Pfizer Inc, La Jolla, CA, USA.
Pfizer Inc, Groton, CT, USA.
Int J Med Sci. 2021 Sep 24;18(16):3718-3727. doi: 10.7150/ijms.65040. eCollection 2021.
The objective for the present analyses was to evaluate the utility of physiologically-based pharmacokinetic (PBPK) modeling for prediction of the pharmacokinetics (PK) in Chinese and Japanese populations with a panel of Pfizer internal compounds. Twelve compounds from Pfizer internal development pipeline with available Westerner PK data and available PK data in at least one of the subpopulations of Japanese and Chinese populations were identified and included in the current analysis. These selected compounds represent various elimination pathways across different therapeutic areas. The Simcyp PBPK simulator was used to develop and verify the PBPK models of individual compounds. The developed models for these compounds were verified by using the clinical PK data in Westerners. The verified PBPK models were further used to predict the PK of these compounds in Chinese and Japanese populations and the predicted PK parameters were compared with the observed PK parameters. Ten of the 12 compounds had PK data in Chinese, and all the 12 compounds had PK data in Japanese. In general, the PBPK models performed well in predicting PK in Chinese and Japanese, with 8 of 10 drugs in Chinese and 7 of 12 drugs in Japanese has AAFE values less than 1.25-fold. PBPK-guided predictions of the relative PK difference were successful for 75% and 50%, respectively, between Chinese and Western and between Japanese and Western of the tested drugs using 0.8-1.25 as criteria. In conclusion, well verified PBPK models developed using data from Westerners can be used to predict the PK in Chinese and Japanese populations.
本分析的目的是评估生理药物动力学（PBPK）模型在预测中国人和日本人药代动力学（PK）方面的应用。从辉瑞内部开发管道中选择了 12 种化合物，这些化合物具有可用的西方人 PK 数据，并且至少在中国人和日本人的一个亚群中具有 PK 数据。这些选定的化合物代表了不同治疗领域的各种消除途径。使用 Simcyp PBPK 模拟器开发和验证了各个化合物的 PBPK 模型。通过使用西方人临床 PK 数据对这些化合物的开发模型进行验证。进一步将验证后的 PBPK 模型用于预测这些化合物在中国人和日本人中的 PK，并将预测的 PK 参数与观察到的 PK 参数进行比较。这 12 种化合物中有 10 种在中国有 PK 数据，所有 12 种化合物在日本都有 PK 数据。总的来说，PBPK 模型在预测中国人和日本人的 PK 方面表现良好，10 种药物中有 8 种在中国，12 种药物中有 7 种在日本的 AAFE 值小于 1.25 倍。使用 0.8-1.25 作为标准，PBPK 指导的预测在中国人和西方人之间以及日本人与西方人之间分别成功预测了 75%和 50%的测试药物的相对 PK 差异。总之，使用西方人数据验证良好的 PBPK 模型可以用于预测中国人和日本人的 PK。
