Comparative efficacy and safety of PD-1/PD-L1 inhibitors in triple negative breast cancer: a systematic review and network meta-analysis of randomized controlled trials

Background: Triple-Negative Breast Cancer (TNBC) is a lethal subtype of breast cancer with limited treatment options. The purpose of this Network Meta-Analysis (NMA) is to compare the efficacy and safety of inhibitors of programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1) in treating TNBC.

Methods: Our search strategy was used in six databases: PubMed, Cochrane Library, Cumulative Index to Nursing and Allied Health Literature database, Embase, Scopus, and Web of Science up to November 2nd, 2022, as well as a thorough search in the most used trial registries. We included phase II and III randomized controlled trials that looked at the efficacy of PD-1/PD-L1 inhibitors in the treatment of TNBC and reported either Overall Survival (OS), Progression-Free Survival (PFS), or pathological Complete Response (pCR). The risk of bias was assessed utilizing Cochrane's risk of bias 2 tool, and the statistical analysis was performed using a frequentist contrast-based method for NMA by employing standard pairwise meta-analysis applying random effects model.

Results: 12 trials (5324 patients) were included in our NMA including seven phase III trials. Pembrolizumab in a neoadjuvant setting achieved a pooled OS of 0.82 (95% Confidence Interval (CI) 0.65 to 1.03), a PFS of 0.82 (95% CI 0.71 to 0.94) and a pCR 2.79 (95% CI 1.07 to 7.24) compared to Atezolizumab's OS of 0.92 (95% CI 0.74 to 1.15), PFS of 0.82 (95% CI 0.69 to 0.97), and pCR of 1.94 (95% CI 0.86 to 4.37). Atezolizumab had less grade ≥ 3 adverse events (OR 1.48, 95% CI 0.90 to 2.42) than Pembrolizumab (OR 1.90, 95% CI 1.08 to 3.33) in the neoadjuvant setting.

Conclusions: PD-1/PD-L1 inhibitors exhibited varying efficacy in terms of OS, PFS, and pCR. They were associated with an increase in immune-related adverse effects. When used early in the course of TNBC, PD-1/PD-L1 inhibitors exert their maximum benefit. Durvalumab as a maintenance treatment instead of chemotherapy has shown promising outcomes. Future studies should focus on PD-L1 expression status and TNBC subtypes, since these factors may contribute to the design of individualized TNBC therapy regimens. Systematic review registration PROSPERO Identifier: CRD42022380712.