Overcoming Drug Resistance in Cancer Therapy using Machine Learning and Artificial Intelligence

Drug resistance remains a central barrier to durable responses across cancer therapies, spanning targeted agents, cytotoxic chemotherapy, endocrine therapies, radiotherapy sensitizers, and immuno-oncology combinations. Resistance can arise through diverse, evolving mechanisms—tumor-intrinsic genomic and epigenomic alterations, adaptive transcriptional and proteomic reprogramming, phenotypic plasticity and lineage switching, microenvironmental pressures, altered drug transport and metabolism, and therapy-induced selection—often culminating in cross-resistance and relapse. At the same time, the increasing scale and complexity of multi-omic, imaging, pharmacological, and real-world clinical data create an urgent need for computational approaches that can integrate heterogeneous evidence and convert it into actionable therapeutic hypotheses.This Research Topic focuses on how machine learning (ML) and artificial intelligence (AI) can be leveraged to understand, predict, and ultimately overcome drug resistance in cancer therapy. We welcome contributions that develop or apply ML/AI methods to identify resistance mechanisms, stratify patients by resistance risk, forecast treatment trajectories, and prioritize rational therapeutic strategies such as drug combinations, scheduling/dosing adaptations, or next-line therapies. Emphasis is placed on clinically and pharmacologically relevant modeling that connects computational outputs to testable biology and translational decision-making.Areas of interest include, but are not limited to: ML/AI-driven discovery of resistance biomarkers and signatures (including single-cell and spatially resolved data); prediction of drug response and resistance evolution from multi-modal inputs (genomics, transcriptomics, epigenomics, proteomics, metabolomics, pathology, radiology, liquid biopsy, and longitudinal monitoring); computational inference of pathway rewiring and target dependencies under drug pressure; AI-enabled repurposing and de novo design of anti-cancer drugs or resistance-modulating agents; and systematic identification of synergistic drug combinations, sequential regimens, or adaptive therapy strategies. Submissions addressing pharmacology-focused questions are especially encouraged, including modeling of exposure–response relationships, PK/PD-informed learning, drug–drug interaction prediction, dose optimization, toxicity-aware combination design, and integration of preclinical pharmacology with clinical outcomes.We also encourage work that advances methodological rigor and interpretability, including uncertainty quantification, causal and counterfactual inference, fairness and bias assessment, external validation across cohorts and platforms, and reproducible pipelines suitable for clinical translation. Studies using real-world data, prospective/retrospective clinical datasets, and experimentally validated computational predictions are of particular interest, as are frameworks that bridge preclinical models (organoids, PDX, CRISPR screens) with patient-level response.In addition, this Topic welcomes studies leveraging computer-aided drug design and molecular modeling approaches integrated with ML/AI to address drug resistance. This includes ML-enhanced quantitative structure–activity relationship (QSAR) modeling, structure-based virtual screening, molecular docking, molecular dynamics simulations, and binding free energy calculations to characterize resistance-associated mutations and identify next-generation inhibitors. Contributions integrating ADMET prediction and pharmacokinetic property optimization using AI-driven frameworks are strongly encouraged. Studies combining molecular simulation data with ML models to improve the prediction of drug efficacy, selectivity, and resistance liability are particularly relevant, especially when linked to experimental validation or translational pharmacology applications.By bringing together oncology, pharmacology, and ML/AI communities, this Topic aims to catalyze approaches that move beyond resistance description toward practical strategies to prevent, delay, or reverse treatment failure and improve patient outcomes.***Please note: Manuscripts consisting solely of bioinformatics, computational analysis, or predictions of public databases which are not accompanied by validation (independent clinical or patient cohort, or biological validation in vitro or in vivo, which are not based on public databases) are not suitable for publication in this journal. Topic editors: Mohammed Er-rajy, Teresita Padilla-Benavides, Hugo F. Posada-Quintero Source: https://www.frontiersin.org/research-topics/78834/overcoming-drug-resistance-in-cancer-therapy-using-machine-learning-and-artificial-intelligence