Article Tools
ORIGINAL REPORTS
Gastrointestinal Cancer
Article Tools
Analysis of EPCAM Protein Expression in Diagnostics of Lynch Syndrome
Abstract
Lynch syndrome is an inherited tumor predisposition syndrome caused by germline mutations of DNA mismatch repair (MMR) genes, mainly MLH1 and MSH2. Recently, germline deletions affecting the epithelial cell adhesion molecule (EPCAM) gene located upstream of MSH2 were identified as a novel mutational mechanism causing Lynch syndrome by epigenetic inactivation of the respective MSH2 allele. Immunohistochemical analysis of MMR protein expression is a hallmark of Lynch syndrome diagnostics, but it cannot distinguish between EPCAM deletion carriers and MSH2 mutation carriers. We hypothesized that EPCAM protein expression might be altered in tumors from patients with a germline EPCAM deletion.
Immunohistochemistry was used to assess EPCAM expression in Lynch syndrome–associated MSH2-negative tumors (n = 26). Multiplex ligation-dependent probe amplification (MLPA) analysis was performed to detect germline deletions of the EPCAM and MSH2 gene loci.
In four MSH2-negative tumors, a concomitant lack of EPCAM expression was detected. MLPA analysis revealed heterozygous EPCAM deletions in all patients with EPCAM-negative tumors. In contrast, EPCAM expression was positive in all cancers from patients with germline alterations affecting MSH2 but not EPCAM. Two EPCAM deletions were detected in patients with an EPCAM-positive tumor.
Lynch syndrome (clinically referred to as hereditary nonpolyposis colorectal cancer) is a frequent dominantly inherited cancer predisposition syndrome caused by germline alterations that affect DNA mismatch repair (MMR) genes.1 Mutation carriers are predisposed to the development of colorectal cancers and extracolonic tumors, mainly endometrial cancers.2 Lynch syndrome–associated cancers are characterized by MMR deficiency, which results from the inactivation of the second allele of the MMR gene mutated in the germline. As a consequence of MMR deficiency, Lynch syndrome–associated cancers with the high-level microsatellite instability (MSI-H) phenotype present an accumulation of insertion and/or deletion mutations at short repetitive microsatellite sequences spread over the whole genome.3 MSI-H mediates tumor progression when mutations of microsatellites located in the coding region of tumor suppressor genes lead to their functional inactivation.4
The majority of Lynch syndrome–causing mutations affect the two MMR genes MLH1 and MSH2; in addition, mutations affecting the MMR genes MSH6 and PMS2 are found. Recently, deletions of the epithelial cell adhesion molecule (EPCAM) gene, which is located upstream of MSH2, have been discovered as a novel mutational mechanism in Lynch syndrome5 by using multiplex ligation-dependent probe amplification (MLPA) analysis containing probes for this region. Deletions affecting the 3′ exons of the EPCAM gene lead to a transcriptional read-through and mediate epigenetic silencing of the respective MSH2 allele in a mosaic pattern because only EPCAM-expressing cells show inactivation of MSH2. Consequently, heterozygous germline EPCAM deletions predispose to the development of MSH2-negative MSI-H cancers.5,6
Lynch syndrome diagnostics encompass tumor analysis before germline mutation analysis. They account for the presence of MSI-H in tumors7–9 and for the expression of MMR proteins.10–12 Immunohistochemical tools are available for four of the Lynch syndrome–associated genes (MSH2, MLH1, MSH6, and PMS2) to analyze the expression of the respective protein in patients' tumors. Lack of MMR protein expression is detected in more than 90% of Lynch syndrome–associated cancers and is closely related to the presence of germline mutations in the respective MMR gene. Therefore, immunohistochemical analysis of MMR protein expression guides the human geneticist in selecting the gene to be analyzed for germline mutations. However, MMR protein immunohistochemistry fails to discriminate between EPCAM deletion carriers and MSH2 mutation carriers.
