Department of Pharmacology and Toxicology, Radboud University Nijmegen Medical Centre, The Netherlands.
Eur J Pharm Sci. 2013 Aug 16;49(5):819-28. doi: 10.1016/j.ejps.2013.06.009. Epub 2013 Jun 24.
We studied if the clinical pharmacokinetics and drug-drug interactions (DDIs) of the sulfonylurea-derivative glibenclamide can be simulated via a physiologically-based pharmacokinetic modeling approach. To this end, a glibenclamide PBPK-model was build in Simcyp using in vitro physicochemical and biotransformation data of the drug, and was subsequently optimized using plasma disappearance data observed after i.v. administration. The model was validated against data observed after glibenclamide oral dosing, including DDIs. We found that glibenclamide pharmacokinetics could be adequately modeled if next to CYP metabolism an active hepatic uptake process was assumed. This hepatic uptake process was subsequently included in the model in a non-mechanistic manner. After an oral dose of 0.875 mg predicted Cmax and AUC were 39.7 (95% CI:37.0-42.7)ng/mL and 108 (95% CI: 96.9-120)ng/mLh, respectively, which is in line with observed values of 43.6 (95% CI: 37.7-49.5)ng/mL and 133 (95% CI: 107-159)ng/mLh. For a 1.75 mg oral dose, the predicted and observed values were 82.5 (95% CI:76.6-88.9)ng/mL vs 91.1 (95% CI: 67.9-115.9) for Cmax and 224 (95% CI: 202-248) vs 324 (95% CI: 197-451)ng/mLh for AUC, respectively. The model correctly predicted a decrease in exposure after rifampicin pre-treatment. An increase in glibenclamide exposure after clarithromycin co-treatment was predicted, but the magnitude of the effect was underestimated because part of this DDI is the result of an interaction at the transporter level. Finally, the effects of glibenclamide and fluconazol co-administration were simulated. Our simulations indicated that co-administration of this potent CYP450 inhibitor will profoundly increase glibenclamide exposure, which is in line with clinical observations linking the glibenclamide-fluconazol combination to an increased risk of hypoglycemia. In conclusion, glibenclamide pharmacokinetics and its CYP-mediated DDIs can be simulated via PBPK-modeling. In addition, our data underline the relevance of modeling transporters on a full mechanistic level to further improve pharmacokinetic and DDI predictions of this sulfonylurea-derivative.
我们研究了磺酰脲类药物格列本脲的临床药代动力学和药物相互作用（DDI）是否可以通过基于生理学的药代动力学建模方法来模拟。为此，我们在 Simcyp 中使用该药物的体外物理化学和生物转化数据构建了格列本脲 PBPK 模型，并随后使用静脉内给药后观察到的血浆消除数据对其进行了优化。该模型经过验证后可用于口服给药后观察到的数据，包括 DDI。我们发现，如果除了 CYP 代谢之外还假设存在活跃的肝摄取过程，则可以充分模拟格列本脲的药代动力学。随后，以非机械方式将该肝摄取过程包含在模型中。口服 0.875mg 后，预测的 Cmax 和 AUC 分别为 39.7（95%CI：37.0-42.7）ng/mL 和 108（95%CI：96.9-120）ng/mL·h，这与观察到的 43.6（95%CI：37.7-49.5）ng/mL 和 133（95%CI：107-159）ng/mL·h 的值一致。对于 1.75mg 口服剂量，预测值和观察值分别为 82.5（95%CI：76.6-88.9）ng/mL 与 91.1（95%CI：67.9-115.9）ng/mL 用于 Cmax，224（95%CI：202-248）ng/mL 与 324（95%CI：197-451）ng/mL·h 用于 AUC。模型正确预测了利福平预处理后暴露量的降低。预测克拉霉素联合治疗后格列本脲的暴露量增加，但由于该 DDI 的一部分是转运蛋白水平相互作用的结果，因此效应的幅度被低估了。最后，模拟了格列本脲和氟康唑联合给药的效果。我们的模拟表明，这种强效 CYP450 抑制剂的联合使用将极大地增加格列本脲的暴露量，这与将格列本脲-氟康唑联合使用与低血糖风险增加相关的临床观察结果一致。总之，通过 PBPK 建模可以模拟格列本脲的药代动力学及其 CYP 介导的 DDI。此外，我们的数据强调了在全机械水平上对转运体进行建模的相关性，以进一步提高该磺酰脲类药物的药代动力学和 DDI 预测。
