Efficacy, safety, and PD-L1-defined outcomes of neoadjuvant immune checkpoint inhibitors in early-stage triple-negative breast cancer: a systematic review and meta-analysis.

BACKGROUND: Neoadjuvant immune checkpoint inhibitors combined with chemotherapy improve pathological complete response in early-stage triple-negative breast cancer, but the consistency of benefit, role of PD-L1 expression, and immune-related toxicity remain clinically relevant. METHODS: PubMed, Embase, and the Cochrane Central Register of Controlled Trials were searched from inception to June 30, 2026, for randomized trials comparing neoadjuvant immune checkpoint inhibitors plus chemotherapy with chemotherapy alone or placebo plus chemotherapy in early-stage triple-negative breast cancer. The primary efficacy analysis was restricted to phase III trials. Risk ratios and 95% confidence intervals were pooled using a random-effects model. Risk of bias and certainty of evidence were assessed using RoB 2 and GRADE. RESULTS: Four phase III trials including 3,377 patients were included in the primary efficacy synthesis. Neoadjuvant immune checkpoint inhibitors plus chemotherapy significantly increased pathological complete response compared with control treatment (risk ratio, 1.27; 95% confidence interval, 1.19-1.36; P < 0.001), with no statistical heterogeneity. Benefit was observed in both PD-L1-positive tumors (risk ratio, 1.32; 95% confidence interval, 1.18-1.48) and PD-L1-negative tumors (risk ratio, 1.45; 95% confidence interval, 1.22-1.72). Grade ≥ 3 treatment-related adverse events were not significantly increased (risk ratio, 1.06; 95% confidence interval, 0.98-1.13), whereas immune-related adverse events were more frequent with immune checkpoint inhibitors (risk ratio, 3.52; 95% confidence interval, 1.88-6.57). CONCLUSIONS: Neoadjuvant immune checkpoint inhibitors plus chemotherapy improve pathological complete response in early-stage triple-negative breast cancer, including across PD-L1-defined subgroups, but increase immune-related toxicity. PD-L1 expression alone appears insufficient for treatment selection. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/recorddashboard#, identifier CRD420261412366.