Gene-specific cancer risks in female Lynch syndrome carriers: A copula-based meta-analysis

Abstract

Background: Lynch syndrome, caused by germline pathogenic variants in mismatch repair genes, markedly increases risks of endometrial, ovarian, and possibly breast cancer in women. We aimed to establish gene-specific risk profiles for these cancers using a unified multivariate model accounting for correlated outcomes. Methods: Eligible studies reported frequencies of endometrial, ovarian, and breast cancer in female with Lynch syndrome carriers, for at least two mismatch repair genes. We applied a Bayesian multivariate random-effects meta-analysis with a copula model to jointly model prevalence and odds ratios across cancers, accounting for between-study heterogeneity and inter-cancer correlation. Results: Using a copula-based multivariate meta-analysis to account for outcome interdependence, the estimated prevalence was 20 for endometrial cancer, 5.7 for ovarian cancer, and 11 for breast cancer. Gene-specific analyses showed increased endometrial cancer risk with MSH6 (OR = 1.46, 95 CrI 1.02-2.04) and reduced risk with PMS2, relative to other Lynch genes (OR = 0.36, 95 CrI 0.14-0.95). Ovarian cancer risk did not differ significantly by gene. For breast cancer, PMS2 (OR = 1.52, 95 CI 1.02-2.25) and MSH6 (OR = 2.27, 95 CrI 1.08-2.49) were associated with higher risk, while MLH1 and MSH2 carried significantly lower risk. Conclusions: This copula-based meta-analysis identifies gene-specific risks of endometrial and ovarian cancer in female Lynch syndrome carriers, supporting personalized gynecologic surveillance. It also notes higher breast cancer risks in MSH6 and PMS2 carriers, but conflicting evidence from large perspective databases prevents definitive conclusions about breast cancer as part of the Lynch syndrome spectrum.