Matti K. Itkonen

Clopidogrel Markedly Increases Plasma Concentrations of CYP2C8 Substrate Pioglitazone

The glucose-lowering drug pioglitazone undergoes hepatic CYP2C8-mediated biotransformation to its main metabolites. The antiplatelet drug clopidogrel is metabolized to clopidogrel acyl-β-d-glucuronide, which was recently found to be a strong time-dependent inhibitor of CYP2C8 in humans. Therefore, we studied the effect of clopidogrel on the pharmacokinetics of pioglitazone. In a randomized crossover study, 10 healthy volunteers ingested either 300 mg of clopidogrel on day 1, and 75 mg on days 2 and 3, or placebo. Pioglitazone 15 mg was administered 1 hour after placebo and clopidogrel on day 1. Plasma concentrations of pioglitazone, clopidogrel, and their main metabolites were measured up to 72 hours. Clopidogrel increased the area under the plasma concentration-time curve (AUC0–∞) of pioglitazone 2.1-fold [P < 0.001, 90% confidence interval (CI) 1.8–2.6] and prolonged its half-life from 6.7 to 11 hours (P = 0.002). The peak concentration of pioglitazone was unaffected but the concentration at 24 hours was increased 4.5-fold (range 1.6–9.8; P < 0.001, 90% CI 3.17–6.45) by clopidogrel. The M-IV-to-pioglitazone AUC0–∞ ratio was 49% (P < 0.001, 90% CI 0.40–0.59) of that during the control phase, indicating that clopidogrel inhibited the CYP2C8-mediated biotransformation of pioglitazone. Clopidogrel increases the exposure to pioglitazone by inhibiting its CYP2C8-mediated biotransformation. In consequence, use of clopidogrel may increase the risk of fluid retention and other concentration-related adverse effects of pioglitazone.

Clopidogrel Has No Clinically Meaningful Effect on the Pharmacokinetics of the Organic Anion Transporting Polypeptide 1B1 and Cytochrome P450 3A4 Substrate Simvastatin

Simvastatin and clopidogrel are commonly used together in the treatment of cardiovascular diseases. Organic anion transporting polypeptide (OATP) 1B1 activity markedly affects the hepatic uptake of simvastatin acid, whereas both simvastatin and simvastatin acid are sensitive to changes in cytochrome P450 3A4 activity. Clopidogrel and its metabolites inhibit OATP1B1 and CYP3A4 in vitro. We studied the effect of clopidogrel on the pharmacokinetics of simvastatin in a randomized crossover study. Twelve healthy volunteers ingested either a dose of placebo (control) or 300 mg of clopidogrel on day 1 and 75 mg on days 2 and 3. Simvastatin 40 mg was administered 1 hour after placebo and after clopidogrel on days 1 and 3. Plasma drug concentrations were measured for up to 12 hours. Clopidogrel 300 mg (day 1) increased the concentrations of simvastatin and simvastatin acid during the absorption phase. After clopidogrel 300 mg, the area under the concentration time curve (AUC) of simvastatin from 0 to 2 hours was 156% (P = 0.02) and its AUC0–12 hours was 132% (P = 0.08) of that during placebo, whereas the AUC0–2 hours and the AUC0–12 hours of simvastatin acid were 148% (P = 0.04) and 112% (P = 0.52) of control. Clopidogrel 75 mg (day 3) had no significant effect on the pharmacokinetic variables of simvastatin or simvastatin acid compared with placebo. The effect of clopidogrel seemed independent of the SLCO1B1 c.521T>C genotype. In conclusion, as clopidogrel did not have significant effects on the total exposure to simvastatin or simvastatin acid, clopidogrel does not seem to inhibit OATP1B1 or CYP3A4 to a clinically relevant extent.

