Does Germline BRCA1/2 Status Modify the Effect of Neoadjuvant Pembrolizumab on Pathological Complete Response in Early-Stage Triple-Negative Breast Cancer? A Systematic Review and Meta-analysis of Treatment-Biomarker Interactions

Abstract

Background: Germline BRCA1/2 pathogenic variants are associated with chemotherapy response, but higher response among pembrolizumab-treated carriers does not establish a treatment-predictive interaction. Objective: To determine whether germline BRCA1/2 status modifies the effect of neoadjuvant pembrolizumab on pathological complete response in early-stage triple-negative breast cancer, estimated as a within-study treatment-by-biomarker interaction on the ratio-of-odds-ratios scale. Methods: Accessible bibliographic, registry, and citation sources were searched from inception through the final accessible-source update for comparative cohorts of neoadjuvant pembrolizumab plus chemotherapy versus chemotherapy alone reporting pathological complete response (pCR) by germline BRCA1/2 status. One author approved screening, extraction, ROBINS-I, and ICEMAN judgements. Treatment-by-biomarker interactions were expressed as ratios of odds ratios (RORs) and pooled by restricted-maximum-likelihood random effects with Hartung-Knapp inference. Results: Three retrospective cohorts (655 participants) were included; two (415 participants) provided complete interaction cells. Study RORs were 3.65 (95% CI 0.52-25.62) and 2.76 (0.39-19.39). The pooled ROR was 3.17 (0.54-18.51; p=0.076), with tau-squared=0 and I-squared=0%. The third cohort remained qualitative because noncarrier cells were irreconcilable. Risk of bias was serious, interaction credibility was low, and certainty was very low. Conclusion: The direction of effect is compatible with greater relative pembrolizumab-associated pCR improvement among carriers, but the evidence does not establish a predictive interaction. Prospective adjusted interaction analyses are required.