基于tumor cord-PBPK偶联模型的抗肿瘤药物药动学预测研究

The pharmacokinetic models of antitumor drugs mainly focus on the prediction of drug distribution in normal tissues instead of tumor tissues in animals. We have confirmed that the physiologically based pharmacokinetic model (PBPK) could predict the average concentration of adriamycin in the tissues of tumor bearing mice, treating heterogeneous tumor tissue as homogeneous tissue, inaccurately. While “tumor cord” model could describe the uptake and distribution of drugs in vascularized tumor tissue, this model has limitation in its predictive study, which is lacking plasma concentration as an input value to predict reliably. We intended to establish a new model, coupling the “tumor cord” model with the PBPK model, which will combine the advantages of describing tumor tissues and interpreting systemic distribution of drugs, to predict the pharmacokinetic behavior of antitumor drugs more accurately. Based on this coupling model, distributions of drugs in the tumor tissues and others in the tumor-bearing mice were predicted precisely. Then, the PBPK model in rat was extrapolated to predict the drug pharmacokinetics in human, and the parameter sensitivity was analyzed to evaluate impact factor on the bio-distribution of antitumor drugs. In addition, the pharmacokinetic behaviors of different drugs in chemotherapy, simultaneously, were predicted within coupling model, providing guide of rational dosing regimen. This model will reveal the interaction between pharmacokinetics at the system level and distribution of drugs in tumor, which could improve understanding of antitumor drugs in the experimental study and clinical application.

目前针对抗肿瘤药物的药动学模型主要集中于对药物在动物正常组织的分布预测，缺少对其在肿瘤组织分布的模型化预测研究。前期工作表明，对阿霉素在荷瘤小鼠的组织分布预测中，生理药动学模型PBPK只能将肿瘤按均质性组织处理，预测其肿瘤组织的平均浓度。Tumor cord模型可描述药物在血管化肿瘤组织的摄取与分布，但其需要将血药浓度作为输入值，限制了其预测性研究。本课题拟建立一种新的模型方法，搭建tumor cord-PBPK偶联模型，对抗肿瘤药物的药动学进行全面描述。基于搭建的偶联药动学模型，预测抗肿瘤药物在荷瘤小鼠肿瘤组织及其它组织分布情况；将鼠药动学模型种属外推，预测人体药动学；基于模型进行参数敏感性分析，探讨影响药物组织分布的因素；对联合用药后药物的药动学进行预测，指导给药方案设计。该模型有助于揭示抗肿瘤药物在系统水平的药动学与肿瘤组织分布间的关系，对抗肿瘤药物的研究和临床应用具有重要指导意义。

抗肿瘤药物药动学性质差、体内分布广泛，目前在临床上应用的大量抗肿瘤药物几乎都有着不同程度的毒副作用。抗肿瘤药物的有效性和安全性很大程度上取决于在治疗剂量下药物分布到肿瘤组织的能力和在其他正常组织的分布情况。建立可靠的生物药剂学模型，评价抗肿瘤药物的药动学性质，对于抗肿瘤药物的研究、发展以及合理应用有着极为重要的作用。然而由于肿瘤组织异质性的生理特征，目前广泛报道的PBPK 模型能够预测药物在体内除肿瘤组织外的其他正常组织的分布，在预测肿瘤组织分布上存在着一定的局限性。本项目将描述肿瘤组织的tumor cord模型与PBPK模型结合，建立了抗肿瘤药物的药动学预测模型PBPK-tumor cord偶联模型。选择临床广泛应用的抗肿瘤药物阿霉素（doxorubicin，DOX）为模型药物，首先对其在小鼠的体内药动学行为进行研究，并探究了联合化疗时紫杉醇（paclitaxel，PTX）对DOX及其活性代谢产物阿霉素醇药动学行为的影响；通过收集DOX相关理化参数、生理学参数和药动学参数等，建立DOX小鼠生理药动学PBPK模型，预测DOX在小鼠体内分布情况；进行种属外推至大鼠和人体，预测DOX人体药动学行为；利用PBPK模型对DOX和PTX联合化疗过程中DOX的药动学行为进行模拟，探究PTX对DOX体内药动学行为的影响；基于成功建立的PBPK模型的预测结果，作为tumor cord模型输入值，搭建tumor cord模型，在搭建tumor cord模型过程中，选择两种处理方式:一种是假设癌细胞密度是均匀的，沿血管方向柱状肿瘤的性质没有差别，只需考虑垂直于肿瘤血管长度的横切面，另一种是为了更好地模拟异质性肿瘤微环境，对围绕在毛细血管的肿瘤细胞根据其与毛细血管的距离进行了区分，基于建立的模型，对DOX在tumor cord模型中肿瘤内分布行为进行模拟研究。