tang J, Zhang X, Zhang Z, Wang R, Zhang H, Zhang Z, et al. Diagnostic value of tumour marker pro-gastrin-releasing peptide in patients with small cell lung cancer: a systematic review. Chin Med J 2011; 124: 1563–8.

12. Takada M, Kusonoki Y, Masuda N, Matui K, Yana T, Ushujama S, et al. Pro-gastrin releasing peptide (31–98) as a tumor marker of small-cell lung cancer: comparative value with neuron-specific enolase. Br J Cancer 1996; 73: 1227–32.

13. Nisman B, Biran H, Ramu N, Heching N, Barak V, Peretz T. The diagnostic and prognostic value of ProGRP in lung cancer. Anticancer Res 2009; 11: 4827–32.

14. Kudo K, Ohyanagi F, Horiike A, Miyauchi E, Hishi R, Satoh Y, et al. Clinicopathological findings on non-small-cell lung cancer with high serum progastrin-releasing peptide concentrations. Lung Cancer 2011; 74: 401–4.

15. Niho S, Nishiwaki Y, Goto K, Ohmatsu H, Matsumoto T, Hojo F, Ohe Y, Kakinuma R, Kodama T. Significance of serum pro-gastrin-releasing peptide as a predictor of relapse of small cell lung cancer: comparative evaluation with neuron-specific enolase and carcinoembryonic antigen. Lung Cancer 2000; 27: 159–67.

16. Wojcik E, Rychlik U, Skotnicki P, Jakubowicz J, Kulpa JK. Utility of ProGRP determinations in cancer patients. Clin Lab 2010; 56: 527–34.

17. Kim HR, Oh IJ, Shin MG, Park JS, Choi HJ, Ban HJ, et al. Plasma proGRP concentration is sensitive and specific for discriminating small cell lung cancer from nonmalignant conditions or non-small cell lung cancer. J Korean Med Sci 2011; 26: 625–30.

Table 28.1-1 Age dependent incidence of frequently encountered cancers in Great Britain



♂ years

♀ years










< 0.1





< 0.1




< 0.1

< 0.1




< 0.1




< 0.1





< 0.1




< 0.1














< 0.1




< 0.1





< 0.1




< 0.1























< 0.1













< 0.1
































* Based on 1 million inhabitants; NHL, non Hodgkin lymphoma

Table 28.1-2 Incidence of most frequently encountered cancers* /12/










































































* Values in thousand per year (113 refers to 113,000)

LL, leukemias and lymphomas; MP, mouth and pharynx

Table 28.1-3 Common types of cancer in Germany /11/

Men (%)

Women (%)































Ovaries, tubes


Urinary tract


Gall bladder


Lymph node




Table 28.1-4 Tumor classification according to the TNM system /3/



Primary tumor



Small tumor, not spread to neighboring tissue



Medium sized tumor, slightly spread



Large tumor, strongly spread to neighboring tissue



Very large tumor, extensively spread to neighboring tissue



Tumor can not be evaluated



Lymph node involvement



Tumor cells absent from regional lymph nodes



Regional lymph node metastasis present



Tumor spread to an extent between N1 and N3



Tumor spread to more distant or numerous regional lymph nodes



Presence of lymph node involvement cannot be evaluated



Presence of distant metastasis



No distant metastasis



Distant metastasis is present



Presence of distant metastasis cannot be evaluated

For example, cancer classified as T3N2M1 refers to a large tumor, strongly spread to environmental tissue and regional lymph node involvement spread to neighboring tissue and organ metastasis.

Table 28.2-1 Oncogenes, oncogene products and tumor suppressor genes

Oncogenes and oncogene products

By definition, an oncogene is a gene whose abnomal expression or altered gene product leads to malignant transformation. Cell derived oncogene products are growth factors, receptor and nonreceptor protein-tyrosine kinases (PTKs), receptors lacking PTK activity, membrane-associated G proteins, cytoplasmic protein-serine kinases, and cytoplasmic regulators /4/.


HER 2, also called Human Epidermal growth factor Receptor 2, c-webB2, ERBB2, CD340. HER2 gene encodes endothelial growth factor receptor (p185). A mutation in the HER2 (oncogene) encodes a truncated form of the endothelial growth factor receptor (EGFR) that has a cytoplasmic domain which is continuously active. Cells expressing this oncogene behave as though they were constantly being signaled to proliferate, even if no EGFR is present. Approximately 20% of women with breast carcinoma have a strong increase in EGFRs inducing an increase in growth signals that cause an uncontrollably increase in cells /12/. An antibody directed against the EGFR inhibits the growth of tumor cells and labels the tumor cells that the cells can be recognized by the immune system.


EGFR is one of the ERBB family receptor tyrosine kinases that consists of four members: EGFR (also known as ERBB1/HER1), ERBB2/HER2/NEU, ERBB3/HER3 and ERBB4/HER4. Specific ligands bind to the extracellar domain of EGFR, which leads to the formation of homodimers and heterodimers. Dimerization stimulates intrinsic tyrosine kinase activity of the receptors and triggers the autophosphorylation of specific tyrosine residues. Signal transducers initiate multiple downstream pathways such as MAPK, PL3K-AKT and STAT 3 and 5, which regulate proliferation and apoptosis /13/.


The ERBB2 gene (v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, commonly referred to as HER-2), is located on chromosome 17q21 and encodes human epidermal growth factor receptor 2 (HER2) a transmembrane glycoprotein belonging to the tyrosine kinase receptor family, whose members play an important role in the regulation of such fundamental processes as cell growth, survival and differentiation. HER2 overproduction plays a pivotal role as a prognostic marker, by itself or in combination with other markers. The tumor cells in 20–30% of women with breast cancer show amplification of ERBB2 /14/.


The gene FLT3 (Fms-like tyrosine kinase 3; FLT3) encodes the receptor of FLT3 on the surface of the cell membrane. The receptor contains a tyrosine kinase that is of importance for the fuction of myeloid cell. For the development of acute myeloid leukemia mutations of the gene FLT3 can be responsible.


The BCR-ABL1 tyrosin kinase (Abelson murine leukemia viral oncogene homolog 1) is an important player in signal transduction. The protein ABL1 is encoded by the gene c-ABL. The exchange of fragments between chromosome 9 (carrier of c-ABL) and chromosome 22 (carrier of BCR) on the newly produced Philadelphia chromosome results in the new gene BCR-ABL. The oncogene BCR-ABL encodes the BCR-ABL tyrosin kinase and is diagnosed in the majority of cases with chronic myeloid leukemia (CML). Tyrosine kinase inhibitors (imatinib, dasatinib, nilotinib) are effective in CML treatment.


The Wnt family of secreted glycoproteins inhibits the phosphorylation of β-catenin. The protein is involved in the regulation of cell-cell adhesion and activation of signaling pathways. The activity of β-catenin is controlled by the adenomatous polyposis coli (APC) protein. In familial adenomatous polyposis, a mutation in the gene APC inhibits phosphorylation and, thus, degradation of β-catenin. Free β-catenin in plasma translocates into the nucleus where it activates cell proliferation.


The BRAF gene (proto oncogene B-Raf or v-Raf murine sarcoma viral oncogene homolog B1 encodes the serine/threonine protein kinase B-Raf) that is activated by the BRAF gene. The mutated proto oncogen can be found in malignant melanoma, colorectal carcinoma and small cell lung cancer (SCLC). About half of the patients with papillary thyroid cancer have tumors with BRAFV600E /15/.


Receptor tyrosine kinases (RTKs) play an important roles in normal development. Several RTK genes (ALK, NTRKs, RET, EGFR and IGFR) are implicated in malignant transformation or progression of neuroblastoma /16/.


The vascular endothelial growth factor (VEGF) and its receptor (VEGFR) play important roles in physiological and pathological angiogenesis, such as cancer. VEGF regulates angiogenesis and vascular permeability by activating VEGFRs. The VEGFRs are typical tyrosine kinase receptors (TKRs) carrying an extracellular domain for ligand binding, a transmembrane domain, and a cytoplasmic domain, including a tyrosine kinase domain.

VEGF plays an important role in hypoxia dependent control of gene transcription. VEGF activity is mediated by the tyrosine kinase receptors VEGFR1 (FLT1), VEGFR2 (FLK1-KDR) and VEGFR3 (FLT4). VEGF stimulates angiogenesis in numerous cancers /17/.

VEGF pathways inhibitors represent rational treatments. VEGF inhibitors and VEGFR inhibitors share the same approach. VEGF inhibitors bind to the signal molecule, VEGFR inhibitors bind directly to the receptor. The VEGF inhibitors belong to either of two main classes; bevacizumab is a monoclonal antibody and all the rest are tyrosine kinase inhibitors (TKIs)

Deletion in exon 19 is found in 10–15% of patients with non small cell lung carcinoma (NSCLC) the main subgroup (approximately 75%) of lung carcinoma. Defects in the ligand binding domain of the VEGFR due to deletion in the gene VEGFR lead to continuous receptor stimulation without ligand binding. The activated receptor phosphorylates tyrosines in the intracellular domain of the receptor, providing interaction sites for cytoplasmic proteins. The interactions deregulate intracellular signaling in various ways. Activating mutations are found in the other three members of the VEGFR family (ERBB2, ERBB3 and ERBB4) and in the kinase domains of HER2/neu signal receptors and KIT signal receptors /17/.


