Bridging Realms: Artificial Intelligence Integrates Omics, Generative Models, and Traditional Medicine for Anticancer Drug Innovation

Artificial intelligence (AI) is transforming anti-tumor drug discovery by addressing key challenges across the development pipeline. It enables multimodal data integration (genomics, proteomics, and imaging) through advanced frameworks like integrated graph convolutional network (IGCN) and CrossAttOmics, enhancing target identification and biomarker discovery. Generative drug design using generative adversarial networks (GANs), diffusion models, and variational autoencoders accelerates the creation of novel molecules with optimized properties, such as human carboxylesterase 2A (hCES2A) inhibitors to mitigate irinotecan toxicity. AI-driven prediction of absorption, distribution, metabolism, excretion, and toxicity (ADMET) leverages graph neural networks and organ-on-chip technologies to improve pharmacokinetic profiling and safety assessment. An emerging area is the integration of traditional Chinese medicine (TCM) with AI. Network pharmacology and machine learning (ML) elucidate multi-target mechanisms of TCM compounds (e.g., β-elemene inducing ferroptosis via ferritin heavy chain 1/glutathione peroxidase 4 (FTH1/GPX4) axis). Bayesian optimization and nanoformulations enhance TCM bioavailability. In clinical translation, AI optimizes trials through radiomics for programmed cell death protein 1 (PD-1) response prediction, circulating tumor DNA (ctDNA) analysis for relapse monitoring, and digital pathology. It also addresses drug resistance via single-cell RNA sequencing (scRNA-seq) to identify resistant subclones and TCM agents. Challenges include data heterogeneity, model interpretability, and clinical validation. Future directions focus on interdisciplinary strategies combining quantum computing for molecular simulations, federated learning for data privacy, and AI-personalized TCM formulations. A 2030 roadmap prioritizes building unified TCM-AI databases and target-specific generative algorithms to bridge empirical knowledge with precision oncology. • Pioneers an AI-driven multimodal framework integrating genomics, proteomics, imaging, and clinical data for precision oncology target discovery and validation. • Unlocks Traditional Chinese Medicine (TCM) anti-tumor potential via AI-powered network pharmacology, identifying multi-target agents (e.g., β-elemene, Huanglian Jiedu Tang) to overcome drug resistance and reduce toxicity. • Advances generative drug design using GANs, diffusion models, and variational autoencoders to create novel anti-cancer compounds with optimized pharmacokinetic properties. • Revolutionizes clinical translation through AI-enhanced radiomics, ctDNA fragmentomics, and digital pathology for patient stratification and treatment response prediction. • Resolves key drug development bottlenecks including irinotecan-induced gut toxicity via AI-designed hCES2A inhibitors and PROTAC-mediated resistance reversal.