Safety and Efficacy of Checkpoint Inhibition in Patients With Melanoma and Preexisting Autoimmune Disease: A Cohort Study

Abstract

Background: Because immune checkpoint inhibition (ICI) can cause immune-related adverse events (irAEs) mimicking immunologic diseases, patients with preexisting autoimmune disease (AID) have been excluded from clinical trials. Objective: To evaluate the safety and efficacy of ICI in patients with advanced melanoma with and without AID. Design: Nationwide cohort study. Setting: The Netherlands. Patients: 4367 patients with advanced melanoma enrolled in the Dutch Melanoma Treatment Registry (DMTR) between July 2013 and July 2018 and followed through February 2019. Measurements: Patient, clinical, and treatment characteristics; irAEs of grade 3 or higher; treatment response; and survival. Results: A total of 415 patients (9.5%) had AID, categorized as rheumatologic AID (n = 227), endocrine AID (n = 143), inflammatory bowel disease (IBD) (n = 55), or "other" (n = 8). Of these, 228 patients (55%) were treated with ICI (vs. 2546 [58%] without AID); 87 were treated with anti-cytotoxic T lymphocyte-associated protein 4 (CTLA-4), 187 with anti- programmed cell death 1 (PD-1), and 34 with the combination. The incidences of irAEs of grade 3 or higher in patients with AID were 30% (95% CI, 21% to 41%) with anti-CTLA-4, 17% (CI, 12% to 23%) with anti-PD-1, and 44% (CI, 27% to 62%) with combination therapy; for patients without AID, the incidences were 30% (CI, 27% to 33%) (n = 916), 13% (CI, 12% to 15%) (n = 1540), and 48% (CI, 43% to 53%) (n = 388), respectively. Patients with AID more often discontinued anti- PD-1 treatment because of toxicity than patients without AID (17% [CI, 12% to 23%] vs. 9% [CI, 8% to 11%]). Patients with IBD were more prone to anti-PD-1-induced colitis (6/31 = 19% [CI, 7% to 37%]) than patients with other AIDs (3% [CI, 0% to 6%]) and patients without AID (2% [CI, 2% to 3%]). The objective response rate was similar in patients with versus without AID who were treated with anti-CTLA-4 (10% [CI, 5% to 19%] vs. 16% [CI, 14% to 19%]), anti-PD-1 (40% [CI, 33% to 47%] vs. 44% [CI, 41% to 46%]), or the combination (39% [CI, 20% to 59%] vs. 43% [CI, 38% to 49%]). Survival did not differ between patients with and those without AID (median, 13 months [CI, 10 to 16 months] vs. 14 months [CI, 13 to 15 months]). Limitation: Information was limited on AID severity and immunosuppressive treatment. Conclusion: Response to ICI with anti-CTLA-4, anti-PD-1, or their combination for advanced melanoma and overall incidence of any irAEs of grade 3 or higher were similar in patients with and without preexisting AID. However, severe colitis and toxicity requiring early discontinuation of treatment occurred more frequently among patients with preexisting IBD, warranting close follow-up.