Outcomes of immune checkpoint inhibitor therapy in patients with pre-existing autoimmune disease: A propensity-matched real-world analysis of baseline corticosteroid use.

Abstract

e23096 Background: Patients with pre-existing autoimmune disease are frequently excluded from immune checkpoint inhibitor (ICI) clinical trials due to concerns regarding immune-related toxicity and reduced efficacy, particularly in the setting of baseline immunosuppression. As a result, real-world evidence guiding the safety and outcomes of ICIs in this population remains limited. We evaluated clinical outcomes associated with baseline prednisone use among ICI-treated patients with autoimmune disease using a large, multi-institutional electronic health record database. Methods: We conducted a retrospective cohort study using the TriNetX Research Network, identifying adult patients with documented autoimmune disease who received immune checkpoint inhibitors. Patients were stratified based on baseline systemic prednisone exposure at the time of ICI initiation (any dose vs none). Propensity score matching (1:1) was performed to balance demographics, comorbidities, cancer characteristics, ECOG performance status, and key laboratory parameters. Outcomes were assessed after matching using Cox proportional hazards models and Kaplan-Meier survival analysis, excluding patients with outcomes occurring prior to the defined time window. The primary outcome was all-cause mortality. Secondary outcomes included emergency department (ED) visits, hospitalization, and sepsis. Results: After propensity score matching, 3,110 patients receiving baseline prednisone were matched to 3,110 patients without prednisone exposure. Median follow-up after matching was 412.5 days in the prednisone group and 310 days in the non-prednisone group. All-cause mortality occurred in 43.0% of patients receiving baseline prednisone compared with 41.7% in the non-prednisone group. Kaplan-Meier analysis demonstrated no statistically significant difference in overall survival between groups (median survival 914 vs 832 days; hazard ratio [HR] 0.93, 95% CI 0.86-1.00; log-rank p = 0.060). Baseline prednisone use was associated with a higher risk of hospitalization (34.9% vs 26.2%; HR 1.29, 95% CI 1.13-1.48; p < 0.001) and sepsis (16.2% vs 12.9%; HR 1.16, 95% CI 1.02-1.32; p = 0.042). Rates of emergency department visits were similar between groups. Conclusions: In this large real-world cohort of ICI-treated patients with pre-existing autoimmune disease, baseline prednisone use was not associated with a statistically significant increase in mortality, despite higher rates of hospitalization and sepsis. These findings may suggest that selected patients with autoimmune disease receiving baseline corticosteroids may still derive survival benefit from immune checkpoint inhibitors. Prospective studies are needed to define optimal immunosuppression strategies and dosing thresholds in this high-risk population.