The Myc proteins are members of a basic region/helix-loop-helix/leucine zipper (bHLHZip) transcription factor family and are implicated in the regulation of cellular proliferation, differentiation, and apoptosis. The Myc proteins are encoded by the Myc gene. Deregulation of Myc expression can play a causual role in the genesis of malignancies. The manner in which Myc regulates cell cycle progression is likely through its capacity to modulate expression or activity of cell cycle specific genes. Up regulated expression of Myc is linked to proliferation. The Myc is countered by intrinsic suppressor mechanisms that most often trigger apoptosis. Therefore, Myc needs genetic or epigenetic effects (e.g., over expression of BCL2) to suppress apoptosis and induce tumorigenesis. Signaling pathways activated by food, growth factors or mitogenic stimuli convey their signals from the phosphorylation cascades directly to Myc.

Deregulation of the Myc expression is a common finding in more than 50% of carcinomas. Mutation induced over expression of Myc prevents the cell from entering the G0 inactivation phase, allowing proliferation to continue unabated. In many cases, the concurrent activities of Myc and Bcl-2 cause a cell to become cancerous /18/.

PI3K (phosphoinositide 3-Kinase) and AKT

Cells use the PI3K-AKT signaling pathway to respond to cytokines, G-protein coupled receptor ligands, growth factors and cellular stress. Proteins involved in cell proliferation and apoptosis suppression are activated. The enzyme PI3K catalyzes the phosphorylation of phosphatidylinositol-3,4-bisphosphonate (PIP2) to create phosphatidylinositol-3,4,5-triphosphonate (PIP3). The binding of PIP3 to the PH domain anchors members of the Akt protein kinase to the cell membrane for phosphorylation. Aberrations in the PIK3K-AKT pathway are the most common genomic abnormalities and include e.g., the PI3K mutation and loss-of-function mutations. In many kinds of cancer (breast, ovarian urothelial cancer) aberrations of the PI3K-AKT signaling pathway are diagnosed /19/.

Ras (Rat sarcoma)

Approximately 30% of human cancers are associated with mutations of Ras proteins. These proteins are small monomeric GTPases encoded by the genes K-Ras, H-Ras and N-Ras. The wild type GTPases change conformation when binding to GTP and become inact encoding key players in the DNA activated by conversion of GTP to GDP. Oncogenic Ras protein remains bound to GTP, is active and binds to the effector proteins PIeK and Raf. Mutant Ras and PIK3CA have the potential to cooperate in PI3K pathway activation. Colorectal cancer with mutations in Ras/Raf/Mek/Erk and PI3k signaling pathways is common /12/. These proteins are cytoplasmic signal transducers. Point mutations in the codons 12, 13 or 61 convert Ras genes into Ras oncogenes causing amino acid changes in the Ras proteins. Ras oncogenes play an important role in the formation of colon cancer and pancreas cancer. A point mutation in codon 12 of this gene is found in 90% of pancreas cancers /20/.


The c-SRC gene is similar to the v-SRC gene of Rous sarcoma virus. This proto oncogene is known to play a role in the regulation of embryonic development and cell growth. Mutations in c-SRC may be involved in the malignant transformation of colon cancer /20/.


Germline mutations in genes encoding key players in the DNA damage response including BRCA1, BRCA2, BLM, FANCA ,TP53, RAD51 und MSH2, result in cancer susceptibility syndromes, in part because failure to adequately protect the genome against endogenous and exogenous sources of DNA damage results in the accumulation of oncogenic mutations. Genomic instability is therefore a hallmark of cancer. Cancer cells often harbor a reduced repertoire of DNA repair and DNA signaling capabilities compared with normal cells, and in some cancer cases also upregulate mutagenic repair pathways that drive oncogenes /7/.

A well recognized sensor of DNA damage is the protein Poly(ADP-ribose) polymerase (PARP), which is best known for its role in base excision repair and repair of DNA single-strand breaks, although it also has a less well-defined role in DNA double strand breaks repair by alternative nonhomologous end-joining /7/.


ETS (Erythroblast Transformation Specific) is a family of transcription factors throughout all stages of tumorigenesis. Specifically in solid tumors, gene arrangement and ampification, feed-forward growth factor signalling loops, formation of gain-of function co-regulatory complexes and novel cis-activating mutations in ETS target gene promoters can result in increased ETS activity. In turn, pro oncogenic ETS signalling enhances tumorigenesis through a broad mechanistic toolbox that includes lineage specification and self-renewal, DNA damage and genome instability, epigenetics and metabolism /21/.

The oncogenic activation of the ETS related gene (ERG) due to gene fusions is present in over half of prostate cancer in Western countries /22/.

Tumor suppressor genes

Tumor suppressor genes, normally function as physiologic barriers against clonal expansion or genetic mutability and are able to hinder the growth and metastasis of cells driven to uncontrolled proliferation by oncogenes /4/.

BRCA /8/

BRCA1 (BReast CAncer) and BRCA2 are tumor suppressor genes and encode proteins e.g., PARP (poly adenosine diphosphate-ribose polymerase). Cancer often presents genome instability and harbor defects in one or more of the six DNA repair pathways, most of them in homolgous recombination (HR)-deficient cells. PARP plays an important role in DNA damage repair and genome stability. PARP inhibitors inhibit DNA repair and of cancer cells, especially in homologous recombination HR-deficient cells /23/.

Patients carrying a mutation in the BRCA are at high risk of breast cancer, ovarian cancer and prostate cancer. If, in a mutation carrier, a copy of the BRCA is damaged in one of the BRCA (BRCA1 or BRCA2), the BRCA is inactivated and cancer can develop. Tumor cells with BRCA1 and BRCA2 mutations have numerous strand breaks of DNA.

p53 /10/

p53 is a sequence specific DNA binding protein of the nucleus, which inhibits uncontrolled cellular mitogenesis and promotes apoptosis in cellular stress events. The concentration of p53 increases in DNA damage due to phosphorylation at serine 13, preventing interaction between p53 and MDM2. The gene MDM2 encodes a nuclear localized E3 ubiquitin ligase (MDM2 oncogene). The MDM2 protein can promote tumor formation targeting tumor suppressor proteins, such as p53, for proteosomal degradation. Tumorigenesis is basically promoted by the failure of p53 to suppress cell proliferation following DNA damage, thus allowing mutant cells to survive and grow. Oncogenes, for example the Adenovirus EIA oncogene, can induce p53 formation, promoting apoptosis and premature cell aging /24/.

Abnormalities of the tumor suppressor gene TP53 are detected in more than half of human cancers. Acquired mutations of the gene occur in a wide spectrum of cancers, for example in breast cancer, lung cancer and colon cancer.

Table 28.2-2 Functional grouping of oncoproteins /13/

Cellular signal transduction

  • Growth factors (PDGF, Wnt)
  • Growth factor receptors (tyrosine kinase receptors ErbB, Her2/Neu)
  • Membrane bound ligand receptors (E-cadherin)

Cytoplasmic signaling molecules

  • Tyrosine kinases (Abl, Src)
  • Serine/threonine kinases (Bcr, Mos Pim 1)
  • Phosphatases (PTEN)
  • Adaptor proteins (Bcl1, Crk)
  • Membrane associated G-proteins (Ras)
  • Regulators of small G-proteins (Dbl, vav)

Phospholipid derivatives, second messengers (Pl3K, PTEN)

Gene transcription

  • Transcriptional activators (Myc, Jun, Myb, Rel)
  • Transcriptional repressors (WT1, Evil, PML-RARα)
  • Cofactors (CBP, Bmil, MLL)

Protein biosynthesis (EF1α)

Cell cycle (cyclin D, cyclin E, CDK4)

Apoptosis (Bcl-2)

Table 28.2-3 Oncogenes, oncogene products and tumor suppressor genes in cancer patients

Clinical and laboratory findings

Breast cancer

Estrogen receptor (ER) positive and HER2 positive breast cancers are the most common subsets of breast cancers with proportions accounting for 65% among women below 50 years of age and 75% of cases among older women. Estrogen binding to ER stimulates receptor-regulated transcription, which in turn promotes tumor cell growth and proliferation. Hereditary cancer genes account for 8–10% of ER-positive cancers. The prevalence of hereditary mutation is approximately 15% in patients less than 40 years, 10% among women 40–60 years and 5% among those higher than 70 years /2526/. Breast cancer risks with mutations in:

  • Genes associated with homologous recombination deficiency (BRCA1 and BRCA2) are 2%, respectively
  • Partner and localizer of BRCA2 (PALB 2) are 0.5–1% /27/
  • Ataxia-telangiectatica mutated (ATM) genes are 0.5–1% /28/
  • Checkpoint kinase gene 2 (CHEK2) is 1%. CHEK2 is a tumor suppressor gene encoding a serine-threonine kinase (CHK2) that acts a tumor suppressor and regulates cell division.

HER-2 gene: HER-2 determination is recommended in all patients with invasive breast cancer. The primary purpose for determining HER-2 is to select patients who may be treated with trastuzumab.

BRCA1 and BRCA2 genes: BRCA1 and BRCA2 mutation testing may be used for identifying women who are at high familial risk of developing breast or ovarian cancer. For those with such mutations, screening should begin at 25–30 years of age.

Mutations in the gene ESR1 or epigenetic changes in c-myc, cyclin D, and EGFR are associated with resistance to endocrine therapy.

Predisposition of hereditary breast cancer: BCRA1, BCRA2

Acute myeloid leukemia

Acute myeloid leukemia (AML) is characterized by clonal expansion of undifferentiated myeloid precursors, resulting in impaired hematopoiesis and bone marrow failure. Cancer develops from somatically acquired driver mutations, which account for the myriad of biologic and clinical complexities of the disease. In a study /29/ 5234 driver mutations across 76 genes or genomic regions, with 2 or more drivers were identified in 86% of patients /29/.

