AI-Assisted Prediction of Absorption-Related Drug Interactions: Mechanisms, Selected Drugs, and Emerging Applications

Abstract

Drug absorption is a critical determinant of systemic drug exposure and therapeutic response. Interactions occurring during the absorption phase can modify the amount and rate of drug entering the circulation through changes in gastrointestinal pH, dissolution, solubility, gastric emptying, intestinal motility, food intake, chelation, transport activity, and intestinal metabolism. Conventional prediction of these interactions depends on pharmacokinetic studies, physicochemical measurements, experimental models, and clinical observations, which can be time-consuming when many drug combinations need to be evaluated. Artificial intelligence (AI) and machine learning methods have emerged as useful computational tools for integrating physicochemical, pharmacokinetic, molecular, and clinical data. These methods can be applied to recognise patterns associated with altered drug absorption and to predict potential interactions before extensive experimental testing. AI-based models may be particularly useful for evaluating pH-dependent absorption, transporter-mediated processes, food–drug interactions, and interactions caused by complex formation or changes in gastrointestinal conditions. Selected drugs such as levothyroxine, ciprofloxacin, tetracycline, metformin, ketoconazole, dabigatran, and tacrolimus demonstrate the clinical relevance of absorption-related interactions and provide suitable examples for computational prediction. The review integrates mechanistic pharmacokinetic concepts with AI-assisted prediction strategies, highlighting how computational models support in early screening of absorption-related interactions while emphasizing the continuous importance of pharmacokinetic validation for clinical interpretation.