Introduction We conducted a pan-gynecologic cancer analysis of the prevalence of somatic pathogenic gene variants (PGVs) in homologous recombination deficiency (HRD) genes.

Methods The American Association for Cancer Research’s (AACR) Project Genomics Evidence Neoplasia Information Exchange (GENIE) database version 13.1 was queried via cBioPortal (http://genie.cbioportal.org) for the following gynecologic tumors: epithelial ovarian, sex-cord stromal, germ cell, endometrial, uterine sarcoma, cervical and vulvar/vaginal tumors. PGV frequencies of 27 HRD genes were descriptively reported among these tumors: ATM, ARID1A, ATRX, BRCA1, BRCA2, BARD1, BRIP1, BLM, BAP1, CHEK1, CHEK2, FANCA, FANCC, FANCD2, FANCE, FANCF, FANCG, FANCL, MRE11, NBN, PALB2, RAD50, RAD51, RAD51B, RAD51C, RAD51D, WRN.

Results A total of 13,312 tumors from 12,804 patients were included for analysis. At least one PGV in an HRD gene was found in 29.6% (3946/13,312) of all samples analyzed, with the highest frequency observed in endometrial tumors (2156/5087, 42.4%), uterine sarcomas (196/704, 27.8%), and epithelial ovarian tumors (1402/6052, 23.2%). There were also substantial rates of HRD PGVs in cervical and vulvar/vaginal tumors, and comparatively lower rates among germ cell and sex cord/stromal tumors (table 1). Across all tumors, HRD genes with the highest frequencies of PGVs were ARID1A (19.6%), BRCA1 (3.9%), BRCA2 (3.7%), and ATM (3.7%).

Conclusion/Implications NGS data demonstrate a substantial rate of somatic PGVs in HRD genes across most types of gynecologic tumors analyzed. These data suggest the need to expand routine functional HRD status assessment beyond epithelial ovarian tumors, and also suggest the need for clinical trials evaluating the efficacy of HRD targeting agents in these cancers.