Clinical characteristics and influencing factors of anti-PD-1/PD-L1-related severe cardiac adverse event: based on FAERS and TCGA databases.

Combining the FDA Adverse Event Reporting System (FAERS) and the Cancer Genome Atlas (TCGA) databases, we aim to explore the factors that influence anti-programmed cell death protein-1 inhibitors/programmed death-ligand-1 (PD-1/PD-L1) related severe cardiac adverse events (cAEs). We obtained anti-PD-1/PD-L1 adverse event reports from January 2014 to December 2022 from the FAERS database. Disproportionality analysis was performed to find anti-PD-1/PD-L1-related cAEs using the proportional reporting ratio (PRR). We were exploring influencing factors based on multivariate logistic regression analysis. Finally, we utilized a strategy that combines FAERS and TCGA databases to explore the potential immune and genetic influencing factors associated with anti-PD-1/PD-L1-related severe cAEs. Reports of severe cAEs accounted for 7.10% of the overall anti-PD-1/PD-L1 adverse event reports in the FAERS database. Immune-mediated myocarditis (PRR = 77.01[59.77-99.23]) shows the strongest toxic signal. The elderly group (65-74: OR = 1.34[1.23-1.47], ≥ 75: OR = 1.64[1.49-1.81]), male (OR = 1.14[1.05-1.24]), anti-PD-L1 agents (OR = 1.17[1.03-1.33]), patients with other adverse events (OR = 2.38[2.17-2.60]), and the concomitant use of proton pump inhibitor (OR = 1.29[1.17-1.43]), nonsteroidal anti-inflammatory drugs (OR = 1.17[1.04-1.31]), or antibiotics (OR = 1.24[1.08-1.43]) may increase the risk of severe cAEs. In addition, PD-L1 mRNA (Rs = 0.71, FDR = 2.30 × 10) and low-density lipoprotein receptor-related protein 3 (LRP3) (Rs = 0.82, FDR = 2.17 × 10) may be immune and genetic influencing factors for severe cAEs. Severe cAEs may be related to antigen receptor-mediated signalling pathways. In this study, we found that age, gender, anti-PD-1/PD-L1 agents, concomitant other adverse events, concomitant medication, PD-L1 mRNA, and LRP3 may be influencing factors for anti-PD-1/PD-L1-related severe cAEs. However, our findings still require a large-scale prospective cohort validation.