Zinnia P Parra-Guillen 1, Erika Bågeman 1, Björn Frendéus 1, Ingrid Karlson 1, Ingrid Teige 1, Philipp Zimmermann 1, Andres McAllister 1, Johan E Wallin 1
1 BioInvent International AB (Lund, Sweden)
Objectives
BI-1206 is a fully human monoclonal antibody that targets the only inhibitory Fcγ receptor (FcγRIIB), a regulator of antibody driven immune responses across tumor types [1]. By blocking this pathway, BI-1206 is designed to modulate antibody–receptor interactions that influence therapeutic antibody activity and immune effector responses [2-3]. Ongoing clinical studies in solid tumors and B cell malignancies are evaluating whether FcγRIIB blockade can potentiate responses to established immunotherapies.
In this context, population modelling plays an important role in informing early clinical drug development [4]. The aim of this work is to develop a population pharmacokinetic (PK), receptor occupancy (RO), and platelet count model to quantitatively characterize BI-1206 pharmacology and identify sources of variability, thus supporting dose optimization across diverse clinical settings.
Methods
Data analyzed included PK, RO, and platelet count data from 109 patients receiving BI-1206 across two clinical trials, NCT03571568 and NCT04219254, here referred to as the 02 and 03 study respectively. Briefly, in the 02 study, BI-1206 was administered at doses ranging from 30-200 mg intravenously (IV) or 150-300 mg subcutaneously (SC) in combination with rituximab, with or without acalabrutinib, in subjects with Indolent B-cell Non-Hodgins Lymphoma (n=55). Patients received a first 28-day induction cycle comprising 3-4 weekly doses of BI-1206 and 4 doses of rituximab. Some patients received an optional second induction cycle consisting of 3 weekly BI-1206/rituximab doses. Maintenance treatment consisted of BI-1206/rituximab Q8W for up to 1 year. Subjects receiving acalabrutinib received it orally once every 12 h throughout treatment. For the 03 study, BI-1206 was administered to patients with advanced solid tumors (n=54) once or twice per 21-day cycle, depending on the cohort, for up to 35 cycles or 2 years from the 1st dose of BI-1206. Doses ranged from 0.5 to 2 mg/kg (IV) and from 150 to 300 mg (SC). Pembrolizumab was administered at its standard flat dose of 200 mg once per cycle on day 1.
Model building was performed in a sequential and integrative manner: first developing the PK model, then adding RO data. Estimated individual model parameters from the previous step were then used to model the platelet count time course, using the IPP method [5]. Model selection and evaluation were performed using state-of-the-art modelling techniques [6]. Data below the limit of quantification were handled using M3 method [7]. Analyses were performed in NONMEM 7.5, using FOCEI and LAPLACE algorithm with the support of PsN, R and RStudio for data preparation and postprocessing of results.
Results
BI-1206 disposition was well described by a two-compartment model exhibiting restricted total volume of distribution ( ~ 5L , IIV =14.6 %), and both linear clearance (0.814 L/d, IIV = 56 %) and target-mediated drug disposition, including receptor dynamics and complex internalization. After SC administration, a delayed nonlinear absorption was estimated with a bioavailability of 62 %. At this level, differences across studies were only observed at the binding constant, which exhibited a value ~70% lower in the 02 study. No impact of combination therapies was identified. Receptor occupancy was better described assuming a slow reversible binding model [8] with sustained occupancy predicted over the dosing interval at doses above 150 mg QW (SC) or 100 mg QW (IV).
To describe platelet count dynamics, the Friberg model – accounting for proliferation at bone marrow, transit and circulation of platelets in homeostasis- was used as a starting point. Based on prior preclinical knowledge, drug effect was incorporated promoting a reversible binding to circulating cells (potentially reflecting transient and reversible platelet adherence to monocytes). This model provided an overall better description of the data compared to different cytotoxic effects, and identified differences across studies at the binding process, potentially due to differences across the studied diseases (hematological versus solid cancers).
Conclusions
A modelling framework quantitatively characterizing the impact of BI-1206 exposure on pharmacodynamic (RO) as well as safety (platelet counts) markers has been successfully developed, integrating information across different doses, routes of administration and clinical trials. The model represents an important tool to guide dose selection and study design for upcoming clinical trials.
References:
[1] Nimmerjahn et al. Nat. Rev. Immunol, 8, 34–47 (2008)
[2] Rankin et al. Blood, 108:2384-91 (2006)
[3] Teige et al. Front Immunol, 10:481 (2019)
[4] Marshall et al. CPT Pharmacometrics Syst Pharmacol, 8 (2019)
[5] Zhang et al. J Pharmacokinet Pharmacodyn, 30:387-404 (2003)
[6] Nguyen et al. CPT Pharmacometrics Syst Pharmacol, 6:87-109 (2017)
[7] Bergstrand and Karlsson. AAPS J, 11:371-380 (2009)
[8] Ren et al. J Pharmacokinet Pharmacodyn, 49:493-510 (2022)
Reference: PAGE 34 (2026) Abstr 12251 [www.page-meeting.org/?abstract=12251]
Poster: Drug/Disease Modelling - Oncology