Immune checkpoint inhibitor infusion times and clinical outcomes in patients with melanoma

Abstract

ABSTRACT Background Circadian rhythms impact immune function; a previous study demonstrated that immunotherapy treatment times taking place later in the day correlated with poorer outcomes in patients with melanoma. However, this finding has not been replicated, and other infusion timing schemas are unexplored. The objective of this retrospective, cohort study was to determine if the time of immunotherapy infusion affects outcomes. Materials and Methods Five hundred and sixteen participants age ≥18 years diagnosed with cutaneous, acral, mucosal, or unknown primary melanoma treated with >1 infusion of nivolumab, pembrolizumab, or combination ipilimumab/PD-1 inhibitors were included. Response rate, toxicity rate, overall survival (OS), and progression-free survival (PFS) were determined based on infusion timing. Results Patients with ≥1 late infusion (after 4 pm) among their first 4 infusions had slightly poorer objective response rate compared with only pre-4 pm infusions (39.7% vs 44.5%), but no significant associations with late infusions and PFS and OS (P = .23, .93, respectively). Multivariable analyses showed no statistically significant association with outcomes for patients with any post-4 pm infusion among the first 4; median infusion time was also not associated with outcomes. However, considering all infusion times, we found inferior PFS (median 10.6 vs 38.9 months, P < .0001), and numerically inferior OS (median 54.6 vs 81.2 months, P = .19) in patients with ≥20% late infusions. Multivariable models had similarly inferior response and PFS for patients with ≥20% late infusions, and later median infusion times were associated with inferior response, PFS, and OS. Conclusions Late immunotherapy infusion times were associated with inferior outcomes when considering all infusions, but not when considering initial (first 4) infusions.