Endometrial cancer

Obesity and conditions associated with metabolic syndrome, including diabetes and polycystic ovary syndrome, are risk factors for develoment of endometrial cancer. Some molecular changes, especially alterations in the phosphatidylinositol 3-kinase (PI3K)-AKT pathway, are common across all endometrioid tumors and can even be detected in some endometrioid cases. Other molecular alterations, such as CTNNB1 mutation and MLH1 loss due to MLH1 gene methylation, are almost exclusively detected in endometrioid carcinomas. TP53 mutations are especially enriched in non-endometrioid carcinomas and a subset of grade 3 endometrioid tumors /30/.

Ovarian cancer (OC)

According to Global CancerObservatory there were an estimated 240.000 new cases of ovarian cancer each year, which account for 152,000 deaths worldwide The overall 5-year survival rate of all ovarian cancer is 47.6% while distant/late staged ovarian cancer has a 5-year survival of only 29.2% in the US. Genome instability is a hallmark of ovarian cancer, with almost half of the ovarian cancers harbor defects in one or more of the DNA repair pathways, and most of them are in homologous recombination (HR) DNA repair pathway. The high mutation rate of HR genes in ovaian cancer provided a unique opportunity for targeted therapy. The current standard care for ovarian cancer patients is debulking surgery followed by platinum-based chemotherapy. Even though ovarian cancer cells are initially sensitive to chemotherapeutic drugs, such as platinum analogues, they become resistant to these drugs over time. Thus, alternative therapy options to platinum-based chemotherapy, such as PARP inhibitor therapy, benefits ovarian cancer patients /31/.

Predisposition of hereditary ovarian cancer: BCRA1, BCRA2

Lung cancer

Advances in the analysis of the lung cancer genome have changed the understanding of this disease on molecular level. The most important genes responsible for the development of lung cancer are: EGFR, KRAS, MET, LKB1, BRAF, PIK3CA, ALK, RET and ROS1 /1332/.

NSCLC: EGFR and RAS mutations contribute to the development of NSCLC. BRAF mutations have been identified in 1–3% of cases. Less frequent mutations, such as ROS1 and RET mutations in patients with NSCLC occur at a rate of 2% and 1%, respectively. ROS1 mutations are almost always exclusively concomitant wit KRAS, EGFR, and ALK mutations

Squamous cell carcinoma: KRAS mutations are rare but may be found in 15–25% of adenocarcinomas.

SCLC: c-Met plays a significant role

Peutz-Jeghers syndrome: Association with loss-of function mutations in LKB1

Prostate cancer

Chromosomal rearrangements resulting in the fusion of TMPRSS2, an androgen related gene and the ETS family transcription factor ERG occur in 60% of prostate cancers with the majority of cases bearing the TMPRSS2-ERG fusion /33/. As a member of the ETS transcription family ERG binds specifically to the conseved ETS DNA-binding motif.

In patients with metastatic castration-resistant prostate cancer who had disease progression while receiving enzalutamide or abiraterone and who had alterations in genes with a role in homologous recombination repair, PARP inhibitor therapy was associated with longer progression-free survival and better measures of response and patient-reported end points than either enzalutamide or abiraterone /34/.

Colorectal carcinoma

Colorectal carcinoma is found to be the forth commonly occuring cancer worldwide. It is importat to investigate the polymorphism is several DNA repairing genes such as ATM, RAD51, XRCC2, XRCC3, and XRCC9. In a study /35/ the level of ATM protein expression in colorectal cancer cell lines was found to be threefold significantly different from that of breast cell line. Decreased ATM expressesion was linked to indigent survival in patients with colorectal cancer.

Predisposition of hereditary nonpolyposis colorectal cancer (HNPCC): MNLH1, MSH2, MSH6, PMS2

More syndromes of tumor predisposition

  • Familial adenomatous polyposis: APC, MUTYH
  • Familial medullary thyroid cancer: MEN, RET
  • Juvenile polyposis: BMPR1A, SMAD4
  • Multiple endocrine neoplasia: MEN, RET
  • Neurofibromatosis type 2: NF2
  • PTEN-associated hamartoma-tumor-syndrome: PTEN
  • Phaeochromocytoma: SDHD, SDHAF2, SDHC, SDHB
  • Paraganglioma: SDHD, SDHAF2, SDHC, SDHB
  • Retinoblastoma: RB1
  • Tuberous sclerosis: TSC1, TSC2
  • Von-Hippel-Lindau syndrome: VHL
  • Wilms tumor: WT1

Table 28.2-4 Oncogene addiction to particular agents /7/






















Head and neck,








Breast, kidney,




Renal cell



Treatment regimen indicates agent alone (monotherapy) or in combination with cytotoxic agents (combination); PDGFR, platelet derived growth factor

Table 28.3-1 Prognostic value (hazard ratio) of circulating tumor cells in breast cancer /14/

Survival (S)

Early stage Ca

Metastic Ca

free (DFS)



free (PFS)



Overall (OS)



Ca, cancer

Table 28.6-1 Clinically important tumor markers





Germ cell tumor
of the testes, liver
cell cancer

Glycoprotein, 70 kDa, carbohydrate fraction 4%

CA 125

Ovarian cancer

Glycoprotein, 200 kDa, carbohydrate fraction 25%, mab OC125

CA 15-3

Breast cancer

Glycoprotein of the milk fat globule membrane mucin family, 300 kDa

CA 19-9

Pancreatic cancer,
biliary tract cancer

Glycolipid, 36 kDa, hapten of Lewis-α blood group determinants

CA 72-4

Gastric cancer,
ovarian cancer

Mucin like glycoprotein TAG 72, 400 kDa


Colorectal cancer,
breast cancer

Glycoprotein, 180 kDa, carbohydrate fraction 45–60%

CYFRA 21-1

Non small cell lung

Cytokeratin 19 fragment, 30 kDa


Germ cell tumor,
trophoblastic tumor

Glycoproteohormone, 2 subunits, α/β-chain, 14/24 kDa



Glycolytic enzyme, isoform of enolase, 87 kDa



Gastrin propeptide


Prostate cancer

Glycoprotein, 33 kDa, kallikrein related serine protease


Squamous cell

Glycoprotein, 42 kDa, tumor antigen-4 isoantigen



Calcium binding protein in the central nervous system, 21 kDa

Table 28.6-2 Indication for tumor marker determination /1/






C-cell tumor

C-cell tumor

Colon, breast, lung, C-cell

Colon, stomach, breast


Risk groups

Germ cell, HCC

Germ cell, HCC

Germ cell

CA 19-9


Pancreas, biliary tract

Stomach, colon

CA 72-4


Stomach, mucinous ovarian


CA 125


Serous ovarian

Serous ovarian

CA 15-3





Small cell lung

Small cell lung, Neuroblastic apudoma



Small cell lung

Small cell lung




Cervix, ENT tumor, esophagus


CYFRA 21-1


Lung NSCLC, bladder



Risk groups

Germ cell

Trophoblastic tumor

Germ cell Trophoblastic tumor

Germ cell

Trophoblastic tumor







C-cell tumor

C-cell tumor

C-cell tumor

C-cell tumor



Differentiation of thyroid cancer




Malignant melanoma

Malignant melanoma

ENT, ear-nose-throat tumor; HCC, hepatocellular carcinoma; NSCLC, non small cell lung cancer

Table 28.6-3 Biological half life* and threshold values of tumor markers


Half life




9 IU/mL

CA 125


35 U/mL

CA 19-9


37 U/mL

CA 15-3


25 U/mL

CA 72-4


4 U/mL



3 μg/L

CYFRA 21-1


2 μg/L



2 IU/L



10 (20) IU/L



4 μg/L



1.5 μg/L

* Biliary and/or renal excretion of tumor markers to half the baseline concentration

Table 28.6-4 Tumor markers in cancers /111, 1213/

Clinical and laboratory findings

Breast cancer

Breast cancer is by far the most common cancer affecting women worldwide, with approximately 1 million new cases diagnosed each year. Although new adjuvant therapy improves patient outcome, 25–30% of women with lymph node negative and 50–60% of those with node positive disease develop recurrent or metastatic disease.

Estrogen receptor (ER) and progesterone receptor (PR): in breast cancer, the determination of ER and PR is recommended for predicting response to hormone therapy and corresponding patient selection. In ER positive patients, 5 years of adjuvant tamoxifen reduced annual breast cancer death rates by 31%. Patients with ER negative tumors, however, do not benefit from adjuvant tamoxifen but are cured of their disease by surgery and radiotherapy.

HER-2 gene: HER-2 determination is recommended in all patients with invasive breast cancer. The primary purpose for determining HER-2 is to select patients who may be treated with trastuzumab.

BRCA1 and BRCA2 genes: BRCA1 and BRCA2 mutation testing may be used for identifying women who are at high familial risk of developing breast or ovarian cancer. For those with such mutations, screening should begin at 25–30 years of age.

CEA, CA 15-3: other serum assays similar to CA 15-3 that detect the same antigen (i.e., MUC-1), include CA 549, MCA, TAG 12, CA m26, CA m29 and CA 27.29. All of these markers may be released to the blood in the setting of breast cancer, but are of no primary diagnostic significance. At the time of breast cancer diagnosis, 15–35% of the patients have CA 15-3 and CEA values above the upper reference interval, with tumor stage related concentrations. Pretherapeutic values of the two markers are independent prognostic parameters for the recurrence free interval and survival.

Baseline values relevant to follow-up are determined 4 weeks after completion of the first phase of therapy (surgery/chemotherapy or radiotherapy) /1/. While the occurrence of distant metastasis is indicated by an increase in CEA and/or CA 15-3 in 70% of cases, this does not apply to loco regional lymph node metastasis or secondary carcinoma. Tumor marker concentrations doubling the individual baseline values almost always indicate the presence of a tumor /1/.

