Background: Trial-level meta-analysis to investigate differences in immune-related adverse event (irAE) profiles between anti-PD-1/PD-L1 antibodies. Materials & methods: Data analyzed from 8730 patients treated with anti-PD-1/PD-L1 monotherapy. Incidence and odds ratios (ORs) were calculated for irAEs overall, selected individual irAEs for individual agents and pooled estimates for anti-PD-1 or anti-PD-L1 antibodies. Results: For anti-PD-L1 versus anti-PD-1 antibodies, we observed a lower risk of any-grade rash, elevated alanine aminotransferase, colitis, grade ≥3 colitis, hypothyroidism and rash. For individual agents, we observed reduced risks of overall any-grade irAEs for atezolizumab versus pembrolizumab and grade ≥3 irAEs for avelumab versus pembrolizumab. Conclusion: irAE risk may vary between anti-PD-1 and anti-PD-L1 antibodies; however, findings are hypothesis-generating.
Keywords: anti-PD-1; anti-PD-L1; immune checkpoint inhibitor; immune-related adverse events; meta-analysis.