A Joint Concentration-Response Model for ABPM Measurements of Systolic and Diastolic Blood Pressure
Michael Heathman (1), Mary Jane Geiger (2), Amparo de la Peña (1)
(1) Eli Lilly and Company, Indianapolis, Indiana, USA; (2) Regeneron Pharmaceuticals, Inc., Tarrytown, New York, USA
Objectives: To characterize the relationship between the concentration of an experimental drug, LY, and ambulatory blood pressure monitoring (ABPM) measurements of systolic (SBP) and diastolic (DBP) blood pressure.
Methods: A multicenter, randomized, double-blind, parallel-arm, placebo-controlled study was conducted to evaluate the effects of LY on blood pressure using ABPM. ABPM was measured at approximately the maximum LY concentration. Patients (755) were randomized to placebo or 1 of 2 LY treatment arms. ABPM was performed prior to randomization, and at 3 other visits over 26 weeks. Five plasma samples were collected from each patient for determination of LY concentrations.
Separate circadian rhythm (CR) models were developed for SBP and DBP, using data from placebo treated patients. A combination of cosine functions was used to model CR, parameterized in terms of amplitude and phase. Inter-patient and inter-occasion variability were assessed on each parameter. NONMEM 7.2 was used for all analyses.
A 2-compartment model was used to describe the pharmacokinetics (PK) of LY. The CR models for SBP and DBP were each combined with the PK model; exposure-response relationships were evaluated using data from all 3 treatment arms. Demographics, smoking status, hypertensive status, and concomitant medications were evaluated as covariates.
Finally, a linked model for SBP and DBP was developed, incorporating covariance in both inter-patient and inter-occasion variability. This model was qualified using bootstrap and visual predictive checks (VPC). The VPC included SBP and DBP, and the derived variables mean arterial pressure (MAP) and pulse pressure (PP).
Results: The final linked model described the CR of SBP and DBP using 2 cosine functions, with periods of 24 and 12 hours. While separate amplitude parameters were used for SBP and DBP, phase parameters were shared.
SBP decreased with increasing LY concentration, while no relationship was found for DBP. Consistent with physiology, baseline SBP was found to increase with age. Gender influenced baseline DBP, with males having 7% higher values.
Conclusions: The combined concentration-response model described SBP and DBP response to LY very well, based on the VPC. The relationship between SBP and DBP was also well characterized, based on the VPC of MAP and PP. The final model provided an understanding of blood pressure response to LY therapy, and allowed prediction of responses to LY in alternative dosing regimens.
