A Phenotyping Tool for Seven Cytochrome P450 Enzymes and Two Transporters: Application to Examine the Effects of Clopidogrel and Gemfibrozil

Abstract

Clinical cocktails for cytochrome P450 (CYP) phenotyping lack a marker for CYP2C8. We aimed to combine the CYP2C8 index drug repaglinide with the Geneva cocktail (caffeine/CYP1A2, bupropion/CYP2B6, flurbiprofen/CYP2C9, omeprazole/CYP2C19, dextromethorphan/CYP2D6, and midazolam/CYP3A4). We also included endogenous organic anion transporting polypeptide (OATP) 1B1 and 1B3 biomarkers glycochenodeoxycholate 3-O-glucuronide and glycochenodeoxycholate 3-sulfate, and investigated the CYP2C8 inhibition selectivity of clopidogrel and gemfibrozil with the full cocktail. In a five-phase randomized cross-over study, the following drugs were administered to 16 healthy volunteers: (i) repaglinide, (ii) the Geneva cocktail, (iii) repaglinide with the Geneva cocktail (full cocktail), (iv) clopidogrel followed by the full cocktail, and (v) gemfibrozil followed by the full cocktail. The Geneva cocktail increased repaglinide AUC_0-23h 1.22-fold (90% confidence interval 1.04-1.44, P = 0.033). The full cocktail accurately captured known inhibitory effects of clopidogrel on CYP2B6, CYP2C8, and CYP2C19 and that of gemfibrozil on CYP2C8. Gemfibrozil decreased the paraxanthine/caffeine AUC_0-12h ratio by 23% (14-31%, P < 0.01) and increased caffeine AUC_0-12h 1.20-fold (1.03-1.40, P = 0.036). Gemfibrozil increased the metabolite-to-index drug AUC_0-23h ratios of flurbiprofen, omeprazole, dextromethorphan, and midazolam 1.59-fold (1.32-1.92), 1.47-fold (1.34-1.61), 1.79-fold (1.23-2.59), and 2.1-fold (1.9-2.4), respectively, without affecting the index drug AUCs (P < 0.01). Gemfibrozil increased the AUC_0-4h of glycochenodeoxycholate 3-O-glucuronide 1.33-fold (1.07-1.65, P = 0.027). In conclusion, the combination of repaglinide, the Geneva cocktail and endogenous biomarkers for OATP1B1 and OATP1B3 yields a nine-in-one phenotyping tool. Apart from strong CYP2C8 inhibition, gemfibrozil weakly inhibits CYP1A2 and OATP1B1 and appears to impair the elimination of the metabolites of several CYP index drugs.

Figure 1

Repaglinide concentrations as geometric mean values ± 90% confidence intervals with and without the Geneva cocktail (a) and after clopidogrel pretreatment and gemfibrozil pretreatment compared with Geneva cocktail + repaglinide (b). Results are presented in two subfigures for clarity.

Figure 2

Individual AUC_0‐12h and AUC_0‐23h ratios of the Geneva cocktail index drugs in study phases with Geneva cocktail + repaglinide (full cocktail, control), clopidogrel pretreatment followed by the full cocktail and gemfibrozil pretreatment followed by the full cocktail represented with symbols for poor metabolizer (PM), intermediate metabolizer (IM), normal metabolizer (NM), rapid metabolizer (RM) and ultrarapid metabolizer (UM) CYP phenotypes. Geometric mean values ± 90% confidence intervals are expressed as horizontal lines with error bars. P < 0.05 are considered statistically significant.

Figure 3

Concentrations of the Geneva cocktail index drugs and their metabolites for six CYP enzymes ((a) CYP1A2, (b) CYP2B6, (c) CYP2C9, (d) CYP2C19, (e) CYP2D6 and (f) CYP3A4) as geometric mean values ± 90% confidence intervals in study phases with Geneva cocktail + repaglinide (full cocktail), clopidogrel pretreatment followed by the full cocktail and gemfibrozil pretreatment followed by the full cocktail.