The effects of heterozygous germline EPCAM deletions on EPCAM protein expression in associated cancers have not been analyzed so far. Given that the EPCAM and MSH2 genes are close neighbors on chromosome 2, we hypothesized that EPCAM protein expression might be altered in colorectal cancer lesions from Lynch syndrome patients with EPCAM deletions. Here, we analyzed the relation between EPCAM expression in tumors and the germline alteration underlying MSH2 inactivation to evaluate the suitability of EPCAM immunohistochemistry for the identification of patients with Lynch syndrome caused by a heterozygous germline EPCAM deletion.
Tumors and blood specimens were collected at the Department of Applied Tumor Biology, Institute of Pathology, University Hospital Heidelberg in the framework of the German Hereditary Nonpolyposis Colorectal Cancer Consortium funded by the Deutsche Krebshilfe (German Cancer Aid). Pathogenic MSH2 germline mutations were identified by direct exon-wise sequencing in 35 of 47 families with MSH2-negative tumors who were undergoing diagnostics for Lynch syndrome in our department. For this study, 26 MSH2-negative tumors (colorectal cancer, n = 25; duodenal cancer, n = 1) were available, including 12 tumors from patients in whom no pathogenic MSH2 germline mutation had been identified and 14 tumors from patients with pathogenic MSH2 mutations detected by exon-wise sequencing. All tumors were from different patients and independent families. Complete exon-wise sequencing of the MSH2 gene was performed for all included patients. The study was approved by the Institutional Ethics Committee. Written consent was obtained from all patients included in the study.
Microsatellite instability analysis was performed as described previously,9 by using markers BAT25, BAT26, CAT25, D2S123, D5S346, and D17S250/Mfd15. MSH2 immunohistochemistry was performed by using the antibody FE11 (Calbiochem, Gibbstown, NJ) as described.13 Immunohistochemistry of EPCAM was performed by using mouse monoclonal antibody BER-Ep4 (Dako, Glostrup, Denmark) at a dilution of 1:100 following standard procedures.
To detect genomic deletions affecting the EPCAM and MSH2 gene loci, MLPA was performed by using the SALSA MLPA kits P003 and P072-B1 (MRC Holland, Amsterdam, the Netherlands) following the manufacturer's instructions.
To analyze a potential relation between the type of MSH2-inactivating germline alteration and EPCAM expression in the tumor, tumors negative for MSH2 protein expression were investigated by EPCAM immunohistochemistry. Twelve MSH2-negative tumors from patients without pathogenic germline MSH2 mutations detectable by direct exon-wise sequencing were compared with 14 MSH2-negative tumors from patients who harbored a pathogenic germline mutation in MSH2 detected by exon-wise sequencing.
EPCAM immunohistochemistry was performed to determine EPCAM protein expression and to evaluate the suitability of the method for discrimination between EPCAM deletion carriers and MSH2 mutation carriers. Four of 26 tumors presented with lack of EPCAM protein expression, whereas EPCAM was expressed in the remaining 22 tumors. Representative results of EPCAM immunohistochemistry are shown in Figure 1. Loss of EPCAM expression was accompanied by loss of MSH2 expression, suggesting that one somatic event caused inactivation of both EPCAM and MSH2 (Fig 2). All tumors that exhibited lack of EPCAM expression were from patients without a pathogenic MSH2 mutation detectable by exon-wise sequencing. Regular EPCAM expression was observed in all tumors derived from patients harboring MSH2 germline mutations that had been identified by exon-wise sequencing (Fig 1).
Fig 1. (A, B) Strong expression of epithelial cell adhesion molecule (EPCAM) is seen in MSH2-deficient carcinomas of patients with germline MSH2 nonsense and frame shift mutations detected by exon-wise sequencing (patient P14, ×25; patient P19, ×40). (C) EPCAM expression is membranous and cytoplasmic in normal epithelium (bottom right) and carcinoma (top left); patient P19, ×100. (D-F) EPCAM expression is retained in normal mucosa but lost in carcinoma cells of patients with germline deletion of the EPCAM locus; (D) patient P1, duodenal carcinoma, ×40; (E) patient P1, duodenal carcinoma, ×100; (F) patient P3, colon cancer, ×100.
