Radiation therapy management in BRCA1/2 carriers diagnosed with early breast cancer: An international cohort study.

The oncological safety of breast-conserving surgery followed by radiation therapy (BCS + RT) in young women carrying pathogenic or likely pathogenic BRCA1/2 variants remains debated, with mastectomy (MS) often favoured despite limited comparative real-world evidence. We evaluated survival and recurrence outcomes associated with different loco-regional strategies in a large international cohort of young BRCA carriers. The BRCA BCY Collaboration (NCT03673306) is a retrospective, multicentre cohort including women aged ≤ 40 years with invasive breast cancer and confirmed germline BRCA1/2 variants treated between 2000 and 2020. Outcomes were compared between patients treated with BCS + RT, MS alone, or MS plus RT (MS + RT). Endpoints were overall survival (OS), breast cancer-free interval (BCFI), and second primary breast cancer events, defined as ipsilateral breast recurrence (IBR) or contralateral breast cancer (CBC). Multivariable Cox models were used for OS and BCFI. Competing-risks models were used for IBR/CBC. Models were adjusted for prespecified prognostic factors, and subgroup analyses were conducted by BRCA gene, stage, and tumour grade. Among 4,837 patients, 1,704 (35.2%) received BCS + RT, 1,488 (30.8%) MS alone, and 1,645 (34.0%) MS + RT. After a median follow-up of 8.2 years (IQR 4.8-12.7), OS did not differ between BCS + RT and MS alone (adjusted hazard ratio [aHR] 1.02, 95% CI 0.78-1.34). BCFI was comparable across groups. BCS + RT was associated with a higher risk of second primary breast cancer events compared with MS alone (aHR 1.33, 95% CI 1.07-1.66), particularly in patients with BRCA1 variants and stage III disease. In young BRCA1/2 carriers, BCS + RT was not associated with worse OS compared with MS alone, despite a higher risk of second primary breast events. Differences in second primary breast cancer events should be interpreted cautiously given differences in BRCA testing timing and treatment era across groups. These data support individualised loco-regional management within a multidisciplinary framework.