C. Falcoz (2), M. Marchand (2), P. Chanu (2), S. Bolze (1)
(1) Poxel SA, Lyon, France; (2) Pharsight Consulting Services™, a division of Certara™, St Louis, MO, USA
Objectives: To set up early on a PK and PKPD framework based on clinical endpoints (fasting plasma glucose (FPG), glycosylated hemoglobin (HbA1c)) in T2DM for imeglimin, the first in a new, tetrahydrotriazine-containing class of oral antidiabetic agents, the Glimins.
Methods: Phase IIa study 1 (39 subjects; 1000 mg BID or 2000 mg OD for 4 weeks) and study 2 (92 subjects; placebo, 500, or 1500 mg BID for 8 weeks) were used for the population (Pop) PK model of imeglimin (99 subjects). Study 2 was selected for the Pop PKPD indirect response (IR) models for FPG with imeglimin inhibiting glucose production, and for HbA1c production from FPG [1,2]. Models developed in NONMEM 7.2 were qualified through Visual and Posterior predictive checks (Trial Simulator v.2.2.1).
Results: PopPK. The model selected for PKPD purposes was a 3-compartment model with zero-order absorption, dose influence on bioavailability F and inter-individual variability (IIV) on CL/F, V1/F and V2/F. Concentrations were well described and parameters well estimated with standard errors (RSE) <= 34%. Pop PKPD. HbA1c data were few (baseline, end of treatment (EoT)) and a joint FPG-HbA1c model was therefore not selected. Parameters of the FPG model were well estimated (RSE <=27%); individual profiles were reasonably well captured considering day-to-day IIV, including the one week washout. In the HbA1c model, HbA1c degradation rate was fixed to a value estimated with denser data [1]; other parameters were well estimated (RSE <= 20%). Model qualification. Predictive checks indicated adequate performance of all models. Observations were within the 90% prediction intervals. Median change from baseline at EoT in the placebo, 500 mg BID and 1500 mg BID groups was predicted respectively at: (i) 0.76, 0.03 and -0.71 mmol/L for FPG (vs. observed 0.55, 0.20 and -0.90 medians); and (ii) 0.5, 0.2 and -0.1 % for HbA1c (vs. observed 0.2, 0.1 and -0.1 medians).
Conclusions: PK data from two Phase IIa monotherapy studies were combined. Data from study 1 were essential for PK model development. IR models could be used to characterize changes in FPG and HbA1c over 8 weeks of treatment. Model development with early limited data should already prove useful in guiding biopharmaceutical development and the design of future imeglimin studies.
References:
[1] Rohatagi S et al. (2008). Model-based development of a PPARgamma agonist, rivoglitazone, to aid dose selection and optimize clinical trial designs. J Clin Pharmacol 48: 1420-9.
[2] Naik H et al. (2013). Pharmacometric approaches to guide dose selection of the novel GPR40 agonist TAK-875 in subjects with type 2 diabetes mellitus. CPT: Pharmacometrics & Systems Pharmacology 2, e22; doi:10.1038/psp.2012.23; advance online publication 9 January 2013.
Reference: PAGE 22 (2013) Abstr 2720 [www.page-meeting.org/?abstract=2720]
Poster: Endocrine