A randomized phase II study of ipilimumab at 3 (ipi3) or 10 mg/kg (ipi10) alone or in combination with high dose interferon-alfa (HDI) in advanced melanoma (E3611).

Abstract

9542 Background: Interferon-α (IFN) favors a Th1 shift in immunity. Combining CTLA4 blockade with IFN may downregulate CTLA4 suppressive elements. Prior data from a phase II of tremelimumab and HDI showed promising efficacy supporting the current study. Methods: E3611 had a 2x2 factorial design (A: ipi10 + HDI; B: ipi10; C: ipi3 + HDI; D: ipi3) to evaluate (i) no HDI vs. HDI (across ipi doses) and (ii) ipi3 vs. ipi10 (across HDI status). We hypothesized that median progression free survival (PFS) would improve from 3 to 6 months (mos) with HDI vs. no HDI and with ipi10 vs. ipi3. Based on the log-rank test for 80 patients (pts) these comparisons would have 82% power at 2-sided type I error of 0.10. Results: Median follow up 26.4 mos. PFS and overall survival (OS) (eligible and treated pts; N = 80: 18 III/M1a, 24 M1b, 38 M1c) are shown in Table 1. There were no significant differences in PFS or OS when evaluating HDI vs. no HDI or ipi10 vs. ipi3 (Table 1). Response (RECIST) among response evaluable pts (and 4 pts with early death) (N = 76) is shown in Table 1. Stable disease (SD) in 7 (A), 6 (B), 8 (C) and 6 (D). Adverse events (AEs) were consistent with the toxicity profiles of ipi and HDI and included 3 grade 5 AEs considered at least possibly related: 1 in A (suicide), 1 in B (lung infection and hemorrhage) and 1 in C (adult respiratory distress syndrome). One patient in B died of gastrointestinal bleed and cardiac arrest while on corticosteroids to treat temporal arteritis and vision loss. Conclusions: Within the limitations of the sample size, there were no significant differences in PFS with HDI vs. no HDI or ipi10 vs. ipi3. Response and PFS with ipi10 were superior to historical controls and similar to the combination. Correlative studies are ongoing. Clinical trial information: NCT01708941. [Table: see text]