S. D. Hall

The Interaction between CYP3A Substrates in The Elderly

Clinical Pharmacology & Therapeutics (1999) 65, 136–136; doi:

Improved prediction of drug interactions using in vivo Ki

Drug‐drug interactions are a leading cause of death in the U.S. CYP3A metabolism is the predominant pathway for drug metabolism. Itraconazole (ITZ) alters CYP3A activity resulting in clinically significant drug interactions. To date, in vitro studies have not been successful in predicting the magnitude of CYP3A drug interactions in humans. We validated the venous equilibrium model using a physiologically stable, chronically catheterized rat, measuring MDZ disposition from multiple, simultaneous aortic and hepatic vein blood samples after duodenal administration. MDZ clearance and hepatic availability were scaled to humans, and predicted published human MDZ pharmacokinetic parameters. MDZ disposition in vivo was characterized in rats after a single and multiple doses of ITZ. MDZ and ITZ concentrations were quantified by LC‐MS. In vivo estimates from rat model for KM of MDZ were 5.7 ± 0.6 ug/ml (n=8) and 1.1 ± 0.2 ug/ml (n=10) in the presence and absence of 0.71 uM ITZ. The extent of inhibition (AUCI/AUC) reported in humans of MDZ clearance by ITZ was 3.2‐fold after a single dose (ITZ CpSS, AVG= 0.17 uM) and 7‐fold after 6 daily doses of 200 mg (ITZ CpSS AVG = 0.71 uM). Based upon the effects of ITZ on MDZ pharmacokinetics in rats, the extent of MDZ inhibition was 2.4‐ and 7‐fold, respectively. The estimate for the ITZ inhibition constant in rats (Ki) accurately predicted the extent of inhibition of MDZ elimination in humans from previously published studies.

Quantitative prediction of CYP3A‐mediated drug interactions

Clinical Pharmacology & Therapeutics (2003) 73, P16–P16; doi:

Population pharmacokinetic modeling and simulation of intravenous and oral midazolam

To evaluate the population pharmacokinetics (PPK) of intravenous (IV) and oral (PO) midazolam (MDZ) with predictive covariates and to optimize dosing and sampling design for a clinical CYP3A inhibition study. Methods. PK data were combined from 5 similar studies consisting of 112 healthy adults (>3000 observations), 68 female, 44 male, 20 ‐ 81 yrs of age and 42 ‐113 kg weight (WT). Subjects were administered MDZ in PO (tablet, 15N solution or syrup) or IV formulations on 2 to 5 occasions. Simultaneous IV and PO (15N‐MDZ) data were available in 82% of subjects. IV MDZ dose (30‐min infusion) was 0.05 mg/kg and PO doses ranged 2.5 ‐ 15 mg. The PPK model was developed using nonlinear‐mixed effect model (NONMEM). Simulations (Pharsight TS2) were used to assess sample size effect on covariate identification and for model validation using Posterior Predictive Checks. Results. A three‐compartment open model described MDZ PK with first‐order absorption and elimination. CL was described by blood flow and intrinsic clearance (CLint). Blood flow was a power function of WT. CYP2D6 genotype and formulation influenced Clint and bioavailability. WT is a predictor on volume. Inter‐occasion variability on CL was 19%. Conclusions. A robust validated PPK model for MDZ has been developed. Simulations supported the power of the study to identify covariate effects. Optimal dosing and sampling schemes were selected based on simulations for a semi‐simultaneous CYP3A inhibition study.

The Effect of Estrogen Replacement Therapy on CYP3A Activity

Clinical Pharmacology & Therapeutics (1999) 65, 192–192; doi:

Verapamil‐induced PR prolongation is influenced by CYP3A5 genotype

Verapamil, a CYP3A4/5 substrate, controls ventricular rate in patients with supra‐ventricular tachycardia by delaying atrioventricular conduction. We hypothesized that subjects with the null CYP3A5 *3/*3 genotype had a more enhanced response to verapamil than *1 carriers.

The intestine is the major site of interaction between midazolam and verapamil

To determine the site of the drug interaction between verapamil and midazolam, we examined the effect of verapamil on intestinal and hepatic CYP3A activity in vivo using midazolam as a CYP3A substrate.