Fig 2. (A, B) Synchronous loss of epithelial cell adhesion molecule (EPCAM) and MSH2 protein expression characterize carcinomas from patients with heterozygous EPCAM germline deletions (patient P4, colon cancer, ×100). (C, D) Expression loss of both EPCAM and MSH2 was already observed in apparently non-neoplastic crypts with slight architectural abnormalities, yet lacking significant cellular atypia (patient P4, colon mucosa, ×100).
To relate EPCAM protein expression in the tumor to the germline alteration underlying MSH2 inactivation, EPCAM and MSH2 deletion analysis was performed in all 12 patients in whom no pathogenic MSH2 mutation could be detected by exon-wise sequencing. MLPA analysis revealed deletions affecting the genomic region that encompassed the EPCAM gene and/or the intergeneic region between EPCAM and MSH2 in all four patients showing lack of EPCAM protein expression in the tumor. In addition, deletions affecting the EPCAM gene locus were identified in two patients with EPCAM-positive tumors. The location of the deletions in relation to the MLPA probes is shown in Figure 3.
Fig 3. Results of epithelial cell adhesion molecule (EPCAM) deletion analysis. Multiplex ligation-dependent probe amplification (MLPA) probes are shown in relation to the genomic region encompassing EPCAM and the 5′ region of MSH2. MLPA probes are denoted by fragment lengths (SALSA MLPA kit P072-B1, MRC Holland, Amsterdam, the Netherlands). Blue circles indicate deletions; white circles indicate biallelic amplification. No deletions were detected in the remaining patients with MSH2-negative, EPCAM-positive tumors. a, exons 1 to 6 of MSH2 deleted; b, exons 1 to 16 of MSH2 deleted; c, exons 1 and 2 of MSH2 deleted.
In this study, we demonstrated that EPCAM protein expression was frequently abrogated in cancers associated with Lynch syndrome caused by heterozygous EPCAM germline deletions. Lack of EPCAM expression in the tumor was detected in four patients with germline deletions affecting the genomic region containing EPCAM. This suggests that no EPCAM product detectable by antibody BER-Ep4 is translated from previously described EPCAM/MSH2 fusion transcripts5 and that a second somatic hit inactivated wild-type MSH2 and EPCAM. Similar to the observation in a previously published finding5of retained EPCAM expression in a colorectal carcinoma from one patient with a heterozygous EPCAM deletion in the germline, we observed retained EPCAM protein expression in two patients with a germline deletion, most likely because the second MSH2-inactivating hit did not affect the EPCAM gene.
In concordance with the abundant expression of EPCAM protein described for the majority of epithelial tissues and epithelial cancers,14,15 EPCAM protein expression was detectable in all analyzed tumors from patients with proven MSH2 germline mutations detectable by exon-wise sequencing and in an additional set of 91 colorectal tumors (data not shown). This indicates that lack of EPCAM expression was highly specific for the detection of EPCAM deletions in the germline.
Immunohistochemical analysis of EPCAM is a standard procedure in routine pathology. EPCAM immunohistochemistry helps to discriminate basal cell carcinoma from squamous cell carcinoma of the skin; it is useful in the distinction of epitheloid mesothelioma and pulmonary adenocarcinoma, and in the identification of lymph node micrometastasis. The consistent expression pattern of EPCAM in epithelial tissues and cancers suggests that lack of EPCAM expression may serve as a specific marker to identify Lynch syndrome patients with heterozygous EPCAM deletions in the germline.
Lynch syndrome diagnosis encompasses three crucial steps: (1) the identification of patients or families who may have the syndrome by using clinical criteria (Bethesda criteria),7,16 (2) microsatellite instability analysis and immunohistochemical MMR protein expression analysis in tumors from affected family members, and (3) if MSI-H and an altered MMR protein expression pattern were detected, mutation analysis of the respective gene in the germline. Whereas microsatellite instability analysis is the most sensitive method for detecting MMR deficiency in the tumor,10,11 it fails to predict which gene in the germline may be affected. Immunohistochemical staining of MMR proteins is a crucial cost-reducing factor that simplifies Lynch syndrome diagnostics, predicting the gene to be analyzed for mutations in the germline. To date, it has not been possible to use immunohistochemical tumor analysis to differentiate between Lynch syndrome patients carrying MSH2 germline mutations from those carrying EPCAM deletions. The results of this study suggest that the inclusion of EPCAM immunohistochemistry might solve this problem, potentially representing a novel diagnostic tool that may add significantly to the clinical value of immunohistochemistry in Lynch syndrome diagnostics.
