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

To the Editor:

A recent drug–drug interaction (DDI) study reported a 3.89–4.67‐fold increase in exposures of dasabuvir by clopidogrel,1 and suggested a discrepancy between observed data and previously published physiologically based pharmacokinetic (PBPK) predictions.2 We have published prospective predictions of dasabuvir and clopidogrel DDI using PBPK, which suggested a 2.8–4.5‐fold increase at steady state. The PBPK simulation design represented a clinically relevant scenario in which dasabuvir and ritonavir are coadministered with clopidogrel maintenance dosage (75 mg q.d.) at steady state (Figure 1 a). The model sensitivity analysis suggested a maximum increase in dasabuvir exposures to 2.6‐fold for peak plasma concentration (C _max) and 4.5‐fold for area under the curve (Figure 1b).2

Figure 1.

Comparison of study designs and results of PBPK simulations by Shebley et al.2 vs. clinical study by Itkonen et al.1 (a) PBPK simulation study design of a clinically relevant scenario; all drugs at steady state, interaction assessment with CLOP maintenance dosage of 75 mg q.d., DAS 250 mg b.i.d., and RTV boosting dose of 100 mg q.d., on simulated study D5 of coadministration. (b) Results from the previously published PBPK predictions of a base case DDI scenario and sensitivity analysis, including worst case. (c) Itkonen et al.1 clinical study design for maximizing DDI; interaction assessment with CLOP 300‐mg loading dose, higher RTV dosage of 100 mg b.i.d., and DAS 250‐mg single dose on study D1 of coadministration. (d) PBPK simulation results replicating Itkonen et al.1 clinical study design using a previously published PBPK model by Shebley et al.2 AUCR, ratio of area under the curve with inhibitor relative to without inhibitor; CLOP, clopidogrel; CmaxR, ratio of peak plasma concentration with inhibitor relative to without inhibitor; D, day; DAS, dasabuvir; DDI, drug‐drug interaction; PBPK, physiologically based pharmacokinetic; RTV, ritonavir.

Itkonen et al.1 suggested an apparent underprediction of DDI by the PBPK model. However, when differences between the PBPK simulation and clinical study designs were taken into consideration (Figure 1 a,c), the clinical data appear to validate our PBPK predictions. PBPK simulations predicted effects of clopidogrel 75‐mg q.d. maintenance dosage on dasabuvir 250 mg b.i.d. with ritonavir 100 mg q.d. at steady state. Itkonen et al.1 evaluated clopidogrel 300‐mg loading dose on dasabuvir 250‐mg single dose, with and without ritonavir 100 mg b.i.d. With clopidogrel 300 mg, plasma concentrations of the acyl‐β‐glucuronide metabolite are threefold to fourfold higher compared with 75 mg. The impact of clopidogrel higher dose on dasabuvir exposures was replicated by the PBPK model, suggesting that the observed higher interaction relative to that predicted previously could be explained by higher exposures of the acyl‐β‐glucuronide (Figure 1 d). The ritonavir regimen was shorter in the clinical study, leading to relatively lower induction of Cytochrome P450 2C8 (CYP2C8), alternatively explaining differences between the results.

When Itkonen et al.1 compared clinical results with PBPK predictions, the authors failed to highlight this important factor of higher clopidogrel dose (300‐mg loading dose) on observed DDI with dasabuvir compared with that predicted by PBPK at lower dosage (75 mg q.d.). The authors highlighted potential concern of corrected QT interval (QTc) prolongation attributable to DDI; however, they did not acknowledge that the 1.82‐fold increase in dasabuvir C _max was with clopidogrel 300 mg, which is used clinically as a loading dose only on day 1 of treatment.3 This suggests that clopidogrel maintenance dosage of 75 mg q.d. may cause lower increases in dasabuvir exposures, as previously demonstrated by the PBPK model.2

Maximum effects on dasabuvir exposures at the higher 300‐mg loading dose of clopidogrel were 1.40–1.82‐fold on C _max and 3.89–4.67‐fold on area under the curve, a range well captured in the sensitivity analysis of the PBPK model by Shebley et al.2 Thus, dasabuvir contraindication with strong CYP2C8 inhibitors remains appropriate.4 These data demonstrate the powerful impact of PBPK modeling in predicting DDIs, deciphering study design differences, and evaluating complex interactions.

Funding

AbbVie, Inc., provided financial support for the studies and participated in the design, study conduct, analysis, and interpretation of data as well as the writing, review, and approval of the publication.

Conflict of Interest

M. Shebley is an employee of AbbVie, Inc., and may hold stock or stock options.