Pathogenic germline variant prevalence and genetic testing outcomes in patients with urothelial carcinoma.

Abstract

741 Background: Urothelial carcinoma (UC) is rarely attributed to hereditary causes, but recent studies suggest inherited factors may play a larger role. Current guidelines recommend considering genetic evaluation in patients with bladder cancer diagnosed at <50 years old or a personal or family history of Lynch Syndrome-associated cancers. However, the full pathogenic germline variant landscape and optimal genetic evaluation criteria in UC remain poorly understood. In this study, we report the genetic testing outcomes and clinicopathologic characteristics of patients with UC who underwent genetic evaluation. Methods: A retrospective analysis of all patients with UC who underwent genetic evaluation and testing at the University of Michigan Cancer Genetics Clinic between 2002 and 2024 was performed. Patients who did not have a UC diagnosis code or did not undergo germline testing were excluded. Genetic testing outcomes, variant prevalence and characterization, and associated clinicopathologic features were recorded. Fisher’s Exact test was used to compare clinicopathologic factors between individuals found to have a pathogenic or likely pathogenic variant (PV/LPV) vs. those with negative or uninformative testing. Results: Among 151 patients referred to cancer genetics, 129 underwent genetic testing. Median age at diagnosis was 59 (IQR: 48-68). Most patients (58%) had non-muscle invasive disease at initial diagnosis; 26% had muscle invasive disease, 9% had metastatic disease, and 7% were unknown. Pure urothelial histology was most frequent (71%). A majority (83%) had primary tumor location in the bladder. The most common reasons for referral to genetic counseling were personal history of multiple cancers (58%), diagnosis at perceived younger age (19%), and suspicion for a hereditary syndrome based on syndromic features (9%). Among all tested patients, 35 (27%) had a confirmed PV/LPV; 18 (14%) had a variant of unknown significance (VUS) only. Most frequent PV/LPV genes were MSH2 (n=8, 6%), BRCA2 (n=5, 4%), CHEK2 (n=4, 3%), APC (n=4, 3%), and MUTYH (n=4, 3%). There were no statistically significant differences in testing outcomes based on age of diagnosis (<65 vs >65, p=0.83; <55 vs >55, p=0.16; <45 vs >45, p=0.61), primary tumor location, stage at initial diagnosis, personal history of other cancers, or first-degree family history of urothelial cancer. Oncoprint visualization will be presented. Conclusions: Among patients with UC referred to cancer genetics who had genetic testing, 27% had a confirmed PV/LPV, although this cohort was a selected population. Age threshold and traditional clinicopathologic features were not associated with genetic testing results. These findings may suggest a broader use of genetic evaluation in UC. Future studies should determine the optimal genetic evaluation criteria in UC given the impact on cascade testing and potential therapeutic implications.