RESUMO
OBJECTIVE: The aim of this study was to characterize the safety of programmed death 1 inhibitors in patients with preexisting autoimmune disease. METHODS: A medical records review study was conducted on adults with solid tumor malignancies who received ≥1 dose of pembrolizumab or nivolumab at Emory Healthcare from September 4, 2014 until December 31, 2019. All autoimmune patients were included (n = 77), whereas the nonautoimmune patients were randomized and the first 156 patients were included in a 2:1 ratio to autoimmune patients. The primary objective was the comparison of incidence of immune-related adverse events (irAEs) between patients with preexisting autoimmune disease and those without. Secondary objectives included irAE characterization, irAE treatment, and survival analyses. RESULTS: Preexisting autoimmune disease was controlled in all of the autoimmune patients before immunotherapy initiation. The rate of irAE was 32.7% in the nonautoimmune group and 42.9% in the autoimmune group (odds ratio, 0.65; 95% confidence interval, 0.37-1.14; p = 0.130). In the patient population diagnosed with a rheumatologic autoimmune disease, 23.81% of irAEs were considered to be a flare of their preexisting autoimmune disease. Less patients in the autoimmune group experienced a grade ≥3 irAE (21.21% vs 37.25%, p = 0.379) and received systemic corticosteroids (54.55% vs 67.35%, p = 0.241) for the treatment of the irAE. CONCLUSIONS: These results suggest that pembrolizumab and nivolumab can be safely administered in patients with controlled preexisting autoimmune diseases without a significant increase in irAE compared with patients without autoimmune diseases. Inclusion of patients with preexisting autoimmune diseases in prospective clinical trials is warranted.