EPCAM is a glycosylated transmembrane protein expressed in a variety of human epithelial cancers, particularly in adenocarcinomas, and has been shown to function as an oncoprotein, which enhances the expression of growth-promoting genes such as c–myc and cyclins.17 Thus, it can be speculated that loss of EPCAM expression by the somatic hit inactivating MSH2 should alter tumor-relevant signaling pathways. On the basis of the involvement of EPCAM in tumor progression, one might assume that the invasive or metastatic potential of EPCAM-deficient tumors may be lower compared with EPCAM-positive cancers. These considerations suggest that loss of EPCAM in EPCAM deletion-induced Lynch syndrome is likely to have an effect on the biologic aspects of tumors. Large multicenter efforts will be required to assess tumor spectrum, penetrance, and prognosis in Lynch syndrome caused by EPCAM deletions and to compare the findings in families with classical MMR gene mutation-related Lynch syndrome.
Although there are no population-based data on the frequency of EPCAM deletions as a cause of Lynch syndrome, studies suggest that up to 30% of Lynch syndrome patients with MSH2-negative tumors or approximately 20% of Lynch syndrome patients without mutations of the MMR genes might be carriers of EPCAM germline deletions.18,19 If this is true, EPCAM deletions are the predisposing factor in more Lynch syndrome families than PMS2 and potentially MSH6 mutations. This underlines the importance of EPCAM deletions in Lynch syndrome and highlights the need for a tumor diagnostics tool that identifies EPCAM deletion carriers before germline mutation analysis. The results of the present study encourage evaluation of EPCAM immunohistochemistry as a diagnostic tool in Lynch syndrome, suggesting that somatic loss of EPCAM expression has a high positive predictive value for the identification of EPCAM deletions in the germline.
Supported by the Deutsche Krebshilfe (German Cancer Aid) and by Grant No. BL-554/3-2 from the Deutsche Forschungsgemeinschaft.
M.K. and A.Y.V. contributed equally to the manuscript.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
The author(s) indicated no potential conflicts of interest.
Conception and design: Matthias Kloor, Anita Y. Voigt, Hendrik Bläker
Financial support: Peter Schirmacher, Magnus von Knebel Doeberitz
Administrative support: Peter Schirmacher, Magnus von Knebel Doeberitz
Provision of study materials or patients: Matthias Kloor, Peter Schirmacher, Magnus von Knebel Doeberitz
Collection and assembly of data: Matthias Kloor, Anita Y. Voigt, Hans K. Schackert, Hendrik Bläker
Data analysis and interpretation: Matthias Kloor, Anita Y. Voigt, Hans K. Schackert, Hendrik Bläker
Manuscript writing: Matthias Kloor, Anita Y. Voigt, Peter Schirmacher, Magnus von Knebel Doeberitz, Hendrik Bläker
Final approval of manuscript: Matthias Kloor, Anita Y. Voigt, Hans K. Schackert, Peter Schirmacher, Magnus von Knebel Doeberitz, Hendrik Bläker
| 1. | HT Lynch, A de la Chapelle: Hereditary colorectal cancer N Engl J Med 348: 919– 932,2003 Crossref, Medline, Google Scholar |
| 2. | T Goecke, K Schulmann, C Engel , etal: Genotype-phenotype comparison of German MLH1 and MSH2 mutation carriers clinically affected with Lynch syndrome: A report by the German HNPCC Consortium J Clin Oncol 24: 4285– 4292,2006 Link, Google Scholar |
| 3. | CR Boland, A Goel: Microsatellite instability in colorectal cancer Gastroenterology 138: 2073– 2087.e3,2010 Crossref, Medline, Google Scholar |
