Artificial intelligence applications in oxaliplatin-based chemotherapy for colon cancer: advancing prognosis, toxicity prediction, and dose personalization.

Colorectal cancer (CRC) remains a significant cause of cancer deaths worldwide. Oxaliplatin-based regimens, such as FOLFOX, which form the cornerstone of therapy, are associated with variability in patient responses and toxicities, most notably peripheral neuropathy, and often limit long-term benefit. In recent years, the use of Artificial Intelligence (AI) in oncology has expanded significantly. The beneficial role of AI lies in its unprecedented ability to rapidly process and integrate high-dimensional datasets (e.g., genomic, radiomic, clinical) to uncover subtle, nonlinear relationships that conventional statistical methods cannot access. Thereby, AI is transforming the empirical approach to chemotherapy into a truly predictive science. Unlike previous reviews that broadly discuss AI in oncology, this review focuses specifically on oxaliplatin, drawing on genomic, transcriptomic, radiomic, and body composition data to refine patient stratification and anticipate potential adverse effects. It also highlights emerging AI-driven strategies for identifying transporter inhibitors and protective agents to mitigate neurotoxicity, particularly in patients with CRC. By moving beyond retrospective prediction, the review illustrates how AI can enable proactive, individualized treatment planning and safer dosing. It also summarizes and clarifies the methodologies used in machine learning models, serving as a reference for readers interested in this field. Collectively, the review highlights AI as a transformative tool for advancing precision oncology in oxaliplatin-based CRC care and provides the first comprehensive synthesis of AI applications in oxaliplatin therapy, with a focus on prognosis, toxicity prediction, and dose personalization.