A Population Pharmacokinetic Analysis of Dextroamphetamine in the Plasma and Hair of Healthy Adults

Abstract

Background and objective: Hair is an attractive matrix for amphetamine drug testing; however, little is known about the rate at which amphetamines are deposited into hair. Therefore, the purpose of this study was to determine the pharmacokinetics of oral dextroamphetamine in plasma and quantify the rate of deposition into hair in healthy adults using a linked population pharmacokinetic model.

Methods: Healthy adults >18 years of age received dextroamphetamine 10 mg orally for 7 days. Plasma samples were collected over 48 h following the final dose, and hair was collected 5 weeks following the first dose. NONMEM 7.2 was used to estimate dextroamphetamine oral absorption rate constant, apparent clearance and volume of distribution of the plasma compartment, the plasma to hair incorporation rate constant, and the apparent volume of distribution in the hair compartment.

Results: Dextroamphetamine pharmacokinetics were well-described by a one-compartment model with combined additive and proportional error for the plasma compartment, which was linked to a single compartment for the hair. Apparent clearance and volume of distribution in the plasma compartment were scaled by current body weight (centered on the mean). Melanin hair concentration was included as a significant covariate on the hair compartment. Absorption rate constant, clearance, and volume of distribution for the plasma compartment were estimated as 0.527 h(-1) (95% CI 0.467-0.586), 28.7 L/h (95% CI 27.1-30.3), and 377 L (95% CI 326-428), respectively. The incorporation rate constant from plasma to hair was 1.60e(-6) h(-1) (95% CI 1.06e(-6)-2.14e(-6)) and apparent volume of distribution in hair was 17.7 mg (95% CI 12.5-22.8).

Conclusions: A one-compartment plasma model linked to a single compartment for hair successfully described the pharmacokinetics of dextroamphetamine in healthy adults. The volume of distribution and clearance of dextroamphetamine increased with weight, and the volume of distribution of the hair compartment increased with greater melanin concentrations.

Conflict of interest statement

Conflicts of interest Jessica K. Roberts, Sarah F. Cook, Chris Stockmann, Douglas E. Rollins, Diana G. Wilkins, and Catherine M. T. Sherwin have no financial relationships with any organisations that might have an interest in the submitted work, nor are there any other relationships or activities that could appear to have influenced the submitted work.

Figures

Fig. 1

Final linked dextroamphetamine model schematic. K12 rate constant from the gut compartment to the plasma compartment, V1/F apparent volume of distribution of the plasma compartment, CL/F apparent clearance from the plasma compartment, K23 incorporation rate constant from plasma to hair, V2/F apparent volume of distribution of the hair compartment

Fig. 2

Diagnostic plots for the final linked dextroamphetamine covariate model. Observed versus a population-predicted and b individual-predicted dextroamphetamine plasma concentrations, and observed versus c population-predicted and d individual-predicted dextroamphetamine hair concentrations. The solid black lines depict the lines of identity (y = x), and the dashed black lines depict the LOESS fits of the data

Fig. 3

Diagnostic plots for the final linked dextroamphetamine covariate model. Conditional weighted residuals of dextroamphetamine plasma concentrations versus a time since previous dose and b population-predicted concentrations. Conditional weighted residuals of dextroamphetamine hair concentrations versus c time since previous dose and d population-predicted concentrations. The solid black lines depict y = 0

Fig. 4

Diagnostic plots for the final linked dextroamphetamine covariate model. Normalized prediction distribution errors of dextroamphetamine plasma concentrations versus a time since previous dose and b population-predicted concentrations. Normalized prediction distribution errors of dextroamphetamine hair concentrations versus c time since previous dose and d population-predicted concentrations. The solid black lines depict y = 0

Fig. 5

Visual predictive check for the final linked dextroamphetamine covariate model. Observations are depicted as gray dots. The solid black line depicts the observed 50th percentile, and the dashed black lines depict the observed 5th and 95th percentiles. The shaded regions depict the 95 % confidence intervals surrounding the predicted 5th, 50th, and 95th percentiles