Immune checkpoint inhibition in sepsis: a Phase 1b randomized study to evaluate the safety, tolerability, pharmacokinetics, and pharmacodynamics of nivolumab

Purpose Sepsis-associated immunosuppression increases hospital-acquired infection and viral reactivation risk. A key underlying mechanism is programmed cell death protein-1 (PD-1)-mediated T-cell function impairment. This is one of the first clinical safety and pharmacokinetics (PK) assessments of the anti-PD-1 antibody nivolumab and its effect on immune biomarkers in sepsis. Methods Randomized, double-blind, parallel-group, Phase 1b study in 31 adults at 10 US hospital ICUs with sepsis diagnosed >= 24 h before study treatment, >= 1 organ dysfunction, and absolute lymphocyte count <= 1.1 x 10(3) cells/mu L. Participants received one nivolumab dose [480 mg (n = 15) or 960 mg (n = 16)]; follow-up was 90 days. Primary endpoints were safety and PK parameters. Results Twelve deaths occurred [n = 6 per study arm; 40% (480 mg) and 37.5% (960 mg)]. Serious AEs occurred in eight participants [n = 1, 6.7% (480 mg); n = 7, 43.8% (960 mg)]. AEs considered by the investigator to be possibly drug-related and immune-mediated occurred in five participants [n = 2, 13.3% (480 mg); n = 3, 18.8% (960 mg)]. Mean +/- SD terminal half-life was 14.7 +/- 5.3 (480 mg) and 15.8 +/- 7.9 (960 mg) days. All participants maintained > 90% receptor occupancy (RO) 28 days post-infusion. Median (Q1, Q3) mHLA-DR levels increased to 11,531 (6528, 19,495) and 11,449 (6225, 16,698) mAbs/cell in the 480- and 960-mg arms by day 14, respectively. Pro-inflammatory cytokine levels did not increase. Conclusions In this sepsis population, nivolumab administration did not result in unexpected safety findings or indicate any 'cytokine storm'. The PK profile maintained RO > 90% for >= 28 days. Further efficacy and safety studies are warranted.