A randomized phase Ib/II study of intermittent androgen deprivation therapy plus nivolumab with or without interleukin-8 blockade in men with hormone-sensitive prostate cancer (MAGIC-8).

Abstract

5082 Background: Immunotherapy has limited efficacy in castration-resistant prostate cancer. Androgen deprivation therapy (ADT) has significant immunomodulatory effects and initially induces a complex immune infiltrate before castration-resistance develops. However, ADT also recruits immunosuppressive myeloid cells to the tumor microenvironment by increasing interleukin-8 (IL-8). We conducted a phase Ib/II clinical trial of immunotherapy plus ADT in men with recurrent castration-sensitive prostate cancer (CSPC). We hypothesized that anti-PD-1 (nivolumab) +/- anti-IL-8 (BMS-986253) given at the time of castration could induce anti-tumor immune responses and decrease disease progression. Methods: MAGIC-8 was a multicenter, phase Ib/II study evaluating nivolumab +/- BMS-986253 combined with a short course of degarelix acetate in patients with recurrent CSPC and rapid PSA doubling time (≤ 12 mos). In the Phase Ib portion, patients were treated with nivolumab (480mg Q4W) for 8 wks followed by nivolumab plus degarelix for an additional 16 wks. In the phase II portion, patients were randomized 1:2 to nivolumab + degarelix (Arm A) versus nivolumab + BMS-986253 (2400mg Q2W) + degarelix (Arm B). The primary endpoints were PSA recurrence at 10 mos following randomization and safety. Key secondary endpoints included biochemical recurrence-free survival (bPFS), time to recovery of testosterone (> 150ng/dl), and bPFS after recovery of testosterone. Results: Between October 16, 2019 and March 9, 2021, 59 patients were enrolled. The first 15 patients were treated on Arm A followed by 1:2 randomization to Arm A (N = 15) versus Arm B (N = 29). Median follow up was 11.6 mos at the data cutoff (1/24/22). Patients treated on Arm A had a significantly lower rate of PSA relapse (17.39%) at 10 mos compared to historical controls (p = < 0.001), including a subgroup of patients (6.67%) with recovery of testosterone and no PSA relapse at > 2 years of follow up. Median time-to-recovery of testosterone was 12.7 mos, median bPFS was 14.0 mos and median bPFS after recovery of testosterone was 5.5 mos. In Arm B, there was no difference in PSA relapse at 10 mos (35%, p = 0.09), median time-to-recovery of testosterone, median bPFS and median bPFS after recovery of testosterone compared to historical controls. Treatment in both arms was well tolerated with a lower rate of grade 3-4 treatment-related adverse events in Arm B compared to Arm A (3.5% vs 12.9%). Conclusions: A short course of ADT plus nivolumab may decrease the rate of PSA relapse and lead to durable long-term responses after recovery of testosterone in a subset of patients. These data support further evaluation of combining nivolumab with ADT in CSPC. Although the addition of BMS-986253 did not improve rate of PSA relapse, we observed significantly less toxicity with the addition of IL-8 inhibition. Clinical trial information: NCT03689699.