Assessing endometrial cancer etiology by histologic subgroups

Introduction: Endometrial cancer incidence has increased substantially in the past two decades, with the increases occurring primarily in tumors of non-endometrioid histology, which have a poor prognosis. Evidence regarding the origin(s) of the increase is sparse. We sought to determine whether women who had particular molecular signatures of endogenous and exogenous exposures had an increased risk of non-endometrioid histology, particularly serous tumors. Methods: We investigated single-base substitution (SBS) signatures in exome sequencing data from The Cancer Genome Atlas (TCGA). Standard algorithms for mutational signature assessment were applied. Signatures were evaluated on a continuous scale, and also using a high/low cut point determined by maximally selected rank statistics. Relative risks (RR) and 95% confidence intervals (CI) were computed for 49 signatures, comparing risk for serous in relation to endometrioid histology. Disease-specific survival (DSS) was also evaluated in Cox proportional hazard models, with calculation of hazard ratios (HR) and 95% CI. The proportional hazards assumption was verified by Schoenfeld residuals. All analyses were adjusted for age at cancer diagnosis, and DSS analyses were also adjusted for stage. Results: Included were n=414 women with endometrioid and n= 134 with serous non-endometrioid tumors. High Mismatch Repair defective (dMMR) (SBS6 and SBS15) and Polymerase Epsilon mutation (POLE) (SBS10a; SBS10b) signatures were less common in serous than endometrioid tumors, as expected from the literature (dMMR: SBS6 RR 0.21;95% CI 0.12-0.39; SBS15 0.22;95% CI 0.12-0.40;SBS26 0.21;95% CI 0.12-0.40); SBS15 RR 0.22;95% CI 0.12-0.40 and POLE SBS10a RR 0.23;95% CI 0.10-0.54 and SBS10b 0.33; 95% CI 0.19-0.58). A signature related to tobacco use (SBS4) was also less common in serous tumors (RR 0.32;95% CI 0.15-0.71). Women with high dMMR (SBS6: HR 0.06;95%CI 0.01-0.47; SBS15 HR 0.23;95%CI 0.08-0.64; SBS26: HR 0.42;95% CI 0.21-0.84)) or high POLE (SBS10a HR 0.06;95% CI 0.01-0.46; SBS10b HR 0.13; 0.03-0.54) signatures had a reduced risk of endometrial cancer-specific mortality. Discussion: Endometrioid and serous tumors differ in mutational signatures. The reduced dMMR and POLE signature risk in relation to both serous tumors and DSS are in agreement with other findings. Our results open up new prospects for exploration of serous tumor etiology