Pathogenic germline variants in patients with endometrial cancer across diverse ancestries.

Abstract

5596 Background: Racial/ethnic disparities in outcomes exist in endometrial cancer (EC), particularly amongst self-reported Black compared to Non-Hispanic White (NH-White) patients. However, the contribution of differences in germline pathogenic variants (gPV) and genetics care to disparities is unknown. We seek to describe the spectrum of gPV in unselected patients with EC by ancestry to characterize variations and describe any differences in subsequent genetic counseling follow-up. Methods: Germline assessment of ≥76 cancer predisposition genes was performed in patients with EC undergoing clinical tumor-normal MSK-IMPACT sequencing from 1/1/15-6/30/21. Self-reported data on race/ethnicity and Ashkenazi Jewish (AJ) ancestry were used to classify patients into mutually exclusive ancestry groups: AJ, Asian, Black, Hispanic, NH-White, or Unknown. Genetic ancestry was inferred from MSK-IMPACT, and those with single population ancestral fraction >0.8 were assigned to African (AFR), European (EUR), East Asian (EAS), Native American (NAM), South Asian (SAS), and AJ (ASJ) ancestry populations or considered admixed (Mixed). Clinicopathologic variables were reported using summary statistics and compared by ancestry using non-parametric tests. Results: Among 1,625 patients with EC, self-reported ancestry [202 AJ (12.4%), 124 (7.6%) Asian, 171 Black (10.5%), 124 Hispanic (7.6%), 927 NH-White (57.1%), and 77 missing (4.7%)] correlated with genetic ancestry. In those identifying as Black, 122 (71%) had >80% African ancestry, 45 (26%) were considered Mixed, and 4 (3%) had missing data. We observed gPV in 216 (13.3%) patients (73 high, 36 moderate and 107 low/uncertain penetrance), with 15 patients having >1 gPV. Rates of gPV varied significantly by self-reported ancestry [AJ: 40 (19.8%), Asian: 15 (12.1%), Black: 12 (7.0%), Hispanic: 15 (12.1%), NH-White: 129 (13.9%), Missing: 5; p=0.009] and genetic ancestry [AFR: 9 (6.4%), AJ: 53 (19.3%), EAS: 9 (11.0%), EUR: 112 (13.3%), NAM: 2 (40%), SAS: 7(20%), Mixed: 17 (8.6%), Missing: 7; p<0.001]. Self-reported Black patients had the lowest rate with 13 gPV observed in 12 patients (7.0%); 5 gPV had biallelic inactivation within the tumor (2 MSH2, 1 MSH6, 1 BRCA2 and 1 FANCA). Biallelic inactivation was observed in 70 (32%) patients overall with highest rates in Black (42%), Asian (40%), AFR (67%) and SAS (71%) populations. In those with new gPV (n=114), 102 (89%) overall were seen in genetics clinic for follow-up care, with the lowest rates in patients identifying as Black (75%) and of AFR ancestry (67%). Conclusions: Although rates of gPV varied by ancestry, they are drivers of disease in all populations, emphasizing the importance of germline assessment in EC. Differences in gPV and subsequent genetics care may contribute to racial/ethnic disparities in outcomes in EC by affecting treatment and cancer prevention and should be studied further.