Risk and predictors of ipilimumab-associated cardiac adverse events among patients treated for melanoma: A national cohort analysis.

Abstract

e14592 Background: Ipilimumab is a CTLA-4 inhibitor widely used to treat advanced melanoma. While ipilimumab-induced cardiac immune-related adverse events (irAE) have been reported, there is a paucity of data from large cohorts. We investigated the risk and predictors of ipilimumab-associated cardiac irAE in a national cohort of patients with cutaneous melanoma. Methods: Using SEER-Medicare linked data, we compared the risk of cardiac irAE between patients treated with ipilimumab with or without concomitant treatment for cutaneous melanoma and controls not treated with ipilimumab following a primary diagnosis of cutaneous melanoma. We excluded patients ≤65 years and patients with cardiac comorbidities diagnosed within one year prior to the initiation of melanoma treatment. The primary endpoint was the incidence of at least one cardiac irAE after ipilimumab initiation including acute pericarditis, myocarditis, cardiomyopathy, conduction disorders, cardiac dysthymias, acute heart failure, and takotsubo syndrome. To estimate the risk of cardiac irAE, we conducted a multivariable competing-risk analysis adjusting for death of any cause within one year of treatment as a competing event. Then, we constructed a stepwise logistic regression to assess the predictors of having at least one cardiac irAE within one year of ipilimumab initiation. Subgroup analysis was conducted among patients who received ipilimumab only. The models were adjusted for patient demographics, disease stage, Charlson comorbidity index (CCI), history of hypertension, autoimmune disease, end stage renal disease (ESRD), chronic anticoagulant, and steroid use. Results: The cohort included 715 patients treated with ipilimumab and 22,070 controls. In the ipilimumab arm, 23.4% had metastatic disease, 9.5% had a history of autoimmune disease, and 2.2% had CCI≥2. The incidence rates of cardiac irAE among patients who received ipilimumab and among the control group were 23.3 and 13.6 per 1,000 person-years, respectively. We found that patients who received ipilimumab had a higher risk of cardiac irAE compared to controls (adjusted hazard ratio 1.87; 95%CI 1.50-2.32; p < 0.001). In addition to ipilimumab treatment, other predictors of cardiac irAE included male gender, older age, patients with metastatic disease, history of autoimmune disease, hypertension, ESRD, anticoagulant use, and CCI≥2. The predictors of cardiac irAE were also consistent in the subgroup analysis of patients who received ipilimumab only. Conclusions: Patients who received ipilimumab with or without concomitant treatment for cutaneous melanoma had a higher risk for cardiac irAE. Predictors of cardiac irAE help tailor therapy according to patients’ risk profiles.