Immune-related adverse events in cancer patients being treated with immune checkpoint inhibitors.

ABSTRACT

INTRODUCTION: With the increased use of immune checkpoint inhibitors (ICI), it is essential to improve our understanding of immune-related adverse events (irAE). To date, most studies describing irAE have been performed in clinical trial populations, which may not be an accurate description of irAE in real-world populations. Also, identification of patients at increased risk of irAE is needed as early recognition may improve irAE outcomes. METHODS: We performed a retrospective analysis of patients who received an ICI between January 2014 and October 2018 at a single institution (Tufts Medical Center). Each patient was followed for up to 12 months for the outcome of a physician-reported irAE. Kaplan-Meier curves were created for the time to development and resolution of initial irAE. A Cox proportional hazards model was created to evaluate whether the following variables were independent predictors of an initial irAE age ≥65 years, female sex, non-Caucasian race, radiation in previous 6 months, current smoking status, melanoma, and combination ICI (ipilimumab and nivolumab). RESULTS: Of 131 patients followed, 57 patients (43.5%) developed at least one irAE at a median of 250 days (95% confidence interval (CI) 132 days-not estimable). The most common irAE included dermatitis, thyroid dysfunction, and pneumonitis. Nearly two-thirds of patients with an irAE had ICI therapy withdrawn, and nearly 60% had immunosuppression initiated. In multivariable analysis, we found a significant association between irAE development and age ≥65 years hazard ratio (HR) 1.80, 95% CI (1.03-3.14) and current smoking status (HR 2.26, 95% CI 1.06-4.82). DISCUSSION: We detected a high rate of irAE and that irAE and subsequent management can be clinically burdensome in this patient population. While further studies are needed to validate these findings, this study provides insights into the magnitude, time course, management of, and possible predictors of irAE in a real-world setting.