CHEK 2 1100 delC Is a SusceptibilityAllele for HNPCC-Related Colorectal

Purpose: The pathogenic CHEK2 1100delC variant is firmly established as a breast cancer susceptibility allele. Dutch CHEK2 1100delC breast cancer families frequently also include colorectal cancer cases, and the variant is particularly prevalent among breast cancer families with hereditary breast and colorectal cancer.Yet, it is still unclear whether CHEK2 1100delC also confers a colorectal cancer risk independent of its breast cancer risk. Experimental Design: CHEK2 1100delC was genotyped in the index cases of 369 Dutch colorectal cancer families that had been excluded for familial breast cancer. The cohort included 132 cases with familial adenomatous polyposis (FAP) and FAP-related disease, and 237 cases with hereditary nonpolyposis colorectal cancer (HNPCC) and HNPCC-related disease. Results: None of the FAP/FAP-related cases carried the CHEK2 1100delC variant. In contrast, CHEK2 1100delC was present in 10 of 237 (4.2%) HNPCC/HNPCC-related cases that was significantly more prevalent than the 1.0% Dutch population frequency (odds ratio, 4.3; 95% confidence interval,1.7-10.7; P = 0.002). Nine of the10 CHEK2 1100delC colorectal cancer cases met the revised Amsterdam and/or Bethesda criteria. The10 CHEK2 1100delC colorectal cancer families had a high-risk cancer inheritance pattern, including 35 colorectal cancer cases, 9 cases with polyps, and 21cases with other tumor types. Conclusion:Our analysis provides strongevidence that the1100delC variant ofCHEK2 confers a colorectal cancer risk in HNPCC/HNPCC-related families, supporting the hypothesis that CHEK2 is a multiorgan cancer susceptibility gene. In 2002, we and others identified CHEK2 1100delC (MIM 604373) as a breast cancer susceptibility allele (1–4). The CHEK2 1100delC variant was present in 5% of non-BRCA1/ BRCA2 breast cancer families compared with only 1% in the general population. Although variants with a population frequency of 1% or more are often considered polymorphisms, the estimated 2-fold increased breast cancer risk for female CHEK2 1100delC carriers had classified CHEK2 1100delC as a pathogenic moderate-risk mutation. Strikingly, the prevalence of CHEK2 1100delC increased with increasing numbers of breast cancer cases in the families, suggesting a rather high-risk breast cancer inheritance pattern for the variant. Also, the CHEK2 1100delC genotype only partially segregated with the breast cancer phenotype, particularly so in families that were more severely affected with breast cancer. Together, these apparent discrepancies have suggested a polygenic breast cancer susceptibility model for the CHEK2 1100delC variant, in which CHEK2 1100delC acts in concert with another as-yet-unknown breast cancer susceptibility allele or alleles (1–5). We had noted that near half of Dutch CHEK2 1100delC breast cancer families also included colorectal cancer cases (11 of 25 families, together with 21 cases; refs. 1, 6). Indeed, when we applied stringent clinical criteria to define a new subtype of breast cancer families with a ‘‘hereditary breast and colorectal cancer’’ (HBCC) phenotype, the CHEK2 1100delC frequency was almost five times higher among the HBCC families compared with breast cancer families not fulfilling the HBCC criteria (18% versus 4%; ref. 6). These results thus suggested that CHEK2 1100delC carriers may also be predisposed to develop colorectal cancer. To evaluate the colorectal cancer risk conferred by CHEK2 1100delC independent of its breast cancer risk, we here have genotyped CHEK2 1100delC in 369 Dutch familial colorectal cancer cases. Materials andMethods Colorectal cancer families. The selected familial colorectal cancer cohort (n = 385) included all colorectal cancer families registered by the Dutch Foundation for Detection of Hereditary Tumors and those that were registered by the Rotterdam Family Cancer Clinic at Erasmus Medical Center. Sixteen hereditary nonpolyposis colorectal cancer (HNPCC)-related families had been excluded because they met our Cancer Prevention and Susceptibility Authors’Affiliations: Departments of Medical Oncology and Clinical Genetics, Josephine Nefkens Institute, Erasmus University Medical Center, Rotterdam, the Netherlands; Department of Gastroentrology and Center for Human and Clinical Genetics, Leiden University Medical Center, Leiden, the Netherlands Received 2/14/08; revised 3/31/08; accepted 4/2/08. Grant support: Dutch Cancer Society grant DDHK 2003-2862. The costs of publication of this article were defrayed in part by the payment of page charges.This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section1734 solely to indicate this fact. Requests for reprints: Mieke Schutte, Department of Medical Oncology, Josephine Nefkens Institute Room Be414, Erasmus University Medical Center, PO Box 2040, 3000 CARotterdam, the Netherlands. Phone: 31-10-7038039; Fax: 3110-7044377; E-mail: a.schutte@erasmusmc.nl. F2008 American Association for Cancer Research. doi:10.1158/1078-0432.CCR-08-0389 www.aacrjournals.org Clin Cancer Res 2008;14(15) August1, 2008 4989 Research. on May 1, 2017. © 2008 American Association for Cancer clincancerres.aacrjournals.org Downloaded from criteria for familial breast cancer [i.e., at least two firstor second-degree relatives (DGR) with breast cancer of whom at least one was diagnosed before age 60 y]. The evaluated Dutch Foundation for Detection of Hereditary Tumors cohort (cohort I) included 114 HNPCC/HNPCCrelated and 91 familial adenomatous polyposis (FAP)/FAP-related families that had been reported as part of another CHEK2 1100delC study (6). The evaluated Erasmus Medical Center cohort (cohort II) included 123 HNPCC/HNPCC-related and 41 FAP/FAP-related families. All cases have consented to search for cancer susceptibility genes and the Medical Ethical Review Board at Erasmus Medical Center has