藥物代謝酵素發育表現於Valproic acid肝損傷之探討

背景：Valproic acid (簡稱VPA)是兒童和成人常用的抗癲癇藥物，藥品動態學(pharmacokinetics，PKs)特性複雜。VPA相關肝毒性明列於仿單開頭之「Boxed Warning」，而在新生兒與嬰兒年齡次群(小於一歲)之發生率又高於成人。本研究旨在探討在個體發育過程中，藥物代謝酵素表現之變化於VPA肝毒性發生之角色，藉以解釋VPA相關肝毒性於兒童較易見之現象。方法：本研究以肝癌HepG2細胞株為實驗模型，將之轉染與VPA代謝相關之酵素cytochrome P450 (CYP) 同功異構酶2C9與2E1，及UDP-glucuronosyltransferase (UGT)2B7，藉以建構細胞研究模式，嘗試探究兒童和成人之VPA代謝途徑可能差異。本研究另分別整理或優化VPA 與其代謝物4-ene-VPA的in vitro、in silico及in vivo各參數值，建構適用於成人和兒童之以生理為基礎的藥品動態學(physiologically based pharmacokinetics，簡稱PBPK)化合物資料，亦彙整VPA臨床藥動學研究文獻，運用Simcyp軟體進行PBPK探討。後續並透過敏感性分析，以靈敏度指標評估CYP2C9與UGT2B7對於VPA與4-ene-VPA之藥動學參數的影響程度。最後將本研究所建立PBPK模型(基於個體發育學)，用於預測兒童各年齡次群之VPA最適劑量，並與傳統Fried氏或Young氏規則（基於年齡）、Clark氏規則（基於體重）及基於體表面積(body surface area, BSA)法所推算兒童用藥劑量進行比較。結果：當HepG2分別轉染CYP2C9或CYP2E1(模擬兒童VPA主要代謝途徑)時，觀察到VPA可導致HepG2細胞的粒線體超氧化物增加與細胞死亡。而當HepG2同時轉染CYP2C9與UGT2B7，或共同轉染CYP2E1與UGT2B7(模擬成人VPA主要代謝途徑)時，均未觀察到前述細胞傷害，或細胞損傷情形較僅轉染CYP2C9或CYP2E1時輕微；惟僅在轉染CYP2C9的HepG2，可以觀察到類似VPA肝損傷之脂質積累。有趣的是，當HepG2與CYP2C9及UGT2B7共轉染時，並無脂質累積現象。本研究彙集人體藥動學研究文獻14篇，並將VPA濃度-時間變化曲線圖之觀測值數位化處理，進而用於測試與評估前所建立PBPK模型之性能。結果顯示，以各臨床研究設計所模擬的血漿濃度-時間曲線，觀察值均落在預測值之5th至95th百分位數範圍內，且所預測的平均濃度-時間曲線下面積(AUC)、平均峰值血漿濃度(Cmax)，均與觀測值之差異介於0.5-2倍範圍間。敏感性分析結果呈現，在嬰幼兒(出生到一個月大)和青少年至成人(13至65歲)時， Clint,UGT2B7對VPA之AUC最具影響力。而在所有年齡層，Clint,CYP2C9則對4-ene-VPA之AUC和Cmax最具影響力。本研究以基於個體發育學PBPK 模型所預測的兒童VPA建議劑量，與Clark氏規則所推估的劑量相若，且兩種預測方式所得之Cmax，均於有效濃度目標範圍內。結論：In vitro研究結果顯示，當HepG2分別轉染CYP2C9或CYP2E1時，VPA 所造成的細胞損傷較HepG2同時轉染CYP2C9與UGT2B7，或共同轉染CYP2E1與UGT2B7時嚴重，得以解釋VPA肝毒性在兒童中更常被觀察到。本研究成功建立兒童與成人VPA及4-ene-VPA之PBPK模型，並應用於探討在個體發育過程中，藥物代謝酵素表現之變化於VPA與4-ene-VPA濃度影響。本研究所建立的PBPK模型和Clark氏規則是估計兒童劑量的推薦方法。然而，仍有一些未知因素可能影響兒童VPA的吸收、代謝、分布及排除，值得進一步研究。