Clopidogrel carboxylic acid glucuronidation is mediated mainly by UGT2B7, UGT2B4 and UGT2B17: Implications for pharmacogenetics and drug-drug interactions

The antiplatelet drug clopidogrel is metabolized to an acyl-β-d-glucuronide, which causes time-dependent inactivation of CYP2C8. Our aim was to characterize the UDP-glucuronosyltransferase (UGT) enzymes that are responsible for the formation of clopidogrel acyl-β-d-glucuronide. Kinetic analyses and targeted inhibition experiments were performed using pooled human liver and intestine microsomes (HLMs and HIMs, respectively) and selected human recombinant UGTs based on preliminary screening. The effects of relevant UGT polymorphisms on the pharmacokinetics of clopidogrel were evaluated in 106 healthy volunteers. UGT2B7 and UGT2B17 exhibited the greatest level of clopidogrel carboxylic acid glucuronidation activities, with a CLint,u of 2.42 and 2.82 µl⋅min−1⋅mg−1, respectively. Of other enzymes displaying activity (UGT1A3, UGT1A9, UGT1A10-H, and UGT2B4), UGT2B4 (CLint,u 0.51 µl⋅min−1⋅mg−1) was estimated to contribute significantly to the hepatic clearance. Nonselective UGT2B inhibitors strongly inhibited clopidogrel acyl-β-d-glucuronide formation in HLMs and HIMs. The UGT2B17 inhibitor imatinib and the UGT2B7 and UGT1A9 inhibitor mefenamic acid inhibited clopidogrel carboxylic acid glucuronidation in HIMs and HLMs, respectively. Incubation of clopidogrel carboxylic acid in HLMs with UDPGA and NADPH resulted in strong inhibition of CYP2C8 activity. In healthy volunteers, the UGT2B17*2 deletion allele was associated with a 10% decrease per copy in the plasma clopidogrel acyl-β-d-glucuronide to clopidogrel carboxylic acid area under the plasma concentration-time curve from 0 to 4 hours (AUC0–4) ratio (P < 0.05). To conclude, clopidogrel carboxylic acid is metabolized mainly by UGT2B7 and UGT2B4 in the liver and by UGT2B17 in the small intestinal wall. The formation of clopidogrel acyl-β-d-glucuronide is impaired in carriers of the UGT2B17 deletion. These findings may have implications regarding the intracellular mechanisms leading to CYP2C8 inactivation by clopidogrel.

Clopidogrel but Not Prasugrel Significantly Inhibits the CYP2C8‐Mediated Metabolism of Montelukast in Humans

The oxidation of montelukast is mainly mediated by cytochrome P450 (CYP) 2C8, but other mechanisms may contribute to its disposition. In healthy volunteers, we investigated the effects of two widely used P2Y12 inhibitors on montelukast pharmacokinetics. Clopidogrel (300 mg on day 1 and 75 mg on day 2) increased the area under the plasma concentration–time curve (AUC) of montelukast 2.0‐fold (90% confidence interval (CI) 1.72–2.28, P < 0.001) and decreased the M6:montelukast AUC0‐7h ratio to 45% of control (90% CI 40–50%, P < 0.001). Prasugrel (60 mg on day 1 and 10 mg on day 2) had no clinically meaningful effect on montelukast pharmacokinetics. Our results imply that clopidogrel is at least a moderate inhibitor of CYP2C8, but prasugrel is not a clinically relevant CYP2C8 inhibitor. The different interaction potentials of clopidogrel and prasugrel are important to consider when antiplatelet therapy is planned for patients at risk for polypharmacy with CYP2C8 substrates.

Clopidogrel Increases Dasabuvir Exposure With or Without Ritonavir, and Ritonavir Inhibits the Bioactivation of Clopidogrel

Dasabuvir is mainly metabolized by cytochrome P450 (CYP) 2C8 and is predominantly used in a regimen containing ritonavir. Ritonavir and clopidogrel are inhibitors of CYP3A4 and CYP2C8, respectively. In a randomized, crossover study in 12 healthy subjects, we examined the impact of clinical doses of ritonavir (for 5 days), clopidogrel (for 3 days), and their combination on dasabuvir pharmacokinetics, and the effect of ritonavir on clopidogrel. Clopidogrel, but not ritonavir, increased the geometric mean AUC0‐∞ of dasabuvir 4.7‐fold; range 2.0–10.1‐fold (P = 8·10−7), compared with placebo. Clopidogrel and ritonavir combination increased dasabuvir AUC0‐∞ 3.9‐fold; range 2.1–7.9‐fold (P = 2·10−6), compared with ritonavir alone. Ritonavir decreased the AUC0‐4h of clopidogrel active metabolite by 51% (P = 0.0001), and average platelet inhibition from 51% without ritonavir to 31% with ritonavir (P = 0.0007). In conclusion, clopidogrel markedly elevates dasabuvir concentrations, and patients receiving ritonavir are at risk for diminished clopidogrel response.

