Estimating cancer risk in carriers of Lynch syndrome variants in UK Biobank

Abstract

Background Lynch syndrome (LS) is an inherited cancer predisposition syndrome caused by genetic variants affecting DNA mismatch repair (MMR) genes MLH1 , MSH2 , MSH6 and PMS2 . Cancer risk in LS is estimated from cohorts of individuals ascertained by individual or family history of cancer, which may upwardly bias estimates. Methods 830 carriers of pathogenic or likely pathogenic ( path_MMR ) MMR gene variants classified by InSiGHT were identified in 454 756 UK Biobank (UKB) participants using whole-exome sequence. Nelson-Aalen survival analysis was used to estimate cumulative incidence of colorectal, endometrial and breast cancer (BC). Results Cumulative incidence of colorectal and endometrial cancer (EC) by age 70 years was elevated in path_MMR carriers compared with non-carriers (colorectal: 11.8% (95% confidence interval (CI): 9.5% to 14.6%) vs 1.7% (95% CI: 1.6% to 1.7%), endometrial: 13.4% (95% CI: 10.2% to 17.6%) vs 1.0% (95% CI: 0.9% to 1.0%)), but the magnitude of this increase differed between genes. Cumulative BC incidence by age 70 years was not elevated in path_MMR carriers compared with non-carriers (8.9% (95% CI: 6.3% to 12.4%) vs 7.5% (95% CI: 7.4% to 7.6%)). Cumulative cancer incidence estimates in UKB were similar to estimates from the Prospective Lynch Syndrome Database for all genes and cancers, except there was no evidence for elevated EC risk in carriers of pathogenic PMS2 variants in UKB. Conclusion These results support offering incidentally identified carriers of any path_MMR surveillance to manage colorectal cancer risk. Incidentally identified carriers of pathogenic variants in MLH1 , MSH2 and MSH6 would also benefit from interventions to reduce EC risk. The results suggest that BC is not an LS-related cancer.