P1‐262: Preliminary population pharmacokinetic modeling of PF‐04360365, a humanized anti‐amyloid monoclonal antibody, in patients with mild‐to‐moderate Alzheimer's disease
Timothy Nicholas, William Knebel, Marc R. Gastonguay, Martin M. Bednar, Clare B. Billing, Jaren W. Landen, James W. Kupiec, Brian Corrigan, Rene Laurencot, Qinying Zhao · Alzheimer s & Dementia · 2009
PF-04360365 is a humanized anti-amyloid monoclonal antibody that recognizes amino acids 33-40 of the beta-amyloid (Aβ) 1-40 peptide and requires a free carboxy terminus for binding. In transgenic mice that overexpress amyloid precursor protein, the murine analog of PF-04360365 has been observed to decrease Aβ levels in the central nervous system and to improve radial arm water maze performance. PF-04360365 is currently undergoing clinical testing in patients with Alzheimer's disease (AD) as a potential therapeutic to reduce brain Aβ burden and improve clinical outcomes. A robust population pharmacokinetic (PK) model at an early stage of development can be critical in helping design more efficient clinical studies. Plasma PK data were obtained from a randomized, double-blind, placebo-controlled, dose-escalation (0.1-10 mg/kg) study in patients with mild-to-moderate AD. Patients received either a single intravenous dose of PF-04360365 (N=26) or placebo (N=11). Plasma drug concentrations were analyzed by ELISA. A population PK model was then developed using non-linear mixed effects modeling. Allometric scaling was implemented using a reference weight of 70 kg. The PK profile of PF-04360365 appeared linear and was best described by a two-compartment model. Diagnostic plots and visual predictive checks for the model indicated a good fit with minimal bias. Predictive checks showed that the model accurately described PF-04360365 exposure across the observed dosing range. Fixed parameters, CL, V1, Q, and V2, were precisely estimated, as seen by low percent standard error (SE) in the range of 3.5% to 10%. Inter-individual variances were estimated with moderate precision (%SE: 35% to 55%). This preliminary model describing the PK profile of PF-04360365 will be refined as more data are collected. Simulated exposure and concentration-time profiles of different dosing regimens based on the model can provide a better understanding of clinical trial designs including optimal doses and dosing frequency.