Response to “Interaction of Dasabuvir With Clopidogrel: Did Predictions by Physiologically Based Pharmacokinetics Modeling Pass the Test?”

To the Editor: We would like to thank Dr. Shebley for valuable comments concerning our study and for the physiologically based pharmacokinetic (PBPK) simulations that fairly accurately reproduced our findings. As pointed out, there are some differences in study design between the previously published PBPK simulations by Shebley et al. and our clinical study. The loading dose of clopidogrel can partially explain the greater interaction in the clinical study. However, in our earlier study by Tornio et al., the area under the curve (AUC) of the cytochrome P450 (CYP)2C8 probe drug repaglinide was increased by 75mg daily doses of clopidogrel only about 20% less (3.9fold vs. 5.1fold) than by a single 300 mg clopidogrel dose. Moreover, it is difficult to mechanistically explain how CYP2C8 induction by ritonavir would attenuate the effect of clopidogrel on dasabuvir AUC, as proposed by Shebley et al., because CYP2C8 induction should increase the fraction of dasabuvir metabolized by CYP2C8 (fmCYP2C8). Consequently, dasabuvir would actually be more susceptible to CYP2C8 inhibition with ritonavir. Unfortunately, Shebley et al. did not demonstrate with the PBPK model how incorporation of induction of CYP2C8 by ritonavir changes the effect of clopidogrel on dasabuvir exposure. Furthermore, clinical evidence regarding CYP2C8 induction by ritonavir is scarce. Accordingly, although the PBPK model of Shebley et al. seemed to work adequately with our clinical study, the model has major mechanistic uncertainties. Thus, until more thorough clinical data is available, care is warranted in extrapolating the results of our clopidogreldasabuvir interaction study to a situation in which clopidogrel is used as maintenance doses with dasabuvir. In addition, it should be kept in mind that a patient already using dasabuvir may require clopidogrel therapy that involves the standard loading dose of 300 mg (or even 600 mg). Last, we would like to underline the substantial interindividual variability in the extent of clopidogreldasabuvir interaction. In our study with 12 healthy young subjects, the largest individual increases in dasabuvir AUC were 10.1fold and 7.9fold in the clopidogrel phase and clopidogrelritonavir phase, respectively. Concentrationdependent adverse effects of dasabuvir (e.g., QTc prolongation), likely occur in patients who have the highest, not the average, drug concentrations and additional risk factors. Therefore, we find it essential that clinicians recognize the large interindividual variation in the magnitude of the clopidogreldasabuvir interaction and its potential consequences in order to choose efficacious and safe drug treatments and followup for the right patients. FUNDING No funding was received for this work.

Effects of Genetic Variants on Carboxylesterase 1 Gene Expression, and Clopidogrel Pharmacokinetics and Antiplatelet Effects

Several single nucleotide variations (SNVs) affect carboxylesterase 1 (CES1) activity, but the effects of genetic variants on CES1 gene expression have not been systematically investigated. Therefore, our aim was to investigate effects of genetic variants on CES1 gene expression in two independent whole blood sample cohorts of 192 (discovery) and 88 (replication) healthy volunteers and in a liver sample cohort of 177 patients. Furthermore, we investigated possible effects of the found variants on clopidogrel pharmacokinetics (n = 106) and pharmacodynamics (n = 46) in healthy volunteers, who had ingested a single 300 mg or 600 mg dose of clopidogrel. Using massively parallel sequencing, we discovered two CES1 SNVs, rs12443580 and rs8192935, to be strongly and independently associated with a 39% (p = 4.0 × 10−13) and 31% (p = 2.5 × 10−8) reduction in CES1 whole blood expression per copy of the minor allele. These findings were replicated in the replication cohort. However, these SNVs did not affect CES1 liver expression, or clopidogrel pharmacokinetics or pharmacodynamics. Conversely, the CES1 c.428G>A missense SNV (rs71647871) impaired the hydrolysis of clopidogrel, increased exposure to clopidogrel active metabolite and enhanced its antiplatelet effects. In conclusion, the rs12443580 and rs8192935 variants reduce CES1 expression in whole blood but not in the liver. These tissue‐specific effects may result in substrate‐dependent effects of the two SNVs on CES1‐mediated drug metabolism.