| 4. | SM Woerner, M Kloor, M von Knebel Doeberitz , etal: Microsatellite instability in the development of DNA mismatch repair deficient tumors Cancer Biomark 2: 69– 86,2006 Crossref, Medline, Google Scholar |
| 5. | MJ Ligtenberg, RP Kuiper, TL Chan , etal: Heritable somatic methylation and inactivation of MSH2 in families with Lynch syndrome due to deletion of the 3′ exons of TACSTD1 Nat Genet 41: 112– 117,2009 Crossref, Medline, Google Scholar |
| 6. | KW Jasperson, TM Tuohy, DW Neklason , etal: Hereditary and familial colon cancer Gastroenterology 138: 2044– 2058,2010 Crossref, Medline, Google Scholar |
| 7. | CR Boland, SN Thibodeau, SR Hamilton , etal: A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: Development of international criteria for the determination of microsatellite instability in colorectal cancer Cancer Res 58: 5248– 5257,1998 Medline, Google Scholar |
| 8. | JW Bacher, WM Abdel Megid, MG Kent-First , etal: Use of mononucleotide repeat markers for detection of microsatellite instability in mouse tumors Mol Carcinog 44: 285– 292,2005 Crossref, Medline, Google Scholar |
| 9. | P Findeisen, M Kloor, S Merx , etal: T25 repeat in the 3′ untranslated region of the CASP2 gene: A sensitive and specific marker for microsatellite instability in colorectal cancer Cancer Res 65: 8072– 8078,2005 Crossref, Medline, Google Scholar |
| 10. | NM Lindor, LJ Burgart, O Leontovich , etal: Immunohistochemistry versus microsatellite instability testing in phenotyping colorectal tumors J Clin Oncol 20: 1043– 1048,2002 Link, Google Scholar |
| 11. | C Engel, J Forberg, E Holinski-Feder , etal: Novel strategy for optimal sequential application of clinical criteria, immunohistochemistry and microsatellite analysis in the diagnosis of hereditary nonpolyposis colorectal cancer Int J Cancer 118: 115– 122,2006 Crossref, Medline, Google Scholar |
| 12. | K Lagerstedt Robinson, T Liu, J Vandrovcova , etal: Lynch syndrome (hereditary nonpolyposis colorectal cancer) diagnostics J Natl Cancer Inst 99: 291– 299,2007 Crossref, Medline, Google Scholar |
| 13. | M Kloor, C Sutter, N Wentzensen , etal: A large MSH2 Alu insertion mutation causes HNPCC in a German kindred Hum Genet 115: 432– 438,2004 Crossref, Medline, Google Scholar |
| 14. | SV Litvinov, MP Velders, HA Bakker , etal: Ep-CAM: A human epithelial antigen is a homophilic cell-cell adhesion molecule J Cell Biol 125: 437– 446,1994 Crossref, Medline, Google Scholar |
| 15. | MJ Winter, ID Nagtegaal, JH van Krieken , etal: The epithelial cell adhesion molecule (Ep-CAM) as a morphoregulatory molecule is a tool in surgical pathology Am J Pathol 163: 2139– 2148,2003 Crossref, Medline, Google Scholar |
| 16. | A Umar, CR Boland, JP Terdiman , etal: Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability J Natl Cancer Inst 96: 261– 268,2004 Crossref, Medline, Google Scholar |
| 17. | M Münz, C Kieu, B Mack , etal: The carcinoma-associated antigen EpCAM upregulates c-myc and induces cell proliferation Oncogene 23: 5748– 5758,2004 Crossref, Medline, Google Scholar |
| 18. | RC Niessen, RM Hofstra, H Westers , etal: Germline hypermethylation of MLH1 and EPCAM deletions are a frequent cause of Lynch syndrome Genes Chromosomes Cancer 48: 737– 744,2009 Crossref, Medline, Google Scholar |
| 19. | ME Kovacs, J Papp, Z Szentirmay , etal: Deletions removing the last exon of TACSTD1 constitute a distinct class of mutations predisposing to Lynch syndrome Hum Mutat 30: 197– 203,2009 Crossref, Medline, Google Scholar |
Acknowledgment
We thank Beate Kuchenbuch, Nina Nelius, Petra Hoefler, and Michael Geissler for their excellent technical assistance.