Ovarian cancer (OC)

Approximately 120,000 of the 200,000 women newly diagnosed with OC die from the disease worldwide each year. A critical factor in the elevated mortality associated with OC is the lack of, or failure to detect, disease specific, early stage symptoms. Surface epithelial tumors are histologically divided into two subgroups: type I neoplasms (low grade malignant) and type II tumors (high grade malignant) /14/.

CA 125: as of yet, the most robust serum marker for detection of ovarian cancer is CA 125. Screening for OC is recommended:

  • In women with pelvic tumor, especially post menopausal women, for differentiating malignant from benign tumors
  • In combination with ultrasonography for diagnosing OC in women with hereditary syndromes.

In either case, further clarification is required if CA 125 levels are above 35 U/mL. Women suffering from epithelial OC and showing levels above 35 U/mL have stage I cancer in 50–60%, stage II cancer in 90% and stage III or IV cancer in more than 90% of the cases. CA 125 is useful for assessing the response to chemotherapy. Samples should be analyzed within two weeks before start of therapy, at intervals of 2–4 weeks during therapy and several times at intervals of 2–3 weeks after the end of therapy. Therapy is deemed successful if the CA 125 concentration decreases by 50% during therapy and by 75% in the last sample compared to a sample taken at least 28 days beforehand. The CA 125 concentration also provides prognostic information. Five year mortality is 6.37-fold higher in preoperative concentrations > 65 U/mL than in lower concentrations. Patients with a CA 125 half life < 20 days during or after chemotherapy have a median survival of 28 months compared to a median survival of 19 months in patients with CA 125 half lives > 20 days.

Germ cell tumor of the testes

Germ cell tumor of the testes account for approximately 1% of malignant tumors in men, but are the most common malignant tumors between age 15 and 35. Approximately 95% of malignant testicular tumors originate in the germ cells; the remaining 5% are lymphomas, Leydig or Sertoli cell tumors and mesotheliomas. Germ cell tumors in adolescents and adults are divided into two subtypes: seminomatous (SGTC) and non-seminomatous germ cell cancers of the testes (NSGCT). According to a collaborative study /15/, 65% of the tumors are NSGCT and 36% are seminomas.

Serum tumor markers play an important role in diagnosis, staging, differentiation, prognosis and assessment of response to therapy and recurrence. Proven markers include AFP, hCG and the enzyme LD. In the setting of NSGCT, at least one of these markers is elevated.

Diagnosis of germ cell tumors of the testes: according to a collaborative study /15/, when presenting,

  • 77% of seminoma patients have stage I cancer and 21% have elevated hCG. hCG level is usually below 300 U/L, while concentrations above 1,000 U/L are in most cases associated with NSGCT. LD is elevated in 40–60% of the cases.
  • 52% of NSGCT patients have stage I cancer and elevated serum marker levels (hCG and AFP in 44%, only AFP in 26% and only hCG in 9% of cases) LD is elevated in 40–60% of the cases.

Prognosis: the markers allow classification into three subgroups with different prognosis:

  • Good: LD < 1.5- fold the upper reference interval value, AFP < 1,000 μg/L, hCG < 5,000 U/L
  • Medium: LD 1.5–10 fold the upper reference interval value, AFP 1,000–10,000 μg/L, hCG 5,000–50,000 U/L
  • Poor: LD > 10 fold the upper reference interval value, AFP > 10,000 μg/L, hCG > 50,000 U/L.

Orchidectomy: cures stage I seminoma in 80–85% and NSGCT in 70% of the cases.

Chemotherapy: the half life of the tumor markers after the first two cycles of chemotherapy is a prognostic factor for outcome. The half lives of hCG and AFP are 1.5 days and 5 days, respectively. Half lives above 3.5 days for hCG and above 7 days for AFP indicate a poor prognosis.

Follow-up: after curative therapy, determination in low risk patients post surgery should be performed every 1–2 weeks within the first 6 months. Recurrence often occurs within the first year following treatment and rarely after the second year.

Colorectal carcinoma (CRC)

CRC is the third most common cancer worldwide, with an estimated incidence of 1 million cases and an estimated 0.5 million deaths globally per year. Approximately 40–50% of the patients develop metastases despite potentially curative surgery.

Diagnostics: CEA is not recommended for the screening of healthy individuals due to lack of diagnostic sensitivity and specificity. At the time of primary diagnosis of CRC, CEA is elevated in up to 20% of stage Dukes A cases, 20–40% of Dukes B cases, 40–60% of Dukes C cases and 60–95% of Dukes D cases. The preoperative CEA level can be used in combination with other findings for surgical therapy planning.

Prognosis: if preoperative CEA levels are above 5 μg/L, diagnostic investigation for the presence of distant metastasis should be performed.

Follow-up: in patients with stage II or III CRC, the CEA level should be determined every three months post surgery for at least three years after diagnosis. An increase by at least 30% compared to the previous value during the CRC follow-up period is considered to be an elevated CEA concentration (no clinically validated statement). However, an increase of this nature must be confirmed within one month. Monthly increases by 15–20% for at least three months also warrant intervention. CEA should be determined every 1–3 months in patients with locally progressive or metastatic CRC receiving systemic therapy. An increase in CEA by more than 30% compared to the individual baseline confirms the progressive nature of the disease. The routine determination of tumor markers other than CEA is not recommended in CRC. Patients receiving intensive follow-up after curative surgery have a 5-year survival advantage.

Pancreatic cancer

In pancreatic cancer, the tumor doubling time is 0.5–3.5 months. Tumor marker determination is recommended in all patients above 45 years of age suffering from epigastralgia for 2–3 weeks. CA 19-9 is the marker of first choice. Its diagnostic sensitivity for pancreatic cancer is 70–80%, independently of the degree of differentiation. There is no correlation between marker concentration and tumor mass. In the setting of pancreatic cancer, levels above 1,000 U/mL indicate involvement of the lymph nodes and values above 10,000 U/mL indicate hematogenous spread.

Patients with a resectable tumor have lower CA 19-9 concentrations than those with non resectable tumors. Within a patient group with tumors of the same stage, median survival time is markedly longer in patients with lower CA 19-9 levels. Following completion of first line therapy, the diagnostic sensitivity of CA 19-9 for the detection of progression is 80–90%.

Hepatocellular carcinoma (HCC)

Besides the more sensitive imaging techniques, alpha fetoprotein (AFP) is the marker of choice among tumor markers in HCC. Small lesions (less than 1–2 cm in diameter) are usually AFP negative. At first diagnosis of HCC with imaging procedures, 60–70% of patients have elevated AFP levels (30% > 1,000 μg/L, 20% > 10,000 μg/L).

In space occupying hepatic lesions, AFP is diagnostically significant at levels above 1,000 μg/L. It is recommended in patients with liver fibrosis/cirrhosis to perform AFP determination every three months for early detection of HCC.

Gastric cancer

CA 72-4 is recommended for monitoring and follow-up because its diagnostic sensitivity and specificity are higher than those of CEA and CA 19-9. However, additional determination of these two markers achieves a markedly higher sensitivity. Patients with a CA 72-4 concentration above 6 μg/L at primary diagnosis of gastric cancer are at 4.2-fold higher risk of dying from the disease than those with lower concentrations.

In patients testing positive for all three markers at primary diagnosis, one of the markers will indicate post therapy recurrence in 100% of the cases.

Lung cancer

The incidence of lung cancer is 1.35 million cases with 1.18 million deaths worldwide per year. The high mortality is due to the fact that 80% of the tumors are not diagnosed until the cancer has spread locally or systemically. Lung cancer is histologically classified into:

  • Small cell lung cancer (SCLC), with a proportion of 20–25%
  • Non small cell lung cancer (NSCLC). This term also includes squamous cell cancer, adenocarcinoma and large cell carcinoma, which represent all other lung cancers. Only 20% of NSCLC patients at stage III and only 5% of NSCLC patients at stage IV have a survival time of 1 year or longer.

Diagnostics: the tumor markers NSE, CEA, SCCA, CYFRA 21-1 and ProGRP are recommended depending on the clinical situation. In the presence of pulmonary nodules of unknown origin, elevated concentrations of ProGRP, as well as NSE, point to SCLC and rule out NSCLC. High SCCA concentrations point to NSCLC (squamous cell cancer). CYFRA 21-1 and CEA are released in lung cancers of all histological types /16/.

Post-surgery: concentrations declining according to their physiological half life indicate successful intervention. More slowly declining concentrations and/or failure to normalize indicate that surgery was not successful.

Follow-up: the following markers have a 70–85% sensitivity and almost 100% specificity in the detection of recurrence: CYFRA 21-1 in NSCLC, NSE and ProGRP in SCLC, and SCCA in squamous cell cancer. Elevated concentrations of these markers indicating the presence of recurrence can precede clinical symptoms and detection with imaging techniques by 2–15 months.

Prostate cancer

Prostate cancer is the second most common cancer in men, but only accounts for 5.85% of cancer mortality. Optimal management of prostate cancer requires PSA determination in all stages of the disease.

Table 28.7-1 Elevated serum AFP concentrations

Clinical and laboratory findings

Physiological conditions

Children less than 1 year of age: mean AFP level at birth approximately 70 mg/L, 0.5–4 mg/L during the 2nd to 3rd week of life, < 20 μg/L after the 10th month of life. Elevated values deviating from those mentioned above are observed in liver and biliary disorders during early childhood.

Pregnant women: rise in AFP level, dependent on the gestational age of the pregnancy from the 10th gestational week; maximum values of 400–500 μg/L between the 32nd and 36th gestational week followed by a moderate decline until delivery (down to a level of maximally 250 μg/L). After delivery, AFP declines with a physiological half life of about 4 days. Elevated levels deviating from these are observed in conjunction with fetal neural tube defects, fetal distress syndrome and fetal intrauterine death.

