Physiologically based pharmacokinetic modeling for development and applications of a virtual celiac disease population using felodipine as a model drug

Abstract

In celiac disease (CeD), gastrointestinal CYP3A4 abundance and morphology is affected by the severity of disease. Therefore, exposure to CYP3A4 substrates and extent of drug interactions is altered. A physiologically-based pharmacokinetic (PBPK) population for different severities of CeD was developed. Gastrointestinal physiology parameters, such as luminal pH, transit times, morphology, P-gp, and CYP3A4 expression were included in development of the CeD population. Data on physiological difference between healthy and CeD subjects were incorporated into the model as the ratio of celiac to healthy. A PBPK model was developed and verified for felodipine extended-release tablet in healthy volunteers (HVs) and then utilized to verify the CeD populations. Plasma concentration-time profile and PK parameters were predicted and compared against those observed in both groups. Sensitivity analysis was carried out on key system parameters in CeD to understand their impact on drug exposure. For felodipine, the predicted mean concentration-time profiles and 5th and 95th percentile intervals captured the observed profile and variability in the HV and CeD populations. Predicted and observed clearance was 56.9 versus 56.1 (L/h) in HVs. Predicted versus observed mean ± SD area under the curve for extended release felodipine in different severities of CeD were values of 14.5 ± 9.6 versus 14.4 ± 2.1, 14.6 ± 9.0 versus 17.2 ± 2.8, and 28.1 ± 13.5 versus 25.7 ± 5.0 (ng.h/mL), respectively. Accounting for physiology differences in a CeD population accurately predicted the PK of felodipine. The developed CeD population can be applied for determining the drug concentration of CYP3A substrates in the gut as well as for systemic levels, and for application in drug-drug interaction studies.

FIGURE 1

Modeling strategy for development and verification of a celiac disease population using felodipine extended‐release tablet PBPK model. BP, blood pressure; CR/MR, controlled/modified release; HV, healthy volunteer; IVIVC, in vitro‐in vivo correlation; MW, molecular weight; PBPK, physiologically‐based pharmacokinetic.

FIGURE 2

Comparison of the gastrointestinal tract physiology differences among healthy, treated, and untreated patients with celiac disease with gluten‐free diet. The error bars show the coefficient of variation on parameters.

FIGURE 3

Ratio of predicted to observed CL, AUC, C_max, and T_max for felodipine after intravenous, and oral solution and extended‐release tablet administration to healthy and celiac disease populations. Shaded area is two‐folds intervals and black dashed lines show 1.5‐folds. AUC, area under the curve; CL, clearance; C _max, maximum plasma concentration; T _max, time to maximum plasma concentration.

FIGURE 4

Predicted dissolution profile, fraction absorbed, portal vein fraction, and bioavailability versus time for felodipine after in vitro‐in vivo correlation in Simcyp V21.

FIGURE 5

Plasma concentration‐time profile of extended release felodipine tablet in healthy volunteers using 10 trials. The black solid line is the mean of predictions, gray lines are the mean of each trial and dashed gray lines are the 5th and 95th percentiles of predictions. The white circles are the mean of observed data reported in clinical studies and the error bars are the SD., , , , Black dotted line and gray dotted lines are the mean and 5th and 95th percentiles after drug–drug interaction with erythromycin. HV, healthy volunteers.

FIGURE 6

Plasma concentration‐time profile of felodipine after extended‐release tablet in celiac disease using 10 trials. The black solid line is the mean of predictions, gray lines are the mean of each trial and dashed gray lines are the 5th and 95th percentiles of predictions. The white circles are the mean of observed data reported in clinical studies and the error bars are the SD.