Population Pharmacokinetic-Pharmacodynamic Analysis of Weight Loss Efficacy of CP-945,598 in Adult Obese Subjects
Patanjali Ravva (1), Karthik Venkatakrishnan (2), Thomas G Tensfeldt (1), Carol Cronenberger (3) and John D Obourn (4)
(1) Global Pharmacometrics, Pfizer Inc, USA; (2) Millennium Pharmaceuticals, USA; (3) Clinical Pharmacology – Primary Care, Pfizer Inc, USA (4) Clinical Development – Emerging Markets, Pfizer Inc, USA
Objectives: To characterize the time course of weight loss efficacy of CP-945,598 (CP) in obese subjects following multiple oral doses for up to 6 months
Methods: This was a double-blind, placebo- and positive-controlled dose-ranging study designed to assess the efficacy and safety of CP in obese subjects (BMI³27 and <40 kg/m2) with co-morbid conditions. A total of 282 subjects were randomized to receive 5, 15, or 25 mg QD of CP, or 15 mg QD of sibutramine (S) or placebo (P). All subjects followed the same disease management program for obesity. Outpatient visits were scheduled for weight measurements on study Days 14, 28, 56, 84, 112, 140, and 168. Serum samples were collected for concentration analysis on 5 clinic visits.
A nonlinear mixed effects modeling approach was utilized to describe the disease progression and the effects of CP on weight loss. The model was structured as an exponential decrease of body weight (WT) over time (first order rate constant a) from the baseline weight (A) resulting in a maximum fractional weight loss (B). This maximum fractional weight loss was described as the sum of placebo (BPbo) and drug effects, with the latter described as a hyperbolic Emax function of estimated individual steady-state average CP concentration (parameters Emax and ECss,50). In addition, model was comprised of additive inter-individual variance (IIV) on B, exponential IIV on A and a, and an additive residual variance.
WTij = A*exp(η1)*(1-(B+ η2)*(1-e(-α*exp(η3)*tij))) + εij
Results: A total of 211 subjects (86.7% Females), equally balanced across treatments; P (51), 5 mg QD (51), 15 mg QD (52) and 25 mg QD (57) contributed data for the longitudinal analysis. An exponential decay function captured the time course of weight loss. A monotonic relationship was characterized between steady-state average concentration of CP and weight loss efficacy across the dose range studied. Key final model parameter estimates (95% bootstrap CI) were: 0.0275 (0.0125-0.0397) for Bpbo, 0.0824 (0.048-0.260) for Emax, 17.6 (3.54-101) for EC50 (ng/mL) and 0.0825 (0.0345-0.1248) for a (week-1). IIV on B and a was 0.00359, 67% CV, respectively. The residual variance was estimated as 0.923. A visual predictive check indicated adequate model performance.
Conclusions: The model adequately captured the time course of weight change from baseline. Doses of 20-25 mg are predicted to produce a clinically meaningful extent of weight loss (4-5% at 6 months).