Genetic and systems-level regulation of cancer metabolism: From metabolic reprogramming to AI-driven precision oncology

The metabolic activities of cancer cells undergo complete transformation because they need to maintain their growth while resisting metabolic challenges and environmental dangers from their tumour surroundings. The metabolic changes that occur in cells depend on specific oncogenes together with tumour suppressor genes and stress-response pathways, which control essential bioenergetic and biosynthetic functions. This review presents the current scientific knowledge about genetic regulators, which include MYC, KRAS, PI3K-AKT-mTOR, EGFR, p53, PTEN, and LKB1-AMPK, that control glucose, amino acid, lipid, nucleotide, and mitochondrial metabolism in different human cancers. The research demonstrates that these pathways connect through common metabolic pathways, which produce metabolic flexibility and create complex metabolic patterns that drive tumour diversity and development and resistance to treatment. We present new systems-level frameworks that exceed pathway-based descriptions to show the intricate nature of cancer metabolism. The review investigates how artificial intelligence (AI) and machine learning methods, combined with multi-omics data and genome-scale metabolic models, enable scientists to enhance metabolic phenotyping and discover specific tumour weaknesses and forecast treatment results and combination methods. The study begins with a discussion of present-day obstacles that impede clinical application of research results, which include data inconsistency and the challenges of understanding and testing models. Then it presents upcoming research paths that will develop AI-powered metabolic assessment into biologically understandable and clinically usable tools. The review creates a comprehensive framework that connects genetic control mechanisms with metabolic network functions and AI-driven precision oncology.