Associate Director Gilead Sciences, Inc., United States
Disclosure(s):
Sonoko Kawakatsu: No relevant disclosure to display
Objectives: Zimberelimab (Zim) is a fully human IgG4 monoclonal antibody targeting the PD-1 immune checkpoint. Zim is being developed in multiple oncology indications administered as monotherapy and in combination with other therapeutics. The objectives of this analysis were to develop a population pharmacokinetics (PK) model of zim in patients with solid tumors, and to evaluate the effects of intrinsic and extrinsic factors on zim PK.
Methods: A previously developed model [1] was used as a starting point and updated with data from studies that included treatment regimens relevant to gastric, gastroesophageal junction, or esophageal adenocarcinoma (GC/GEJC/EAC) and non-small cell lung cancer (NSCLC) indications. Data were pooled from three Phase 1 studies in solid tumors, four Phase 1 and 2 studies in NSCLC, and one Phase 2 study in GC/GEJC/EAC. Preliminary data were used for ongoing Phase 2 studies ARC-10 (NSCLC), EDGE-Lung (NSCLC), and EDGE-Gastric (GC/GEJC/EAC). The PK analysis dataset included 6559 PK observations from 738 patients treated with zim doses from 80 mg to 960 mg. Patients were treated with zim administered intravenously as monotherapy or in combination with other agents including domvanalimab (dom) and/or chemotherapy. The population PK analysis was conducted in NONMEM v7.5 [2]. Population and individual model parameters were estimated using the stochastic approximation expectation maximization method followed by Monte Carlo importance sampling. A full covariate modeling approach was implemented to evaluate covariate effects of clinical interest including demographic, laboratory, and disease-related variables.
Results: Zim PK was characterized by a two-compartment model with time-varying clearance. For a typical 63 kg white male participant with GC/GEJC/EAC, albumin of 38 g/L, sum of lesion diameter of 61 mm, and PD-L1 expression < 50%, the parameter estimates (95% CI) for baseline clearance (CL) and central volume were 0.251 L/day (0.229, 0.276) and 3.49 L (3.33, 3.65), respectively. The effects of body weight were included using allometric scaling with fixed coefficients (0.75 for clearances, 1 for volumes). Covariates of interest tested in the model included tumor type (GC/GEJC/EAC vs NSCLC vs other solid tumors) and treatment regimen (monotherapy vs zim and dom vs zim and chemo vs other). Forest plots indicated tumor type and treatment regimen did not have clinically meaningful effects on zim PK. The effect of administration (zim/dom coadministration vs sequential) was examined using post-hoc estimates and did not have a clinically meaningful effect on zim PK.
Conclusions: A population PK model with a time-varying CL was adequate to describe zim PK in patients with solid tumors. The PK characteristics of zim were consistent with those of other PD-1 checkpoint inhibitors. Tumor type or combination regimen did not have an impact on zim PK, supporting the same dosing regimen can be applied across different tumor types and treatments. Coadministration of zim with dom, as opposed to sequential administration, did not impact zim PK. This administration approach can enhance patient convenience and reduce clinic time.
Citations: [1] Yu X., Chaturvedula A., Jin L., Wu H., Foster P., Colburn D., Criste R., Agoram B. Population Pharmacokinetics of Zimberelimab (AB122) and Dose Justification by Model Informed Drug Development (MIDD) Approach. ASCPT Annual Meeting 2022. [2] Beal S.L., Sheiner L.B., Boeckmann A.J. & Bauer R.J. (Eds). NONMEM 7.4 users guides (ICON plc, Gaithersburg, MD, 1989. –2018).
Keywords: Zimberelimab, solid tumor, population PK model
