Nivolumab dose selection: challenges, opportunities, and lessons learned for cancer immunotherapy

Abstract

Background: Immuno-oncology (I-O) therapies target the host immune system, providing the potential to choose a uniform dose and schedule across tumor types. However, dose selection for I-O agents usually occurs early in clinical development and is typically based on tumor response, which may not fully represent the potential for improved overall survival. Here, we describe an integrated approach which incorporates clinical safety and efficacy data with data obtained from analyses of dose-/exposure-response (D-R/E-R) relationships, used to select a monotherapy dose for nivolumab, a programmed death-1 inhibitor, in clinical studies of different tumor types.

Methods: Dose was selected based on anti-tumor activity and safety data from a large phase 1b, open-label, dose-escalation study of nivolumab at doses ranging from 0.1 to 10 mg/kg administered every 2 weeks (Q2W) in 306 patients with advanced malignancies, and quantitative analyses were performed to characterize D-R/E-R relationships for pharmacodynamic, safety, and efficacy endpoints.

Results: A maximum tolerated dose for nivolumab was not identified, and the safety profile was similar across tumor types and dose levels (0.1-10 mg/kg). Objective response rates (ORRs) were similar across doses in melanoma and renal cell carcinoma (RCC), while higher ORRs were observed in non-small cell lung cancer (NSCLC) at 3 mg/kg and 10 mg/kg versus 1 mg/kg. Peripheral receptor occupancy was saturated at doses ≥ 0.3 mg/kg. In D-R/E-R analyses, a positive dose-dependent objective response trend was observed for each tumor type, but appeared to plateau at nivolumab doses of ≥ 1 mg/kg for melanoma and RCC, and at ≥ 3 mg/kg for NSCLC. Although there was no apparent relationship between tumor shrinkage rate and exposure, tumor progression rate appeared to decrease with increasing exposure up to a dose of 3 mg/kg Q2W for NSCLC.

Conclusions: Nivolumab monotherapy at 3 mg/kg Q2W provides unified dosing across tumor types. This dose and schedule has been validated in several phase II/III studies in which overall survival was an endpoint. Integrating D-R/E-R relationships with efficacy data and a safety profile that is unique to I-O therapy is a rational approach for dose selection of these agents.

Keywords: Dose selection; Immunotherapy; Melanoma; Nivolumab; Non-small cell lung cancer; Renal cell carcinoma.

Fig. 1

Exposure-response efficacy analysis of nivolumab by tumor type. Vertical lines represent 90 % prediction intervals for each dose level. CI = confidence interval; Cminss = steady-state nivolumab trough concentration; MEL = melanoma; NSCLC = non-small cell lung cancer; Obs Prob = observed probability; OR = objective response; Pred Prob = predicted probability; RCC = renal cell carcinoma

Fig. 2

Integrated dose-response for treatment-related grade ≥3 AEs and AEs leading to discontinuation. AEs = adverse events

Fig. 3

Peripheral a PD-1 occupancy and b receptor occupancy (RO) of patients treated with nivolumab. MEL = melanoma; PD-1 = programmed death–1; pts = patients

Fig. 4

Exposure-response for efficacy by dose level in melanoma. a Probability of OR vs Cminss, overall and by dose. b Probability of OR vs CL, overall and by dose. CL = clearance; Cminss = steady-state trough concentration; OR = objective response

Fig. 5

Relationship between a exposure and tumor shrinkage rate, and b exposure and tumor progression rate. Cminss = steady-state trough concentration; MEL = melanoma; NSCLC = non-small cell lung cancer; RCC = renal cell carcinoma