Abstract 3771: Mismatch repair deficiency in the etiology of ovarian cancer: A systematic review of microsatellite instability, mismatch repair staining, and hMLH1 methylation

Abstract

Abstract Objective: The etiology of ovarian cancer is poorly understood and only few risk factors have been consistently identified. Loss of mismatch repair capacity by mutations or methylation of mismatch repair genes has been described as a cancer-initiating mechanism in ovarian cancer; however, frequency estimates are heterogeneous and associations with ovarian cancer risk factors have been rarely described. We conducted a systematic review to analyze the role of mismatch repair (MMR) deficiency in unselected ovarian cancers as determined by microsatellite instability (MSI), immunohistochemical (IHC) staining for mismatch repair proteins, and hypermethylation of the hMLH1 promoter. Methods: Studies examining MSI, MMR deficiency by IHC staining and hMLH1 promoter hypermethylation in unselected ovarian cancers were identified by a systematic literature search of the PubMed electronic database through August 31, 2009, using keywords “ovarian cancer” or “ovarian carcinoma” and “MMR,” “microsatellite instability,” “MMR,” “mismatch repair,” “IHC,” “immunohistochemistry,” “methylation,” or “MSP”. Variables extracted from all pertinent studies included: sample size, histology, MSI results using the Bethesda panel (including BAT25, BAT26, DS123, D5S346, D17S250), staining for four MMR genes (hMLH1, hMSH2, hMSH6, hPMS2), and hMLH1 methylation status. Pooled proportions with confidence intervals were computed for these variables. Results: A total of 2,151 studies were identified by keyword search. After manual selection according to above-defined criteria, 19 studies could be included in this analysis. The pooled proportion of MSI detection was 0.10 (95% CI, 0.07-0.13) among 377 cases; the pooled proportion of loss of staining of any MMR protein was 0.07 (95% CI, 0.05-0.10) among 473 cases; the pooled proportion of hMLH1 methylation was 0.10 (95% CI, 0.08-0.12) among 810 cases. Although information on histologic features was limited in the studies, we did not observe an association between MMR-deficiency and a specific histologic subtype. Conclusions: Despite a wide body of literature on MMR deficiency and ovarian cancer, only few studies could be included to calculate estimates of MMR frequency in unselected ovarian cancers. Most studies were excluded because they reported only cases selected by family history or by histologic subtype. Only few studies reported on MSI, IHC and methylation results in the same cases, limiting a combined analysis. We observed substantial heterogeneity with regard to the assays used, potentially introducing measurement error. None of the included studies reported risk factor information. Based on the published literature, it is not possible to link the MMR phenotype to specific risk factors for ovarian cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 3771.