PI-073

X. Wang, G. Bajaj, Y. Feng, M. Gupta, S. Agrawal, A. Yang, J. Park, A. Roy; Bristol-Myers Squibb, Princeton, NJ

BACKGROUND: Nivolumab is a fully human IgG4 monoclonal antibody that selectively binds to the programmed death-1 receptor to promote antitumor immune responses. Nivolumab is currently being developed for the treatment of patients with advanced melanoma who have progressed post anti-CTLA-4 therapy. This analysis aimed to examine the benefit-risk assessment of nivolumab for this patient population by characterizing the relationship between nivolumab exposure and efficacy and safety.
METHODS: The E-R analyses used data from a randomized phase III study of nivolumab in advanced melanoma patients with available measures of nivolumab exposure. The E-R of efficacy was characterized by a logistic-regression model relating nivolumab exposure (trough concentration after dose 1 [Cmin1]) to the probability of achieving an objective response (Pr[OR]; per RECIST 1.1), assessed by Independent Radiology Review Committee (N=115). The E-R of safety was characterized by two separate Cox proportional hazards models relating nivolumab exposure (average steady-state concentration [Cavg ss]) to the hazard of Grade 3+ drug-related adverse events (DR AEs) and AEs leading to discontinuation or death (AE-DC) (N=230).
RESULTS: Cmin1 produced following administration of nivolumab 3 mg/kg every 2 weeks (Q2W) was not a significant predictor of Pr(OR) in subjects with advanced melanoma. Pr(OR) decreased with increasing baseline tumor burden (odds ratio [OR], 0.425; 95% CI, 0.233-0.773 for every unit increase). Nivolumab Cavg ss at 3 mg/kg Q2W did not have a significant effect on the hazard of Grade 3+ DR-AEs or AE-DC (95% CI of both ORs included 1).
CONCLUSION: The nivolumab E-R relationship of both efficacy and safety is relatively flat over the exposure range produced by a dose of 3 mg/kg Q2W.