M.J. van Esdonk (1,2), I. Lindeman (1), P. Okkerse (1), M.L. de Kam (1), J. Stevens (3), G.J. Groeneveld (1)
(1) Centre for Human Drug Research, Leiden, The Netherlands, (2) Division of Systems Biomedicine and Pharmacology, Leiden Academic Centre for Drug Research, Leiden University, Leiden, The Netherlands (3) University of Groningen, University Medical Center Groningen, Department of Clinical Pharmacy and Pharmacology, Groningen, The Netherlands
Introduction:
Pain models are an objective way to study analgesic effects in addition to the commonly used pain questionnaires in clinical trials. These pain models can be performed, multiple times a day, to investigate analgesic effects and to establish the concentration-effect relationship of novel analgesics. A previous study investigated analgesic effects using a battery of multi-modal pain models [1] for six known compounds, following intravenous or oral administration [2]. A mixed model analysis of variance showed significant analgesic effects on four pain models after oral administration of pregabalin. However, the investigation and quantification of the concentration-effect relationships would provide more information on drug response and the sources of variability in endpoints, compared to the results of statistical testing alone. Therefore, the aim of the current study was to quantify the concentration-effect relationship of pregabalin on the previously identified significant effects on pain thresholds in the cold pressor, electrical stimulation, pressure and contact heat models using population non-linear mixed effects pharmacokinetic/pharmacodynamic (PK/PD) modeling.
Methods:
A single oral dose of 300 mg pregabalin was administered to 16 healthy subjects (8 male and 8 female) in a placebo controlled, randomized, cross-over fashion. On each occasion, a battery of pain models was performed 10 times, up to 10 hours after dosing. Population PK model development evaluated the performance of 1-, 2- and 3-compartment models with (non-)linear elimination kinetics. Lag time and transit compartments were explored to describe the absorption phase of pregabalin. Population PD model development evaluated direct and indirect (turnover) effect models in which a linear- or maximal effect relationship (Emax) was tested for the concentration-effect relationship of pregabalin. NLME modeling was performed using NONMEM V7.3 [3].
Results:
The PK of oral pregabalin was best described with a 1-compartment model with lag time, linear absorption and linear elimination. The use of transit compartments for the absorption phase was not superior. Significant inter-individual variability (IIV) was identified on all population PK parameters. The maximum relative standard error (RSE) was 42.4%. The inclusion of weight dependent changes on the volume of distribution and clearance significantly lowered the objective function value (OFV) with 22.4 points.
The cold pressor and electrical stimulation measurements showed a high day-to-day variability within individual subjects. Therefore, between occasion variability was included in both models at the start of model development. For the cold pressor, 148 placebo and 143 pregabalin treated measurements were available for model development. The placebo data were best described using a turnover compartment. A learning effect between all visits, tolerance over time, or circadian rhythmicity could not be identified. A linear relationship between the pregabalin concentrations and the kin best described the concentration-effect relationship (ΔOFV = -75). All parameters were estimated with low RSE’s (<30%).
For the electrical stimulation, 160 placebo, 153 treated measurements were available for model development. The use of a turnover compartment with an Emax effect gave a significant reduction in the OFV (ΔOFV = -91.5). However, a low EC50 was estimated, indicating that the maximal effect was already reached at the lowest pregabalin concentrations. As such, model development was continued with an on-off effect model, which resulted in a similar OFV and reduced RSE’s compared to the full Emax model. Population parameters were estimated with low RSE’s (<30%).
The PD model development of the pressure and contact heat pain models did not result in the identification of a stable model, due to difficulties in parameter estimation and high levels of variability.
Conclusions:
Two population PK/PD models with significant concentration-effect relationships were developed to describe the effect of oral pregabalin on the cold pressor and electrical stimulation response in healthy subjects. The previously identified significant effects of pregabalin on the pressure and heat pain models could not be confirmed using a population PK/PD modelling approach. The high variability in the baseline response to all pain models suggests that sufficient baseline measurements should be performed at each occasion.
References:
[1] Hay JL, Okkerse P, van Amerongen G, Groeneveld GJ. Determining Pain Detection and Tolerance Thresholds Using an Integrated, Multi-Modal Pain Task Battery. J Vis Exp 2016. doi:10.3791/53800.
[2] Okkerse P, Van Amerongen G, De Kam ML, Stevens J, Butt RP, Gurrell R, et al. The use of a battery of pain models to detect analgesic properties of compounds: A two-part, four-way, randomised, placebo-controlled, crossover study. Br J Clin Pharmacol 2017. doi:10.1111/bcp.13183.
[3] Beal SL, Sheiner LB, Boeckmann AJ, and Bauer RJ (eds) NONMEM 7.3.0 Users Guides. (1989–2013). ICON Development Solutions, Hanover, MD.
Reference: PAGE 27 (2018) Abstr 8653 [www.page-meeting.org/?abstract=8653]
Poster: Drug/Disease Modelling - CNS