Dextromethorphan to dextrorphan urinary metabolic ratio not sensitive enough to assess modest changes in CYP2D6 activity

The ability of the dextromethorphan (DTM) to dextrorphan (DT) urinary metabolic ratio (UMR) to detect subtle changes in CYP2D6 activity is not well defined. 10 extensive metabolizers and 1 poor metabolizer were studied on two separate occasions. Serum and urine were collected over 24 hours after DTM (30 mg) oral dosing. DTM and DT were quantitated by LC‐MS (serum) and HPLC‐fluorescence (urine). The DTM oral clearance (CLPO), DTM/DT AUC ratio, and DTM/DT UMR were determined. DTM CLPO was significantly correlated with the DTM/DT AUC ratio (r = 0.98) and DTM/DT UMR (n= 22; r = 0.765). Exclusion of the poor metabolizer resulted in loss of correlation between DTM CLPO and DTM/DT UMR (r <0.001) but not DTM/DT AUC ratio(r = 0.84). Intra and inter subject variability for DTM CLPO, DTM/DT AUC ratio, and DTM/DT UMR was estimated to be 15%, 26%, 35% and 32%, 68%, 100%, respectively. Assuming an effect size of 30%, type I error rate at 5%, 80% power and a crossover design requires a sample size of 14, 34 or 56 patients when the primary outcome measure is DTM CLPO, DTM/DT AUC ratio, DTM/DT UMR, respectively. If 14 subjects are used in a crossover study design and the DTM/DT UMR is used, the effect size would have to be 395% to maintain 80% power. Cross‐sectional studies are similarly affected when the DTM/DT UMR is the principle outcome. The limitations of the DTM/DT UMR should be taken into account when modest effect sizes are expected. Supported by NIH Grants M01RR00750 and FDT001756.

Lack of correlation between lymphocytic mrna expression and activity of CYP3A4 and CYP2E1 In‐vivo in alcoholics and non‐alcoholics

Lymphocyte mRNA levels may serve as a surrogate for activity of drug metabolizing enzymes. We conducted a study to examine the relationship between CYP2E1 and CYP3A4 lymphocyte mRNA levels and the clearances (CL) of oral chlorzoxazone (CHZ) and oral and IV midazolam (MDZ), respectively. 20 alcoholics (ALC) (> 140gm of alcohol/wk) and 20 non‐alcoholics (N‐ALC) (20 African‐Americans 20 Caucasians) enrolled in the study. MDZ was given IV (0.05mg/kg over 30 min) and orally (5mg) on consecutive days. CHZ (500mg) was administered orally 8 hr after IV MDZ. mRNA was isolated from blood lymphocytes, CYP3A4 and 2E1 mRNA was quantitated using real time PCR. Serum CHZ and MDZ concentrations were determined by HPLC‐UV or LC‐MS. There was no significant difference (p>0.05) between ALC and N‐ALC for CYP3A4 mRNA, MDZ CLORAL (149.4±87.8 vs 120.3±84.1 L/hr) and CLIV (36.9±12 vs 36.6±14.1 L/hr). No correlation was observed between lymphocytic CYP3A4 mRNA and CYP3A activity. CYP2E1 mRNA levels were not significantly different between ALC and N‐ALC. In contrast, CHZ CLORAL was significantly greater (p<0.05) in ALC (31.5±11.9) compared to N‐ALC (23.4±8.7) but no correlation was observed between CYP2E1 activity and mRNA. No significant race based differences were observed for either CYP mRNA or activity. In conclusion, lymphocyte CYP2E1 and CYP3A4 mRNA levels did not reflect hepatic CYP2E1 activity nor hepatic and intestinal CYP3A activity.

Genetic polymorphisms of the CYP3A5 and MDR‐1 genes and effect of clarithromycin (CLAR) on midazolam (MDZ) pharmacokinetics

CLAR significantly reduces the intravenous and oral clearance of MDZ. In this study, we investigate the possible role of the CYP3A5 and MDR‐1 gene variants in this drug‐drug interaction.