Background

Food is known to affect drug absorption by delaying gastric emptying time, altering gastrointestinal pH, stimulating bile flow, increasing splanchnic blood flow, or physically interacting with drugs. Although food is known to affect the pharmacokinetics of oral antineoplastic drugs, the relationship between the effects of food and the physicochemical properties of drugs remains unclear.

Methods

In this study, we surveyed the literature on three kinds of pharmacokinetic changes, AUC ratio, C_max ratio and T_max ratio, in the fasted and fed state for 72 oral antineoplastic drugs that were listed on the drug price standard in May 2018 in Japan. We further predicted the physicochemical properties from the 2D chemical structure of the antineoplastic drugs using in silico predictions.

Results

As a result of analyzing the relationship between the effects of food and physicochemical properties, we found that compounds that show increased absorption in the fed state had higher logP and lower solubility in fasted-state simulated intestinal fluid (FaSSIF). However, compounds with delayed absorption had higher solubility in FaSSIF. Furthermore, as a result of decision tree analysis, it was classified as AUC increase with logP ≥4.34. We found that an AUC increase in the fed state did not occur with compounds with low lipid solubilities (logP < 1.59). From these results, it is predicted that 7 compounds out of the 24 compounds for which the effects of food are unknown are at risk for increased absorption in the fed state and that no increase in absorption would occur in 13 compounds.

Conclusion

In this study, we found that drugs that will show increased absorption in the fed state and drugs for which absorption is not dependent on food can generally be predicted by logP. These results suggest that logP can be a useful parameter for predicting the effects of food on drug absorption.