PI-082 - POPULATION PHARMACOKINETIC (PK)-PHARMACODYNAMIC (PD) MODELING OF ANUMIGILIMAB, A MONOCLONAL ANTIBODY THAT BLOCKS G-CSF BINDING TO THE G-CSF RECEPTOR

S. Kathman¹, S. Nuthalapati¹, F. Glassman¹, B. De Miguel Lillo¹, D. Kormann¹, B. Wang², P. Nandy¹; ¹CSL Behring, King of Prussia, Pa, USA, ²Amador Bioscience Inc, Pleasanton, Amador Bioscience Inc, Pleasanton, CA, USA.

Background: Anumigilimab (CSL324) is a novel, fully human mAb that binds to granulocyte colony stimulating factor (G-CSF) receptor, blocks G-CSF binding and inhibits G-CSF-mediated survival, activation, and enhanced migration of neutrophils. The purpose of this analysis was to develop an integrated population pharmacokinetic (PPK) and absolute neutrophil count (ANC) response model to characterize the concentration-time profile of Anumigilimab and its effect on ANC in healthy subjects and hidradenitis suppurativa (HS) or palmoplantar pustulosis (PPP) patients.

Methods: Anumigilimab concentration-time data (1120 observations) and longitudinal ANC data (1880 observations) from three clinical studies (n=87 subjects), including two studies in healthy subjects and one study in patients with HS or PPP, were pooled and analyzed using a population approach. The predictive performances of PK and PD models were assessed by goodness-of-fit diagnostics and prediction-corrected visual predictive check (pcVPC).

Results: Anumigilimab PK was adequately described by a two-compartment disposition model with parallel linear and non-linear elimination pathways, and first order absorption for SC administration including a lag time. The effect of weight on Clearance and Volume of Distribution for the central compartment were found to be statistically significant. A combined additive and proportional error model was used to describe residual variability. The effect of anumigilimab on ANC was characterized based on a semi-mechanistic model of chemotherapy-induced myelosuppression. The model showed that anumigilimab decreased ANC with IC50 in low nM range for proliferating cells with lower IC50 for maturing cells over proliferating cells. Maximum decreases from baseline for maturing and proliferating cells resulted in clinically meaningful effects.

Based on simulation results, SC doses were identified that matched exposures of tested IV doses while decreasing neutropenia risk. The model predicted that the neutropenia risk would be clinically manageable at several pharmacologically active doses.

Conclusion: The PPK-ANC model provided quantitative insight into the dose-response relationship of the drug-induced decline in ANC. The model is adequate for determining doses and regimen for future clinical trials that reduce the risk of neutropenia.