Abstract A54: Can clinical criteria reliably distinguish between sporadic and Lynch Syndrome-associated endometrial carcinomas with immunohistochemical loss of MLH1?

Abstract

Abstract Background: Lynch syndrome is a hereditary cancer syndrome caused by a germline defect in one of 4 DNA mismatch repair genes: MLH1, MSH2, MSH6, or PMS2. Hallmark tumors for this syndrome, endometrial and colorectal cancers, have a prevalence of 60-80% in affected individuals. Accurate identification of patients with this syndrome allows for cancer prevention measures to be instituted in the index patient as well as relatives that test positive for the mutation. Referral for genetic testing for Lynch syndrome is often based on clinical screening criteria, which includes young age of cancer diagnosis and a family history of Lynch syndrome associated tumors. Immunohistochemistry and microsatellite instability analysis have emerged as additional screening tools to assist in evaluating a tumor's probability of being sporadic or secondary to Lynch syndrome. This type of testing can either be used universally on all endometrial carcinomas as an initial triage step, or it can be used after clinical screening criteria has identified someone as at-risk. One limitation of using only microsatellite instability testing and immunohistochemistry is that 15-20% of sporadic endometrial and colorectal cancers have immunohistochemical loss of the DNA mismatch repair protein MLH1 and the presence of high levels of microsatellite instability due to methylation of MLH1 rather than germline mutation of MLH1. A follow up test, the PCR-based MLH1 methylation assay, assists in differentiating sporadic endometrial cancers from probable Lynch syndrome tumors in this subset exhibiting immunohistochemical loss of MLH1. A limitation of this assay is that it is PCR-based and may not be accessible to many smaller clinical laboratories. The objective of this study was to determine if Society of Gynecologic Oncology (SGO) clinical criteria were able to correctly distinguish between sporadic endometrial cancers (MLH1 methylated) from probable Lynch syndrome-associated endometrial cancer (MLH1 unmethylated). Methods: Patients with endometrial carcinomas that had undergone hysterectomy at our institution and had immunohistochemical testing for MLH1 and MLH1 promoter methylation analysis were included in this study. Electronic medical records were reviewed to obtain complete clinical and pathologic data for each case. SGO criteria, a modified form of the Revised Bethesda guidelines which uses endometrial cancer as the cancer of reference, were applied to all patients. Results: Of 337 endometrial carcinomas that underwent immunohistochemical evaluation of the four DNA mismatch repair proteins, 54 had immunohistochemical loss of MLH1. 40/54 had MLH1 methylation and were designated as sporadic, while 14/54 lacked MLH1 methylation and were designated as probable Lynch syndrome. In our cohort, 22/54 endometrial carcinoma patients had clinical features that fulfilled SGO criteria for further genetic evaluation, including germline testing for mutation of MLH1. The SGO criteria correctly identified 10/14 probable Lynch syndrome patients. The sensitivity and specificity of SGO criteria within this population were 71.4% and 69.2%, respectively. Conclusions: SGO criteria correctly identified 71% of Lynch syndrome-associated endometrial carcinomas with immunohistochemical loss of MLH1. In addition to missing nearly 30% of endometrial cancer patients with a likely diagnosis of Lynch syndrome, use of SGO criteria alone without the PCR-based MLH1 promoter methylation assay would result in an additional 12 individuals undergoing unindicated germline MLH1 mutational testing. Using the MLH1 methylation assay decreases the number of cases referred for germline testing to 14, thus making the clinical evaluation of these patients more cost effective. Our study highlights the importance of the MLH1 methylation assay in the evaluation of patients undergoing molecular diagnostic testing and evaluation for Lynch syndrome. Citation Format: Amanda S. Bruegl, Bojana Djordjevic, Diana Urbauer, Shannon N. Westin, Pamela T. Soliman, Karen H. Lu, Rajyalakshmi Luthra, Russell R. Broaddus. Can clinical criteria reliably distinguish between sporadic and Lynch syndrome-associated endometrial carcinomas with immunohistochemical loss of MLH1? [abstract]. In: Proceedings of the Eleventh Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2012 Oct 16-19; Anaheim, CA. Philadelphia (PA): AACR; Cancer Prev Res 2012;5(11 Suppl):Abstract nr A54.