Risk of de novo immune-related adverse events in patients with cancer and autoimmune disease receiving immune checkpoint inhibitors

Immune checkpoint inhibitors (ICIs) can cause immune-related adverse events (irAEs). Patients with preexisting autoimmune disease are often excluded from clinical trials because of concerns about disease exacerbation or new immune-mediated toxicities, limiting comparative safety evidence. We compared the risk of de novo irAEs among patients with cancer and preexisting autoimmune disease initiating ICIs versus chemotherapy or targeted therapy. A target trial was emulated using Merative MarketScan claims from 2014–2023 with a new-user, active-comparator design. Adults with cancer and preexisting autoimmune disease initiating ICIs, chemotherapy, or targeted therapy were identified. Propensity score matching (1:1) was used to balance measured baseline characteristics. The primary outcome was a de novo irAE. Cox proportional hazards models estimated hazard ratios and 95% confidence intervals overall and across prespecified subgroups by key clinical characteristics. Matched cohorts included 3601 pairs for ICI versus chemotherapy and 1479 pairs for ICI versus targeted therapy. ICI initiation was associated with a higher risk of de novo irAEs than chemotherapy (HR = 1.71, 95% CI = 1.49–1.97) and targeted therapy (HR = 1.43, 95% CI = 1.20–1.70). Estimates varied by ICI class, cancer type, and baseline autoimmune disease category. CTLA-4 inhibitors had the highest estimates among ICI classes, while liver cancer and baseline endocrine autoimmune disease had elevated estimates across both comparisons. ICI initiation was associated with a higher risk of de novo irAEs in this population, with variation by ICI class, cancer type, and autoimmune disease category. These findings may inform individualized treatment decisions, although residual confounding remains an important consideration. Immune checkpoint inhibitors are cancer treatments that help the immune system attack cancer cells, but they can also cause immune-related side effects. People who already have autoimmune diseases are often excluded from clinical trials, so less is known about the safety of these treatments in this population. We used US health insurance claims to compare patients with cancer and preexisting autoimmune disease who started immune checkpoint inhibitors with similar patients receiving chemotherapy or targeted therapy. Patients receiving immune checkpoint inhibitors had a higher risk of developing new immune-related adverse events. These findings provide additional real-world safety information that may help patients and clinicians better understand treatment risks and guide future research.