Immunogenomic correlates of immune-related adverse events for anti-programmed cell death 1 therapy

Despite impressive antitumor efficacy of programmed cell death 1 (PD-1) inhibitors, this inhibition can induce mild to severe autoimmune toxicities, termed immune-related adverse events (irAEs). Yet, predictive pretreatment biomarkers for irAEs development across cancer types remain elusive. We first assessed cellular and molecular factors. To determine factors predicting the risk of irAEs for anti-PD-1 immunotherapy across multiple cancer types, an integrative analysis of cellular and molecular factors from 9104 patients across 21 cancer types and 4865522 postmarketing adverse event reports retrieved from adverse event reporting system was then performed. Accuracy of predictions was quantified as Pearson correlation coefficient determined using leave-one-out cross-validation. Independent validation sets included small cell lung cancer and melanoma cohorts. Out of 4865522 eligible adverse events reports, 10412 cases received anti-PD-1 monotherapy, of which, 2997 (28.78%) exhibited at least one irAE. Among established immunogenomic factors, dendritic cells (DC) abundance showed the strongest correlation with irAEs risk, followed by tumor mutational burden (TMB). Further predictive accuracy was achieved by DC and TMB in combination with CD4⁺ naive T-cells abundance, and then validated in the small cell lung cancer cohort. Additionally, global screening of multiomics data identified 11 novel predictors of irAEs. Of these, IRF4 showed the highest correlation. Best predictive performance was observed in the IRF4 - TCL1A - SHC-pY317 trivariate model. Associations of IRF4 and TCL1A expression with irAEs development were verified in the melanoma cohort receiving immune checkpoint inhibitors. Collectively, pretreatment cellular and molecular irAEs-associated features as well as their combinations are identified regardless of cancer types. These findings may deepen our knowledge of irAEs pathogenesis and, ultimately, aid in early detection of high-risk patients and management of irAEs.