Population Pharmacokinetic/Pharmacodynamic Modeling of PF-06480605, an Anti-TL1A Antibody, in Healthy Subjects and Ulcerative Colitis Patients
Srividya Neelakantan (1), Steven W Martin (1), Gang Li (2), Kenneth Hung (1), Deepa Elizabeth Chandra (1), Natalie Rath (2), Christopher Banfield (1)
(1) Pfizer Inc., Cambridge, MA (2) Pfizer Inc., Collegeville, PA
Objectives: The tumor necrosis factor(TNF)-like ligand 1A(TL1A)/ Death Receptor 3 (DR3) pathway has been implicated in the regulation of pathogenic helper T lymphocyte (Th) subsets 1, 2, and 17, respectively (Th1, Th2, and Th17), T cells and in natural killer (NK) and natural killer T (NKT) cell responses. The TL1A expression on antigen presenting cells (monocytes, macrophages, dendritic cells) and DR3 expression on effector cells (ILC2, T-cells, NK and NK-T cells) is highly dependent on pro-inflammatory conditions. Significant literature data in nonclinical species and humans implicate TL1A in the pathophysiology of inflammatory bowel disease (IBD). PF-06480605 is a fully human neutralizing antibody against TL1A which is expected to neutralize the binding and subsequent signaling of TL1A to its functional receptor DR3 and is currently being developed for the treatment of IBD.
The objectives of the analysis were to characterize the pharmacokinetics (PK) and pharmacodynamics (PD, serum total sTL1A) of PF-06480605; and identify covariates of interest to explain the inter-individual variability.
Methods: The PK and PD of PF-06480605 were analyzed in two studies: a Phase 1, randomized, double-blind, placebo-controlled, single and multiple dose escalating study in healthy subjects as well as a Phase 2a multicenter, single arm (non-placebo controlled), two-stage study in subjects with moderate to severe Ulcerative Colitis (UC). Both intravenous (IV) and subcutaneous (SC) routes of administration were tested in the Phase 1 study. The Phase 2a study evaluated PF-06480605 500 mg IV administered every 2 weeks for a total of 7 doses. Serum PF-06480605 concentrations and the serum total TL1A data were characterized jointly using a target mediated drug disposition model with Michaelis-Menten (MM) approximation. Covariate analyses were performed to identify intrinsic and extrinsic factors that explain the inter-individual variability in the PK of PF-06480605. Body weight, age, baseline albumin, and gender were explored as potentially influential covariates on the linear clearance (CL) of PF-06480605. Body weight was tested as a covariate to explain the inter-individual variability in the central volume of distribution (V1). Goodness of fit plots, visual predictive check (VPC) and nonparametric bootstrap were performed on the base and final models to evaluate the model performance.
Results: The PK of PF-06480605 after single and multiple IV or SC dosing in healthy subjects and UC patients were adequately described by a 2-compartment model with additional target-mediated clearance with MM approximation. The population PK/PD analysis indicated that PF-06480605 displayed the attributes of a typical IgG monoclonal antibody with a long terminal half-life of ~20 days. The CL was slow with an estimate of 0.00754 L/hr, while the inter-compartmental clearance was 0.00521 L/day. The central and peripheral volumes of distribution were 3.03 L and 1.64 L, respectively.
Body weight was identified as an important covariate on the CL of PF-06480605 with a power exponent of about 0.7.
The results from the VPCs as well as the goodness of fit plots indicate that the population PK/PD model adequately described both the PF-06480605 and total sTL1A concentrations.
Conclusions: The target mediated drug disposition model with MM approximation adequately described the serum PK and total sTL1A following PF-06480605 administration. Covariates were identified that explained the inter-individual variability in PK.