Kinjal Sanghavi,1 Pradeep Vuppala,1 Vijay Ivaturi,2 Lora Hamuro,1 Amit Roy,1 Satyendra Suryawanshi1
1Bristol-Myers Squibb, Princeton, NJ, USA; 2University of Maryland School of Pharmacy, Baltimore, MD, USA
Objectives: The purpose of this analysis was to assess the benefit–risk profiles of flat-dose regimens of nivolumab 240 mg every 2 weeks (Q2W) and 480 mg every 4 weeks (Q4W) relative to 3 mg/kg Q2W for the adjuvant treatment of melanoma. We compared key summary measures of nivolumab exposure, and characterized exposure–response (E–R) relationships of efficacy (recurrence-free survival [RFS], and distant metastasis-free survival [DMFS]) and safety (Grade 2+ immune mediated adverse events [Gr2+ IMAEs], and Grade 3+ adverse events [Gr3+ AEs]). We evaluated the impact of differences in exposure on the benefit–risk profile of 240 mg Q2W and 480 mg Q4W dosing relative to 3 mg/kg Q2W (the dose tested in the pivotal study CheckMate 238 [1]).
Methods: Previously developed population pharmacokinetic (PPK) models [2, 3] were used to predict measures of nivolumab exposure (peak, trough, and time-averaged concentrations after the first dose, over the first 28 days, and at steady state) for the 240 mg Q2W, 480 mg Q4W, and 3 mg/kg Q2W dosing regimens”. Nivolumab clearance was time-invariant for adjuvant melanoma, but time-varying across different advanced solid tumors and hematologic malignancies. The E-R relationship for efficacy (RFS and DMFS) was characterized with data from CheckMate 238; and the E-R relationship for safety (Gr3+ AEs and Gr2+ IMAEs) with data from CheckMate 238 and 13 additional studies in patients with advanced cancers (including, melanoma, colorectal cancer, squamous and non-squamous non-small cell lung cancer, renal cell carcinoma, urothelial cancer, squamous cell carcinoma of head and neck, and classic Hodgkin’s lymphoma). The E–R relationships for efficacy and safety were characterized by Cox proportional-hazards models, adjusted for the potential modulatory effects of covariates. E–R relationship for efficacy was characterized with respect to trough concentration on Days 14 and 28 (Cmind14 and Cmind28), and E-R of safety with respect to peak concentration after 1st dose (Cmax1). The potential impacts of a change in dosing regimen on efficacy and safety were assessed by predicting the probabilities of RFS and DMFS at 1- and 2-years, and Gr2+ IMAEs and Gr3+ AEs at 12 months, respectively.
Results: Predicted nivolumab exposures in adjuvant melanoma following 240 mg Q2W dosing were similar to those predicted for 3 mg/kg Q2W (<2% difference in geometric means [GM] of all key summary measures of nivolumab exposures). In contrast, predicted peak and time-averaged exposures with 480 mg Q4W dosing were higher, and trough concentrations were lower, relative to 3 mg/kg Q2W. The greatest increase in exposure with 480 mg Q4W was for Cmax1 (~103% higher GM), whereas the greatest decrease was for Cmind28 (~15% lower GM).
Analysis of E–R relationships showed that nivolumab exposure (Cmind14 and Cmind28) was not a significant predictor of risk for RFS or DMFS in adjuvant melanoma, indicating a flat E–R relationship for efficacy over the exposure range. Predicted 1- and 2-year probabilities of RFS and DMFS for the nivolumab 240 mg Q2W and 480 mg Q4W regimens were similar to those for the 3 mg/kg Q2W regimen (<1% difference). Additionally, the differences in the risk of recurrence, occurrence of distant metastases, and death compared to the ipilimumab comparator arm in CheckMate 238 were similar.
The risk of Gr2+ IMAEs was greater in patients with higher Cmax1, but the difference in the probability of experiencing one or more Gr2+ IMAEs was small for both 240 mg Q2W and 480 mg Q4W regimens relative to 3 mg/kg Q2W (<1% and ≤3% difference over 12 months, respectively). The risk of Gr3+ AEs was not significantly associated with Cmax1. Moreover, the difference in the probability of experiencing one or more Gr3+ IMAEs was also small for both 240 mg Q2W and 480 mg Q4W relative to 3 mg/kg Q2W (<1% and ≤ 2% difference over 12 months, respectively).
Conclusion: The benefit–risk profiles of nivolumab 240 mg Q2W and 480 mg Q4W are predicted to be similar to that of the 3 mg/kg Q2W regimen for the adjuvant treatment of melanoma.
References:
[1] Weber, J, et al. New England Journal of Medicine 2017;377:1824–1835. https://doi.org/10.1056/NEJMoa1709030
[2] Bajaj, G, et al. CPT: Pharmacometrics and Systems Pharmacology 2016;4:58–66. https://doi.org/10.1002/psp4.12143
[3] Hamuro, L, et al. Clinical Pharmacology and Therapeutics 2019;106(5):1018–1027. https://doi.org/10.1002/cpt.1502
Reference: PAGE () Abstr 9410 [www.page-meeting.org/?abstract=9410]
Poster: Drug/Disease Modelling - Oncology