A Population Pharmacokinetic Analysis of PF-5190457, a Novel Ghrelin Receptor Inverse Agonist in Healthy Volunteers and in Heavy Alcohol Drinkers

Abstract

Background and objectives: The ghrelin receptor (GHS-R1a) is a potential target for alcohol use disorders. PF-5190457 is the first inverse agonist of GHS-R1a to progress to clinical development with potential to treat alcohol use disorder. We present a population pharmacokinetic model for PF-5190457 in non-heavy (alcohol consumption status = 0) and heavy alcohol drinkers (alcohol consumption status = 1), and identify relevant factors that can influence its pharmacokinetics.

Methods: Plasma concentration-time data from non-heavy (n = 35) and heavy drinkers (n = 12) were pooled for the population pharmacokinetic model development. The influence of various covariates including alcohol consumption status was evaluated. The accuracy, precision, and robustness of the model were also evaluated using bootstrapping and visual predictive checks.

Results: A two-compartment model best described the pharmacokinetics of PF-5190457. The apparent volume of distribution of 44.5 L, apparent clearance of 72.0 L/h, apparent peripheral volume of distribution of 271 L, apparent distributional clearance of 28.7 L/h, and first-order absorption rate constant of 0.27/h were accurate and precise. The apparent volume of distribution was 3.8-fold higher (169 L) in heavy drinkers, and correlated with a lower maximum plasma concentration in heavy drinkers compared with non-heavy drinkers at the same dose; and a corresponding reduced incidence of somnolence in heavy drinkers at doses > 50 mg.

Conclusions: This work provides an accurate, precise, and robust two-compartment model that describes the pharmacokinetics of PF-5190457 and suggests a possible link of PF-5190457 pharmacokinetics with somnolence.

Trial registration: ClinicalTrials.gov identifier numbers NCT01247896 and NCT02039349.

Conflict of interest statement

CONFLICT OF INTEREST: Enoch Cobbina, Mary R. Lee, Lorenzo Leggio, and Fatemeh Akhlaghi declare that they have no potential conflicts of interest that might be relevant to the contents of this manuscript.

Figures

Figure 1.

Box and Whisker plots of ETAs versus ALC. After incorporating the alcohol consumption status of the population, the inter-individual variability on V2 reduced by 26% from 1.41 to 1.03. ALC alcohol consumption status, HD: Heavy Drinkers

Figure 2.

Goodness-of-fit and Diagnostic Plots. All data were log transformed. Observed concentration (DV) was strongly correlated with the predicted concentration (IPRED). CWRES conditional weighted residuals, DV observed plasma concentration, IPRED individual predicted plasma concentration, PRED population predicted plasma concentration.

Figure 3.. Visual Predictive Check for the Final Population PK Model.

Open Circle: Observed Concentrations; Red solid Line: Median of Observed Concentrations; Red dashed Lines: 5th and 95th percentile of observed concentrations. Red Shaded Region: 90% Prediction Interval for Median of Predicted Concentrations; Blue Shaded Regions: 90% Prediction Intervals for the 5th and 95th percentiles of Predicted Concentrations. Time is expressed in minutes.

Figure 4:

Graphical representation of simulation of 100 mg b.i.d doses of PF-5190457 for 3 days using the final model. Solid lines show the median simulated concentration. The shaded area is the 95% confidence interval for the three simulations: Base, Phase 1a and Phase 1b. There is an overlap of the 95% CI. b.i.d. twice daily, conc concentration, HD heavy drinker.