A Mechanistic Model for the Effects of A Novel Drug on Glucose, Glucagon and Insulin Applied to Adaptive Phase II Design
J.Z. Peng (1), W.S. Denney (2), R. Liu (3), K. Tsai (3), K.G. Kowalski (4), M.L. Reitman (5), M. Troyer (5), S. Engel (6), J. Stone (1), A.S. Stoch (5), S. Allerheiligen (1)
(1) Modeling & Simulation, Merck & Co., Inc., USA; (2) Clinical PK/PD, Merck & Co., Inc., USA; (3) Early Development Statistics, Merck & Co., Inc., USA; (4) Ann Arbor Pharmacometrics Group, Inc. (A2PG), USA; (5) Clinical Pharmacology, Merck & Co., Inc., USA; (6) Clinical Research, Merck & Co., Inc., USA
Objectives: Drug A was developed for the treatment of Type 2 diabetes (T2D). A mechanistic model was developed to describe Drug A pharmacokinetics and glucagon, insulin and glucose profiles in healthy subjects during a glucagon challenge. The model was adapted for the T2D patient population to assess the need for dose adjustment at the interim analysis of a Phase IIa study.
Methods: Single oral doses of Drug A (0-900 mg) were given to 36 healthy subjects in a Phase I study. Starting from 3, 12 or 24 hr post dose, glucagon, sandostatin and basal insulin were infused for 2 hrs (glucagon challenge). A published model was expanded incorporating drug, glucagon and sandostatin. The model was then modified using steady-state analysis for patients accounting for differences in the PD parameters between healthy subjects and T2D patients. Clinical trial simulations (CTS) were subsequently performed to extrapolate drug effects to T2D patients in a Phase IIa study setting where no glucagon challenge was given. NONMEM and R were used for modeling and NONMEM and SAS were used for CTS.
Results: The model assumed that the glucose production rate (GPROD) was modulated by glucose and glucagon independently. The drug effect was modeled by using an inhibitory Emax model (Imax=0.96 and IC50=13.7 nM) on the ability of glucagon to increase GPROD. In addition, an Emax model (Emax=0.79 and EC50=575 nM) to increase glucagon secretion by the drug was used to account for the increased glucagon concentrations pre-challenge (via compensatory feedback). The model adequately captured the observed profiles of glucose, glucagon and insulin pre- and post-challenge. For CTS (1000 trials), the parameter estimates were adapted using baseline covariate data for the ongoing Phase IIa study in T2D patients and prior knowledge from a lead compound in the same class. Because the model PD output was fasting plasma glucose (FPG), but weighted mean glucose (WMG) was the PD endpoint for the Phase IIa study, a linear model between FPG and WMG was developed using the data from the Diabetes Control and Complications Trial. The CTS results suggest that the dose selection for the Phase IIa study was adequate and that there was no need for a dose adjustment.
Conclusion: A PK/PD model was developed to capture the interplay between glucose, glucagon and insulin. A linear model to correlate FPG to WMG was developed and provided robust predictions to assist with the dose adjustment for the interim analysis.
 Silber HE, Jauslin PM, Frey N, Gieschke R, Simonsson USH, Karlsson MO. An integrated model for glucose and insulin regulation in healthy volunteers and Type 2 diabetic patients following intravenous glucose provocations. Journal of Clinical Pharmacology, 2007;47:1159-1171.
 The Diabetes Control and Complications Trial (DCCT) was a clinical study conducted from 1983 to 1993 funded by the National Institute of Diabetes and Digestive and Kidney Diseases with 1441 Type 1 diabetic patients treated with insulin.