Benign liver disease

Acute viral hepatitis, alcoholic hepatitis, chronic hepatitis, liver cirrhosis. While acute diseases are more likely to be associated with transient AFP elevations, chronic diseases often show constantly low abnormal AFP levels, usually < 500 μg/L, and rarely (1%) higher ones.

Malignant disease

Gastrointestinal and other tumors: elevated AFP concentrations in up to 21% of gastric cancer, colorectal cancer, biliary cancer, pancreatic cancer, lung carcinoma; often seen in conjunction with liver metastases. AFP values are usually < 500 μg/L, and rarely (4%) greater than this.

Hepatocellular carcinoma (HCC): at the time of diagnosis, normal AFP levels in about 40% of cases and abnormal elevations in 60%; levels found within the latter category: > 100 μg/L in 50%, > 1 mg/L in 32% and > 10 mg/L in 20% (in contrast to cholangiocellular carcinoma, as part of the differential diagnosis: normal AFP values). Extremely elevated AFP levels of up to 10 g/L are possible. There is no correlation between the AFP concentration and the size, growth and stage of the tumor nor the degree of malignancy. AFP levels > 2 mg/L almost always indicate the existence of HCC.

Germ cell tumors (testes, ovary, extra gonadal location): the prevalence of abnormal AFP concentration depends on the tumor type. The elevations tend to be lower than in hepatocellular carcinoma. Pure seminomas, dysgerminomas, dermoid cysts and mature teratomas as well as pure choriocarcinomas are always AFP negative, whereas yolk sac tumors (entodermal sinus tumors) are always AFP positive. Embryonic carcinomas show different AFP expression, with MTUs presenting AFP positivity in 70%, MTIs in up to 50% and combination tumors at a rate somewhere in-between those two. hCG is the second obligatory tumor marker.

Table 28.7-2 AFP concentrations during therapy

Clinical and laboratory findings

Decline in AFP concentration

Post-surgery: a relatively rapid decrease in AFP, associated with a rate of decline of < 5 days (physiological half life), indicates complete tumor removal. In the case of significantly slower rates of decline: impaired catabolism? Concomitant liver disease? A small residual tumor? A course monitoring is indicated.

Radiotherapy/chemotherapy: a rapid decline in AFP back to normal values indicates complete elimination of the tumor marker producing cells which may not include the entire tumor. In tumors with mixed-cell composition, the tumor cells of other types may show concordant or discordant patterns of AFP decline. Therefore, other tumor markers (e.g., hCG) should also be determined. Beware of the pattern of tumor cell types not producing tumor markers and/or changes in the cell type during therapy. Negative tumor marker results do not rule out progressive disease.

Persistence and/or further increase in AFP concentration

Post surgery or after/during radiotherapy or chemotherapy: the determination of AFP is indicated if residual tumor and/or metastases are suspected. A progressive rise in the concentration of AFP during therapy suggests that the tumor is not responding.

Recurrent increase in AFP concentration

Post-surgery and/or after/during radiotherapy or chemotherapy: the determination of AFP is indicated if tumor recurrence and/or metastases are suspected. After the first determination of minimal AFP elevations, further monitoring is required (first check-up after an interval of at least 14 days). A rise in the AFP concentration can indicate tumor recurrence and/or metastases weeks to months prior the detection by other methods (lead time 1–6 months).

Table 28.8-1 Diagnostic sensitivities of CA 19-9 in benign and malignant diseases




Benign diseases

Many benign diseases are not associated with a CA 19-9 increase, except for acute and active diseases of the liver-biliary-pancreatic system. In these cases, in 10–30%, transitory increases occur depending on the activity and extent of the disease. This increase is usually below 100 U/mL, rarely very high (cholestasis) with normalization in the case of clinical improvement. Monitoring is necessary at a minimum interval of two weeks.





Cholecystitis, obstructive jaundice


Toxic hepatitis/CAH


Liver cirrhosis


Liver cell necrosis




  • chronic active


  • chronic inactive


Malignant diseases

In tumor induced CA 19-9 elevations (in case of no treatment), exponential increases to > 1,000, maximally 100,000 U/mL are often observed.

The most important group are patients with excretory pancreatic cancer; 72–90% diagnostic specificity, the sensitivity depends on tumor stage and tumor extent; furthermore, patients with hepatobiliary cancer and gastric cancer. After complete tumor resection, normalization occurs after 2–4 weeks; tumor recurrence or metastases are often indicated early on by renewed or further increases (1–6 months). Increasing CA 19-9 levels indicate a non response.

Ductal pancreatic carcinoma


Hepatocellular carcinoma


Biliary cancer


Gastric cancer


  • Stages I–IV


  • Colorectal cancer


  • Dukes A–D


Gastrointestinal cancer + liver metastases


Lung cancer


Breast cancer


Ovarian cancer


  • Mucinous type


Uterine cancer


* Threshold level 37–40 U/mL

Table 28.9-1 Diagnostic sensitivities of CA 125 in benign and malignant diseases


sensitivity* (%)


> 35*

> 65*

Benign disease

Mild CA 125 elevations occur only in rare cases (adnexa, liver, pancreas); normalization along with clinical improvement. Monitoring at a time interval of at least 2 weeks.

In general



(Acute) adnexitis









Acute pancreatitis




Acute hepatitis


Chronic liver disease


Renal insufficiency

Adnexal tumors


Malignant disease

CA 125 is the most important tumor marker in ovarian cancer with a diagnostic specificity of 99% for healthy people, 83% for adnexal diseases and 92% for benign ovarian tumors. In 90% of ovarian cancer patients, CA 125 correlates well with the disease course; lead time: 1–8 months. In the case of complete tumor resection, normalization within 2–3 weeks. During remission, only 1–4% of patients have elevated values while in progressive disease the rate of elevated levels is 80%. Normal levels do not rule out a small residual tumor (< 1 cm), levels > 65 U/mL suggest a residual tumor > 2 cm. Increasing CA 125 levels indicate a non response.

Metastatic ovarian



Primary ovarian cancer



  • Epithelial type



  • Serous type



  • Undifferentiated



  • Endometrioid type



  • Mucinous type



Breast cancer



Cervical cancer



Uterine cancer



Table 28.10-1 Diagnostic sensitivity of CA 72-4 in benign and malignant disease


sensitivity %*


Benign disease

Benign diseases are more likely to be associated with mild or transient increases in marker concentration with normalization after remission.

In general


Liver cirrhosis






Rheumatic disease


Gynecological disease


Ovarian disease


Ovarian cyst


Malignant disease

Malignant diseases without treatment are associated with gradual to exponential increases correlated with the tumor mass, tumor stage and localization of metastases. CA 72-4 is not useful for screening or for diagnostic purposes but valuable for monitoring the treatment outcome and the course of metastatic gastric cancer (first line marker; CEA or CA 19-9 as a second line marker).

Esophageal cancer


Pancreatic cancer


Biliary cancer


Gastric cancer


  • Stage I


  • Stage II


  • Stage III


  • Stage IV


Tumor recurrence


Colon cancer


  • Dukes A


  • Dukes B


  • Dukes C


  • Dukes D


  • Tumor


Breast cancer


Ovarian cancer


  • Serous


  • Mucinous


Cervical cancer


Endometrial cancer




* Threshold > 3–4 U/mL: NED, no evidence of residual disease

Table 28.11-1 Diagnostic sensitivity of CA 15-3 in benign and malignant diseases


sensitivity* (%)


Benign disease


In general


Renal insufficiency**








  • Mastopathy


  • Fibroadenoma


Malignant disease

Malignant diseases without treatment are associated with gradual to exponential increases correlated with the tumor mass, tumor stage and localization of metastases. CA 15-3 is not useful for screening or for diagnostic purposes but valuable for monitoring the treatment outcome and the course of metastatic breast cancer (first-line marker) in addition to its use in ovarian cancer (as a second line marker after CA 125). Increasing CA 15-3 levels indicate a non-response.

Breast cancer

  • Preoperative


  • M0


  • Node negative


  • Node positive


  • Metastases


  • NED


  • Progressive disease

Up to 100

  • Stage I


  • Stage II


  • Stage III


  • Stage IV


  • T ½


  • T ¾


  • Skin metastases


  • Connective tissues


  • Bone metastases


  • Liver metastases


Ovarian cancer


Endometrial cancer


Uterine cancer


Lung cancer


Gastric cancer


Pancreatic cancer


Liver cancer


* Threshold level 25–50 U/mL; ** dialysis-dependent

NED, no evidence of residual disease

Table 28.12-1 Thresholds below which a medullary thyroid carcinoma is small /256/

Basal CT value in adults: below 10 ng/L in the German cohort and below 15 ng/L in the US cohort /3/

Assay- and sex-specific cutoffs are important.

Basal CT value in children less than 6 months of age: below 40 ng/L /7/

Children less than 3 years of age: below 15 ng/L /7/

Pentagastrin stimulation test: below 30 ng/L /4/

Table 28.12-2 Elevated calcitonin (CT) without the presence of medullary thyroid cancer (MTC) /12/

Clinical and laboratory findings

Chronic alcoholism

Even after several weeks of alcohol abstention, chronic alcoholics may have CT levels within the gray zone without the presence of MTC.

Smoking /6/

Healthy cigarette smokers have CT concentrations above 10 ng/L more often than non smokers.

Oral calcium intake

In oral calcium intake, CT levels are elevated for approximately 2 h. The elevation is more pronounced in patients with resorptive hypercalciuria than in healthy individuals.

Proton pump inhibitors (PPI)

CT values are elevated in 6–32% of patients taking omeprazole, pantozol, lanzoprazole or other PPIs, depending on the test.

Salmon calcitonin

Treatment with salmon calcitonin can cause detectable or elevated values, depending on the test.

Systemic bacterial infection (sepsis)

Elevated CT concentrations may be obtained because procalcitonin is determined to a varying degree, depending on the test.

