Risk-reducing gynecologic surgery in non-BRCA cancer predisposition mutations: a multi-institutional study.

Search for high-grade serous carcinoma and serous tubal intra-epithelial carcinoma in prophylactic salpingectomy or salpingo-oophorectomy specimens from carriers of pathogenic variants in low- to moderate-risk genes (PALB2, RAD51C, RAD51D, BRIP1, CHEK2, BARD1, ATM).

This multi-institutional retrospective study compared histopathologic findings and oncologic outcomes after risk-reducing gynaecologic surgery between carriers of pathogenic variants in low- to moderate-risk genes (PALB2, RAD51C, RAD51D, BRIP1, CHEK2, BARD1, ATM) and BRCA1/BRCA2 carriers, across three academic centres from 2013 to 2022, with all specimens processed using the SEE-FIM protocol. Among the 207 patients operated for a low- or moderate-risk mutation — PALB2 (42%), BRIP1 (31%), RAD51D (9%), RAD51C (8%), ATM (7%), CHEK2 (3%) — only three serous borderline tumours (1%) were identified, the remaining 97% being benign, and all patients had no evidence of disease at a median follow-up of 17 months. In the BRCA1/BRCA2 cohort (n = 388), 18 of 388 specimens (5%) showed pathologic findings, including isolated serous tubal intra-epithelial carcinoma (0.7%) and high-grade serous carcinoma (2%), with a median follow-up of 45 months. BRCA1/BRCA2 variants were associated with increased odds of incidental high-grade serous carcinoma (odds ratio 9.42; 95% CI 1.08-1222; P = .04). The absence of occult carcinoma among non-BRCA carriers supports an individualised, gene-specific approach rather than extrapolating BRCA-based strategies.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

Zero occult serous carcinoma across 207 operations in carriers of low- to moderate-risk genes is consistent with the expected ovarian penetrance of these genes and argues against extrapolating the BRCA timetable to all predispositions. Two caveats temper the conclusion: a median follow-up of 17 months, very short for an incidence endpoint, and a confidence interval running from 1.08 to 1222 that mainly reflects how rare the events are. The message is gene-specific in principle, not yet in its age thresholds.

Analysis by Dr Thibaut Benquey