A Systematic Review of AI-Driven Approaches for Biomarker Discovery and Drug Response Prediction in Precision Cancer Treatment

Abstract

The field of AI has transformed precision oncology by providing new methods for discovering biomarkers and predicting how patients will respond to drugs.The applications proved essential for developing personalized treatment plans which addressed the intricate nature of cancer.The study collected data from research articles published between 2015 and 2025 which examined AI techniques that used genetic transcriptomic and proteomic information to enhance cancer diagnosis and treatment development and patient care.AI-driven approaches brought major progress in two areas through their work on finding useful biomarkers and their ability to predict how individual patients would respond to medications with deep learning (DL) models, such as convolutional artificial neural networks (CNNs) and recurrent artificial neural networks (RNNs) and graph-based artificial neural networks demonstrating better results with multi-omics data which includes different types of cancer data from breast cancer to lung cancer to colorectal cancer to glioblastoma cancer and other cancers.The challenges of data integration and interpretability and clinical implementation have persisted despite the progress made in this field.AI shows great potential to transform cancer treatment according to this review which presents methods for using multi-omics data together with solutions to ethical problems which will lead to better use of medical technology in healthcare.Information from credible research repositories such as PubMed,Scopus, IEEE Xplore, Google Scholar, Nature, and Web of Science were used.