Chronic kidney disease (CKD)

Increased CT concentrations are found in 9.8–24.4% of patients with CKD stages 2–4 and in 1.6–71.4% in CKD stage 5, depending on the test and the applicable cutoff value /2/.

Hashimoto’s thyroiditis

Slightly increased CT concentrations are found in 1–3% of cases depending on the test.

C-cell hyperplasia

C-cell hyperplasia is more common in men than in women. In many cases, it is associated with autoimmune thyroiditis, hypergastrinemia or hyperparathyroidism. CT concentration is within the gray zone or higher.

Table 28.12-3 Serum calcitonin (CT) in medullary thyroid carcinoma (MTC)

Clinical and laboratory findings


The negative predictive value of a CT concentration below 10 ng/L or below the test manufacturer’s test- and gender-specific cutoff value is high and MTC can be ruled out at high probability. CT levels below the cutoff value have been reported in micro carcinomas (less than 1 cm in diameter) and in extensive MTC /10/. However, CT levels in part of the patients with hereditary MTC are below the cutoff level and those in part of examined patients without MTC are within the gray area. If other endocrine tumors and biologic are excluded clinically and the biological influence factors are ruled out, MTC and C-cell hyperplasia are the main differential diagnoses in CT concentrations between 10 and 100 ng/L /1/. According to a study /14/, the positive predictive value for the presence of MTC in patients with thyroid nodular disease is 8% at basal CT concentrations of 20–50 ng/L, 25% at concentrations of 51–100 ng/L and 100% at concentrations above 100 ng/L. If CT concentrations are higher in the presence of the corresponding clinical symptoms, total thyroidectomy is recommended.

If neoplasia does not extend beyond the thyroid, chances of complete cure by total thyroidectomy and removal of the central lymphatic compartments are high. Therefore, early diagnosis of MTC is important. MTC is more aggressive and more difficult to treat than papillary and medullary thyroid carcinomas. Screening for MTC in individuals with thyroid nodules is recommended in some countries.

Pentagastrin stimulation test

If basal CT levels are 10–100 ng/L, CT should be analyzed by pentagastrin stimulation testing. In patients with stimulated CT values > 100 ng/L, thyroidectomy is advised in these individuals /12/. Concentrations above 100 ng/L suggest the presence of both MTC and C-cell hyperplasia, with an overlap occurring especially in the range of 100–200 ng/L /13/. The overlap is more pronounced in men than in women. For instance, the postoperative result in patients with concentrations > 100 ng/L showed that 82% of the males harboured C-cell hyperplasia and 80% of the females were identified with MTC /14/. The positive predictive value in the diagnosis of MTC is always 100% for CT concentrations > 1,000 ng/L /11/.

Postoperative calcitonin levels

CT decreases pronouncedly within a few hours postoperatively. The patient is considered to be cured if the concentration declines below the cutoff value or the detection limit and the pentagastrin stimulation test is negative. In this case, basal concentration should be monitored every 6 months and a pentagastrin stimulation test should only be performed every 2–5 years.

Persistently elevated postoperative CT concentrations suggest tumor persistence or the presence of metastases. There is a correlation between CT concentration and tumor mass. In CT concentrations < 1,000 ng/L, a definite tumor correlate cannot be determined with conventional imaging procedures. Selective venous catheterization for serum sampling along with serum CT determination has proven itself in these cases. Based on a concentration gradient, this method allows the localization of tumor tissue in the neck and/or upper mediastinum as well as the detection of liver metastasis /15/.

Molecular genetic analysis

Molecular genetic analysis is performed:

  • Primarily in all individuals with a corresponding family history because most of these individuals have CT concentrations below the cutoff value and are not detected by screening. Since inheritance is autosomal dominant, 50% of the family members are not genetically affected and do not require regular biochemical examinations following molecular genetic analysis. If gene carriers are identified, there are two options: either prophylactic thyroidectomy is performed preferably around 6 years of age, independently of the pentagastrin test result, or pentagastrin tests are performed every year and a decision for surgery is made if the results are pathological /16/.
  • If MTC is detected. In this case, it is important to differentiate between the sporadic and the hereditary forms. The patient is examined for one of 14 point mutations in the RET proto oncogene. Mutations in codons 634, 768, 790, 791, 804 and 891 have been detected to date. The mutation in codon 634 is frequently found in Europe. It usually becomes clinically manifest in patients younger than 10 years of age, and not older than 20 years of age /17/. The other mutations are associated with an older age.

Molecular genetic analysis also shows the limits of the pentagastrin test /18/. Both false negative and false positive tests have been observed. Some of the children on which thyroidectomy was performed solely based on the typical mutation detected in the RET proto oncogene had a micro carcinoma although the pentagastrin test result was false negative. On the other hand, MEN 2 family members where no mutation was detected by molecular genetic analysis were submitted to surgery because of an abnormal pentagastrin test result (false positive pentagastrin test). Histologically, C-cell hyperplasia, which is not a preliminary stage of C-cell carcinoma, gives false positive pentagastrin test results. It can, for example,occur in the presence of Hashimoto’s thyroiditis and is detected when CT is routinely determined in all cases of cold thyroid nodules and ­a pentagastrin test is performed /19/.

Approximately 3–5% of all seemingly sporadic MTCs are classified as familial based on the detection of germ line RET mutations /16/.

In the presence of elevated CT, there are two indications for a presurgical molecular genetic analysis /1/: extent of resection (no C-cells to remain in the residual thyroid) and detection of C-cell carcinoma with precancerous lesions.

Table 28.13-1 Serum CEA concentration in non malignant and malignant disease

Clinical and laboratory findings

Non malignant disease

CEA elevations are observed most frequently in inflammatory liver disease. In active alcoholic liver cirrhosis, diagnostic specificity of CEA determination can be 30%. Pancreatitis, inflammatory bowel disease (e.g., ulcerative colitis, diverticulitis) and inflammatory lung disease can be associated with CEA elevations. In general, levels do not exceed 4-fold the upper reference interval value.

Malignant disease

The presence of a malignant tumor is probable when CEA concentration exceeds the 4-fold the upper reference interval value. If concentration increases during monitoring or CEA level is above a concentration corresponding to 8-fold the upper reference interval value, malignant disease is present with high probability.

Colorectal cancer

Because of the limited diagnostic sensitivity and specificity of CEA and considering the incidence of colorectal cancer, the determination of CEA is not suitable for screening purposes. In colorectal cancer, the rates of CEA elevations according to tumor stage are as follows: Dukes A: 0–20%; Dukes B: 40–60%; Dukes C: 60–80%; Dukes D: 80–85%. Concentrations exceeding 4-fold the upper reference interval value are not compatible with a Dukes A stage. Concentrations in this range are observed in 15–20% of Dukes stages B and C and in 60–70% of Dukes stage D.

Postoperative monitoring: if preoperatively elevated CEA concentrations do not reach a stable level immediately after tumor resection and increase subsequently, residual tumor is present. If CEA determinations are applied to the diagnosis of local tumor recurrence and/or other metastatic sites in postoperative monitoring, determinations should be performed during the first two years every 2–3 months irrespective of whether the preoperative CEA level was elevated or not. If an increase is suspected, CEA should be measured at shorter time intervals. For the diagnosis of tumor progression, the positive predictive value of an increase in CEA is in the range of 65–84%, the negative predictive value is in the range of 85–90%.

Table 28.14-1 Use of CYFRA in combination with other tumor markers in lung cancer /3/


Before treatment



Cyfra 21-1

Post surgery:
histology dependent


Without surgery:
dependent on the
tumor marker results

Small cell

Cyfra 21-1

NSE and Cyfra 21-1

Non small cell carcinoma (NSCLC)


Cyfra 21-1,

CYFRA 21-1
and/or CEA

Squamous cell

Cyfra 21-1

Cyfra 21-1

Large cell

Cyfra 21-1,

Cyfra 21-1
and/or CEA

Table 28.14-2 Diagnostic sensitivity of Cyfra 21-1 in malignant disease /3/

Maligne organ disease

sensitivity (%)*

Ears-nose-throat (ENT)












Ovary, serous type


Ovary, mucinous type



37, 41



Bladder, superficial type


Bladder, muscle, invasive type


Lung, overall


Lung, SCLC




  • Squamous cell carcinoma


  • Adenocarcinoma


  • Large cell type


* Decision level at 95% diagnostic specificity in comparison to the diagnostically relevant reference cohort.

Table 28.14-3 Cyfra 21-1 cutoffs in healthy individuals and in benign disease /2/


Cutoff (μg/L)

Healthy individuals


Patients with pulmonary disease


Patients with gastrointestinal disease


Patients with gynecological disease


Patients with urological disease


Patients with renal insufficiency


Cutoffs expressed as 95th percentiles

Table 28.14-4 Cyfra 21-1 in benign diseases

Clinical and laboratory findings

Pulmonary disease

Approximately 95% of patients with benign lung tumors, chronic bronchitis, COPD, bronchial asthma, pneumonia, sarcoidosis, tuberculosis and emphysema have Cyfra 21-1 concentrations below the defined cutoff value of 3.3 μg/L /7/.

Gastrointestinal disease

Cyfra 21-1 concentrations below 3 μg/L are found in 80% of patients with acute and chronic hepatitis, liver cirrhosis, pancreatitis, cholangitis, gastritis, Crohn’s disease, ulcerative colitis and colon polyps. In some cases, levels can be higher and the cutoff value is 6.9 μg/L. In cholestatic disease, no cholestasis induced elevations occur /7/.

Gynecological disease

Postulating a diagnostic specificity of 95%, the cutoff value is 3.1 μg/L for the following malignant diseases: endometriosis, ovarian cysts, adnexitis, benign ovarian tumors, benign urological diseases such as urinary tract infection, renal cysts, renal and ureteric stones and benign tumors of the bladder /6/.

