Tanaka Kai, Mochizuki Tatsuki, Baba Shogo, Kawai Shigeto, Nakano Kiyotaka, Tachibana Tatsuhiko, Uchimura Kohsuke, Kato Atsuhiko, Miyayama Takashi, Yamaguchi Tomohito, Nishihara Hiroshi, Terao Kimio, Kato Yasutaka
Translational Research Division, Chugai Pharmaceutical Co., Ltd, 216 Totsuka Totsuka-Ku Yokohama, Kanagawa, Japan.
Department of Biology and Genetics, Laboratory of Cancer Medical Science, Hokuto Hospital, 7-5 Kisen, Inadacho, Obihiro, Hokkaido, Japan.
Sci Rep. 2025 Apr 3;15(1):11403. doi: 10.1038/s41598-025-95823-z.
Predicting the absorption of orally administered drugs is crucial to drug development. Current in vitro models lack physiological relevance, robustness, and reproducibility, thus hindering reliable predictions. In this study, we developed a reproducible and robust culture method to generate a human intestinal organoid-derived monolayer model that can be applied to study drug absorption through a step-by-step approach. Our model showed similarity to primary enterocytes in terms of the drug absorption-related gene expression profile, tight barrier function, tolerability toward artificial bile juice, drug transporter and metabolizing enzyme function, and nuclear receptor activity. This method can be applied to organoids derived from multiple donors. The permeability of launched 19 drugs in our model demonstrated a correlation with human Fa values, with an R value of 0.88. Additionally, by combining the modeling and simulation approaches, the estimated FaFg values for seven out of nine drugs, including CYP3A substrates, fell within 1.5 times the range of the human FaFg values. Applying this method to the drug discovery process might bridge the gap between preclinical and clinical research and increase the success rates of drug development.
预测口服药物的吸收对于药物开发至关重要。当前的体外模型缺乏生理相关性、稳健性和可重复性，从而阻碍了可靠的预测。在本研究中，我们开发了一种可重复且稳健的培养方法，以生成一种源自人肠道类器官的单层模型，该模型可通过逐步方法应用于研究药物吸收。我们的模型在药物吸收相关基因表达谱、紧密屏障功能、对人工胆汁的耐受性、药物转运体和代谢酶功能以及核受体活性方面与原代肠上皮细胞相似。该方法可应用于来自多个供体的类器官。我们模型中19种已上市药物的渗透率与人体Fa值相关，R值为0.88。此外，通过结合建模和模拟方法，包括CYP3A底物在内的9种药物中有7种的估计FaFg值落在人体FaFg值范围的1.5倍之内。将该方法应用于药物发现过程可能会弥合临床前和临床研究之间的差距，并提高药物开发的成功率。