關鍵字

藥物代謝酵素； HepG2細胞；個體發育； PBPK模型；兒童劑量； 2-丙(基)戊酸

並列摘要

Background: Valproic acid (VPA) is a commonly used anticonvulsant in pediatrics and adults with a complex pharmacokinetics (PKs) profile. Hepatotoxicity is enlisted in the “Boxed Warning” of the VPA package insert. Real-world data indicate neonates and infants have higher incidence of valproate hepatotoxicity than adults. This study aimed to investigate the ontogeny role in valproate hepatotoxicity and to elucidate possible mechanisms for the observed higher incidence in children. Methods: Human hepatocellular carcinoma HepG2 cells transfected with VPA metabolism-associated enzymes, cytochrome P450 (CYP) 2C9 and 2E1, and UDP-glucuronosyltransferase (UGT) 2B7, were used as experimental models to simulated different metabolic pathways between pediatrics and adults. Respective in vitro, in silico, and in vivo parameter values for constructing VPA and its metabolite, 4-ene-VPA, compound files were collected and optimized. The study also compiled VPA clinical PKs study data. The physiologically based pharmacokinetics (PBPK) models for VPA and 4-ene-VPA in adults and pediatrics were constructed and performed using the Simcyp PBPK Simulator. Sensitivity analysis was performed to characterize the influences of CYP2C9 and UGT2B7 on PK parameters of VPA and 4-ene-VPA by sensitivity index. Finally, the predicted doses for pediatric subgroups by the developed PBPK model (ontogeny-based) were compared with doses estimated by Fried’s or Young’s rule (age-based), Clark’s rule (weight-based), and body surface area (BSA)-based method. Results: The VPA increased mitochondrial superoxide and cell death in both HepG2 cells transfected with either CYP2C9 or CYP2E1 (representing major VPA metabolic pathway in pediatrics). When HepG2 cells were co-transfected with CYP2C9 and UGT2B7, or co-transfected with CYP2E1 and UGT2B7 (representing major VPA metabolic pathway in adults), the aforementioned cell injuries were minimal as compared to HepG2 transfected with either CYP2C9 or CYP2E1. VPA-associated lipid accumulation was only observed in HepG2 cells transfected with CYP2C9. Interestingly, no lipid accumulation was observed when HepG2 cells were co-transfected with CYP2C9 and UGT2B7. Observed data digitized from fourteen published literatures were utilized to develop and evaluate PBPK model performance. The observed data of plasma concentration-time profiles fell within the 5¬¬¬th to 95th percentile of PBPK predicted value. All of the predicted mean area-under-the-curve (AUC) and peak plasma concentration (Cmax) values were within two-fold of the observed data. Sensitivity analysis showed that Clint,UGT2B7 was the most influential parameters on the AUC of VPA in newborns and neonates (after birth up to 1 month), and adolescents to adults (13-18 years old). Clint,CYP2C9 was the most important key parameter with significant impact both on AUC and Cmax of 4-ene-VPA. The predicted doses by the ontogeny-based PBPK model were similar to estimations by the weight-based Clark’s rule among various age subgroups. All the predicted Cmax were within targeted therapeutic ranges. Conclusion: The in vitro data showed that the VPA-induced cell injuries in HepG2 cells transfected with either CYP2C9 or CYP2E1 were more severe than in HepG2 cells co-transfected with CYP2C9 and UGT2B7 or CYP2E1 and UGT2B7, indicating young children are prone to valproate hepatotoxicity. The developed VPA and 4-ene-VPA PBPK models also pointed to the importance of ontogeny in the differential metabolic pathways observed between pediatrics and adults. The ontogeny-based PBPK model and Clark’s rule are recommended methods for estimating pediatric doses. However, further investigation is required to elucidate more unidentified factors that may affect absorption, distribution, metabolism, and excretion of VPA in pediatrics.

並列關鍵字

drug-metabolizing enzymes ； HepG2 cells ； Ontogeny ； PBPK model ； pediatric dose ； valproic acid

參考文獻

Aarons, L. (2005). Physiologically based pharmacokinetic modelling: a sound mechanistic basis is needed. Br J Clin Pharmacol, 60(6), 581-583. https://doi.org/10.1111/j.1365-2125.2005.02560.x

Google Scholar

Abbott, F. S. Anari M.R. (1999). Chemistry and biotransformation. In: W. Löscher (Ed). Milestones in Drug Therapy-Valproate(pp. 47-75). Basel, Birkhäuser Verlag.

Google Scholar

Addison, R. S., Parker-Scott, S. L., Eadie, M. J., Hooper, W. D., Dickinson, R. G. (2000). Steady-state dispositions of valproate and diflunisal alone and coadministered to healthy volunteers. Eur J Clin Pharmacol, 56(9-10), 715-721. https://doi.org/10.1007/s002280000211

Google Scholar

Allegaert, K., van den Anker, J. (2019). Ontogeny of Phase I Metabolism of Drugs. J Clin Pharmacol, 59 Suppl 1, S33-S41. https://doi.org/10.1002/jcph.1483

Google Scholar

Andrade, R. J., Chalasani, N., Bjornsson, E. S., Suzuki, A., Kullak-Ublick, G. A., Watkins, P. B., Devarbhavi, H., Merz, M., Lucena, M. I., Kaplowitz, N., Aithal, G. P. (2019). Drug-induced liver injury. Nat Rev Dis Primers, 5(1), 58. https://doi.org/10.1038/s41572-019-0105-0

Google Scholar

主題瀏覽