Renal insufficiency

Patients with renal insufficiency have elevated mean Cyfra 21-1 levels, independently of their dialysis requirement and stage of renal insufficiency. Only 67% have concentrations below 3 μg/L; concentrations up to 10 μg/L are found sporadically /8/.

Table 28.14-5 Cyfra 21-1 in non small cell lung cancer (NSCLC)

Clinical and laboratory findings

Primary diagnosis of NSCLC – Generally

The diagnostic sensitivity is 40–64%. The marker correlates well with the depth of tumor invasion (T1, 15%; T2, 49%; T3, 68%; T4, 55%) as well as the tumor stage (I, 29%; II, 56%; III, 63%; IV, 63%) /6/.

– Pulmonary round lesion of uncertain etiology /3/

In some cases, for various reasons, no histological data can be obtained on pulmonary round lesions of uncertain etiology. The combined determination of Cyfra 21-1 as a first line marker in NSCLC and of NSE as the leading parameter in SCLC may be helpful and serve as a diagnostic guideline.

The presence of benign pulmonary disease is very unlikely given a Cyfra level > 10 μg/L and an NSE concentration > 20 μg/L. For instance, pulmonary metastases of other primary tumors (e.g., cancer of the colon, breast, stomach and testes) are associated with relatively low Cyfra 21-1 and NSE concentrations (both < 30 μg/L). A pulmonary round lesion of uncertain etiology in conjunction with Cyfra 21-1 levels > 30 μg/L therefore suggests with high probability the presence of primary lung cancer, although no distinction can be made between NSCLC and SCLC. If, however, NSE levels are > 70 μg/L, the presence of SCLC is very likely. The clinical significance of CYFRA 21-1 compared to other markers for differentiation between lung cancer and benign pulmonary disease is shown in Fig. 28.14-1 – Comparison of tumor markers for the differentiation between SCLC and benign lung disease.

– Squamous cell cancer

The diagnostic sensitivity of Cyfra 21-1 is 52–79%, while that of SCC and CEA is only 30% and 20%, respectively. None of the possible tumor marker combinations provides any significant additive increases in diagnostic sensitivity without a substantial loss of specificity.

– Adenocarcinoma, large cell carcinoma

The diagnostic sensitivity is 42–54% in adenocarcinoma and 44–65% in large cell carcinoma. CYFRA 21-1 is the leading marker. In these tumors, a marked additive increase of approximately 10% results from combination with CEA.

Prognosis in NSCLC

In NSCLC, Cyfra 21-1 is a good prognostic marker for tumor stages I-IIIA versus IIIB; this does not apply to CEA. The relative mortality risks are 2.1 in Cyfra 21-1 concentrations > 3.3 μg/L versus lower concentrations, and for stages IIIB, IV versus I–IIIA.

The 2-year survival rate in patients with the best prognosis (stages I–IIIA and CYFRA 21-1 < 3.3 μg/L) is 60%, whereas in patients with tumor stages IIIB, IV and Cyfra 21-1 levels > 3.3 μg/L it is less than 10% /13/.

Course monitoring of NSCLC /2/

Cyfra 21-1 has proven to be a sensitive and specific marker in the control of therapy and in monitoring patients with lung cancer. Since the half life is short and concentrations decrease considerably within a few days after curative therapy, estimates of the effectiveness of therapy can be performed soon (approximately 48 h) after first treatment (surgery).

According to a follow-up study on the monitoring of NSCLC, all patients with positive pretreatment Cyfra 21-1 levels showed Cyfra 21-1 expression again at the time of tumor recurrence or up to 15 months before /14/. Approximately 50% of patients who originally had been Cyfra 21-1 negative became positive at the time of tumor recurrence.

Following curative surgery (R0 resection) of a tumor which originally expressed cytokeratin-19 fragments, costly invasive diagnostic procedures could, therefore, be avoided during the follow-up period until an increase in Cyfra 21-1 indicates progressive primary disease.

Table 28.14-6 Cyfra 21-1 in the diagnosis of urinary bladder cancer /15/


CYFRA-21-1 (μg/L)

Positive rate (%)

No cancer or early
invasive cancer

1.43 ± 0.75


Muscle invasive cancer

2.14 ± 2.57


Lymph node positive/
metastatic cancer

18.6 ± 17.7


Values expressed as x ± s

Table 28.16-1 Diseases and conditions predominantly associated with specific hCG forms /2/


hCG forms


Intact hCG with various degrees of glycosylation. However, during the first several weeks of pregnancy, the β-core fragment (hCGβcf) is the predominant form of hCG and appears in the urine. Therefore, it must be possible to detect hCGβcf with urine tests.

trophoblastic disease

Intact hCG, hCGβ, nicked hCG and nicked hCGβ, hyperglycosylated hCG.

Germ cell tumors of
the testes

Predominantly hCGβ and intact hCG. A third of patients with non trophoblastic neoplasms secrete only hCGβ.

Trisomy 21


Table 28.16-2 Reference interval for hCG /56/



Total hCG (β-hCG)

Intact hCG

< 5 IU/L (15 pmol/L)

  • Premenopausal
    < 5 IU/L (15 pmol/L)


Total hCG (β-hCG)

< 0.5 IU/L (2 pmol/L)

< 5.05 IU/L (17 pmol/L)

  • Postmenopausal
    < 10 IU/L (30 pmol/L)

Table 28.16-3 Prevalence (%) of increased β-hCG and hCGβ in malignant disease /8/

Non seminomatous testicular
germ cell tumor




Testicular or placental


Hydatidiform mole


Tumors of the small intestine


Colon cancer




Lung cancer


Pancreatic cancer

  • Adenocarcinoma


  • Islet-cell carcinoma


Gastric cancer


Ovarian cancer, epithelial


Breast cancer


Renal carcinoma


Table 28.16-4 Prevalence of testicular tumor subtypes* /9/


Prevalence (%)



Embryonal carcinoma






Yolk sac tumor




Embryonal carcinoma + seminoma


Teratocarcinoma + seminoma


Choriocarcinoma + seminoma


Mixed teratoma + seminoma


* British testicular tumor panel

Table 28.16-5 Histological classification of germ cell tumors according to Ref. /11/

Tumor type

β-hCG and


Carcinoma in situ




  • Embryonal carcinoma



  • Yolk sac tumor (5%)


  • Polyembryoma (30%)



  • Trophoblastic tumor
    (choriocarcinoma, placental
    trophoblastic tumor)



Teratoma (5%)

  • Mature teratoma

  • Dermoid cyst

  • Immature teratoma

  • Teratoma, malignant



Mixed forms**



* Seminomas: classical seminoma, seminoma with high mitotic index, seminoma with syncytiotrophoblast, spermatocytic seminoma

** Teratocarcinoma (embryonal carcinoma and teratoma), choriocarcinoma in combination with other germ cell tumors, other combinations such as mixed germinal tumor, germinoma (germ cell tumor)

Table 28.16-6 Use of serum tumor markers in adult males with germ cell tumors (GCTs)* 

Clinical and laboratory findings


The screening of asymptomatic male adults is not recommended because there is no evidence to support screening for GCTs with any blood test.


The Guidelines recommend drawing blood to measure serum AFP and hCG before orchiectomy for all patients suspected of having a testicular GCT to help establish the diagnosis and interpret post orchiectomy levels. However, the Panel recommends against use of serum tumor marker assay results to guide decision making on need for an orchiectomy. Concentrations in the normal range do not rule out testicular neoplasms or the need for diagnostic orchiectomy.

Cancers of unknown primary (CUP)

The Guidelines recommend against using serum tumor marker results to guide treatment of patients with CUP and indeterminate histology, because evidence is lacking to support this use. Consider treatment with a chemotherapy regimen for disseminated GCT in patients presenting with undifferentiated midline carcinoma even if serum hCG and AFP concentrations are within normal ranges.

Non seminomatous germ cell tumors (NSGCT) – Staging and prognosis before chemotherapy and/or additional surgery

The Guidelines recommend measuring serum AFP, hCG, and LD after orchiectomy and before any subsequent treatment for all patients with testicular NSGCT shortly after orchiectomy and before any subsequent treatment. The magnitude of post orchiectomy serum tumor marker elevations is used to stratify risk and select treatment but must be interpreted appropriately. Serial serum tumor marker measurements may be needed to determine whether serum tumor marker levels are rising or falling and, if falling whether the decline approximates the marker’s biological half life.

The Guidelines recommend measuring serum AFP and hCG shortly before retroperitoneal lymph node dissection (RPLND) and before chemotherapy begins for patients with mediastinal or retroperitoneal NSGCT to stratify risk and guide treatment.

The Guidelines recommend measuring serum AFP and hCG shortly before RPLND in patients with clinical stage I or II NSGCT; those with rising concentrations are beyond stages IA or IB and require systemic therapy similar to the regimens used for patients with stage III disease.

The Guidelines recommend measuring serum AFP and hCG immediately prior to chemotherapy for stage II/III testicular NSGCT. The magnitude of marker elevations guides chemotherapy regimen choice and treatment duration.

– Monitoring response or progression soon after therapy

The Guidelines recommend measuring AFP and hCG at the start of each chemotherapy cycle and again when chemotherapy concludes. However, the Panel sees no indication to delay the start of chemotherapy until after results of serum tumor marker assays are known. Rising AFP and/or hCG levels during chemotherapy usually imply progressive disease and the need to change the regimen. However, tumor lysis from chemotherapy, particularly during first cycle, may result in transient spike in serum tumor marker levels, and such a spike does not represent treatment failure. Resect all residual disease for patients whose serum tumor marker levels have normalized and who have resectable residual mass following chemotherapy. Slow decline during treatment conveys higher risk of treatment failure but does not indicate need to change therapy. Persistently elevated but slow declining post chemotherapy levels do not indicate immediate need for additional chemotherapy; resection of residual masses need not be delayed until serum tumor marker levels normalize.

– After presumably definitive therapy

The Guidelines recommend measuring AFP and hCG at each visit during surveillance after definitive therapy for NSGCT, regardless of stage using intervals within the range used by the available uncontrolled series: every 1 to 2 months in the first year, every 2 to 4 months in the second year, every 3 to 6 months in the third and fourth years, every 6 months in the fifth year, and annually thereafter. The Panel also recommends that surveillance should continue for at last 10 years after therapy is completed.

Seminoma – For staging and prognosis before retroperitoneal lymph node dissection (RPLND), radiation or chemotherapy

Although direct evidence is lacking to determine whether measuring serum tumor marker concentrations improves survival or other health outcomes of these patients, the Panel recommends measuring pre orchiectomy serum concentrations of hCG and/or LD for patients with testicular pure seminoma and pre orchiectomy elevations. However, the Panel recommends against using post orchiectomy concentrations of either hCG or LD to stage or predict prognosis of patients with involved nodes and/or metastasis.

– To predict response to or benefit from treatment

The Guidelines recommend against tumor marker levels to guide treatment decisions for seminoma. Evidence is lacking that selecting therapy based on tumor marker levels yields better outcomes.

– To monitor response or progression during or soon after therapy

The Guidelines recommend against using tumor markers to monitor response or progression of seminomas during treatment. However, serum hCG and AFP should be measured when seminoma treatment concludes. Rising concentrations usually indicate progressive disease and the need for salvage therapy (usually chemotherapy).

– After presumably definitive therapy

The Guidelines recommend against using intervals within the range in the available uncontrolled series: every 2 to 4 months in the first year, every 3 to 4 months in the second year, every 4 to 6 months in the third and fourth years, and annually thereafter. The surveillance should continue for at last 10 years after therapy is completed.

Table 28.16-7 Behavior of serum markers in testicular cancer /8/

Clinical and laboratory findings


Seminomas account for 30–50% of testicular germ cell tumors. Approximately 10–20% of the patients show elevated β-hCG and/or hCGβ concentrations. β-hCG levels are usually below 2000 IU/L. Elevated hCGβ levels are found in approximately one third of marker positive seminomas /16/. In a study /17/, specific determination of hCGβ increased the frequency of marker positive seminomas from 17% to 57% and of marker positive relapses from 32% to 59%.

β-hCG levels > 5,000 IU/L or elevated hCGβ let suspect the presence of a combination tumor and make the presence of a pure seminoma most unlikely, irrespectively of histological classification. This is relevant for therapeutic decisions about orchiectomy and radiation therapy or retroperitoneal lymphadenectomy and/or chemotherapy.

Spermatocytic seminomas represent less than 1% of testicular cancers and have a reduced propensity to metastasize. Median age at diagnosis is 48 years. Immunohistochemical studies showed that spermatocytic seminomas are negative for human placental alkaline phosphatase (PLAP) /18/. In classic seminoma, the serum PLAP concentration is elevated in 58.5% of the cases /19/.

Non seminomatous cancer

In non seminomatous tumors of the undifferentiated teratoma type (WHO: embryonic cancer) and in those of the intermediate type (WHO: embryonic cancer with teratoma, teratocarcinoma), hCG is positive in 60% and 57%, respectively. β-hCG levels up to 1,000 IU/L are measured. The prevalence of elevated serum β-hCG concentrations depends on the tumor stage (stage I 45%, stage II 55%, stage III 84%); however, this does not apply to the absolute serum values /20/. False negative marker results have been reported in 10–33% of cases prior to orchiectomy. Furthermore, according to immunohistological investigations, metastases may synthesize markers less often than the primary tumor. Thus, normalization of hCG and AFP post orchiectomy could be mistaken for complete tumor removal.

Brain metastases occur in about 6% of teratocarcinomas, resulting in high serum β-hCG concentrations (above 10,000 IU/L), with higher values being significantly more frequent than low ones. Additional determination of β-hCG in the cerebrospinal fluid may more reliably detect brain metastases.

Neurohypophyseal germinoma /21/

Differential diagnosis must consider neurohypophyseal germinoma in the presence of suprasellar lesions in children and young adults. In many cases, neurohypophyseal germinoma must be differentiated from tuberofundibular hypophysitis. The determination of hCG in cerebrospinal fluid (CSF) is useful in the diagnosis. It is important to use an assay that detects as many hCG forms as possible, especially hCGβ. According to several authors, the β-hCG threshold in CSF is ≥ 50 IU/L. Other authors recommend measurement in serum and CSF and propose a CSF/serum ratio ≥ 2 as indicative of the presence of neurohypophyseal germinoma. In a study /21/, three patients diagnosed with neurohypophyseal germinoma had a hCG concentration in CSF above 0.7 IU/L determined in an assay that measures intact hCG and hCGβ.

Choriocarcinoma /22/

Testicular choriocarcinoma is very rare and accounts for less than 1% of testicular cancers. According to immunohistochemical investigations, the tumor consists of proliferative syncytiotrophoblast and cytotrophoblast cells. It is a highly malignant cancer and mainly occurs as a component of mixed tumors. In many cases (83%), patients present with multiple metastases that are mainly located in the lung, brain and liver. The prognosis for male choriocarcinoma is poor. Serum hCG concentrations can be elevated up to several million IU/L and are correlated with the tumor mass. Approximately one million tumor cells will result in a plasma hCG concentration of 10 IU/L.

Table 28.16-8 Diagnosis and treatment germ cell cancer: classification of prognostic groups*


Non seminoma


survival 90%)

Testis or primary extra gonadal retroperitoneal tumor and low markers

AFP < 1,000 μg/L

and β-hCG < 5,000 IU/L

and LD < 1.5 × upper reference interval value

and no non-pulmonary visceral metastases

Any primary localization

Any marker level

No non-pulmonary visceral metastases

survival 80%)

Testis or primary extra gonadal retroperitoneal tumor and intermediate markers

AFP 1,000–10,000 μg/L

and/or β-hCG 5,000–50,000 IU/L

and/or LD (1.5–10) × upper reference interval value

and no presence of non-pulmonary visceral metastases

Any primary localization

and presence of non-pulmonary visceral metastases (liver, CNS, bone, intestinum)

Any marker level

survival 50%)

Primary mediastinal germ cell tumor with or without further risk factors

Testis or primary retroperitoneal tumor

and presence of non-pulmonary visceral metastases (liver, CNS, bone, intestinum)

and/or high markers

AFP > 10,000 μg/L

and/or β-hCG > 50,000 IU/L

and/or LD > 10 × upper reference interval value


* European consensus on diagnosis and treatment of germ cell cancer /13/

Table 28.16-9 FIGO staging and classification for trophoblastic tumors /23/

FIGO anatomical staging

Stage I: Tumor confined to the uterus

Stage II: GTN (post-hydatiform mole trophoblastic neoplasia) extends outside of the uterus, but is limited to the genital structures

Stage III: GTN extends to the lungs, with or without genital tract involvement

Stage IV: All other metastatic sites

Modified WHO prognostic scoring system as adapted by FIGO







< 40

≥ 40

> 40







Interval months from
index pregnancy

< 4

4–< 7

7–< 13

≥ 13

Pre treatment serum

hCG (IU/L)

< 103

103–< 104

104–< 105

≥ 105

Largest tumor size
(including uterus)


3–< 5 cm

≥ 5 cm


Site of metastases




Brain, liver

Number of metastases




> 8

Previous failed



Single drug

2 or
more drugs

GTN, gestational trophoblastic tumor; FIGO, Fédération Internationale de Gynécologie et d’Obstétrique

Table 28.16-10 hCG conversion /6/


MW (kDa)



hCG intact












Table 28.16-11 Reference intervals for hCG in males /6/


< 60 yrs.

≥ 60 yrs.


0.7 (2.0)

2.1 (6.1)


0.04 (2.0)

0.05 (2.1)

Data expressed in IU/L (pmol/L)

Table 28.17-1 HER-2/neu in healthy controls depending on age and gender /2/







< 35












≥ 55





< 35












≥ 55




Data expressed in μg/L; *percentile

Table 28.17-2 Levels of HER-2/neu and clinical outcomes after trastuzumab therapy /11/

Levels from
baseline to

rate (%)

duration of

Days to


> 20%





≤ 20%





Breast cancer patients with 3+ immunohistochemistry and fluorescence in situ hybridization-amplified HER-2/neu, * days

Table 28.18-1 Reference intervals for NSE


Serum /1234/

≤ 10 or ≤ 20

Cerebrospinal fluid /5/


Children /1/


≤ 1 year

≤ 25

1–8 years

≤ 20

Cerebrospinal fluid /6/

1–8 years

≤ 4.8

Values expressed in μg/L

Table 28.18-2 Diagnostic sensitivity of NSE in benign and malignant diseases



Benign disease

Benign diseases are more likely to be associated with mild or transient increases in marker concentration with normalization after remission.



Extra pulmonary


Neural tube defect




Malignant disease

Malignant diseases without treatment are associated with gradual to exponential increases correlated with the tumor mass, tumor stage and localization of metastases. NSE is not useful for screening or for diagnostic purposes but valuable for monitoring the treatment outcome and the course of small cell lung cancer (first line marker), neuroblastoma and APUDoma.

Lung cancer

Small cell lung cancer


  • Limited disease


  • Extensive disease


Non small cell lung carcinoma


  • Large cell


  • Adenocarcinoma


  • Squamous cell cancer


Non lung cancer