Modern clinical oncology faces an unprecedented data complexity that exceeds human analytical capacity, making artificial intelligence (AI) integration essential rather than optional. This review examines the dual impact of AI on productivity enhancement and creative discovery in cancer care. We trace the evolution from traditional machine learning to deep learning and transformer-based foundation models, analyzing their clinical applications. AI enhances productivity by automating diagnostic tasks, streamlining documentation, and accelerating research workflows across imaging modalities and clinical data processing. More importantly, AI enables creative discovery by integrating multimodal data to identify computational biomarkers, performing unsupervised phenotyping to reveal hidden patient subgroups, and accelerating drug development. Finally, we introduce the FUTURE-AI framework, outlining the essential requirements for translating AI models into clinical practice. This ensures the responsible deployment of AI, which augments rather than replaces clinical judgment, while maintaining patient-centered care.

The emergence of novel cytotoxic agents, multikinase inhibitors, and various antibody-based therapies has significantly expanded salvage therapy options for metastatic colorectal cancer (mCRC). Consequently, establishing the optimal treatment sequence for patients has become a formidable clinical challenge. Emerging evidence highlights the value of comprehensive biomarker assessment, particularly longitudinal monitoring of circulating tumor DNA (ctDNA), to capture dynamic molecular changes during treatment. Liquid biopsy-based technologies now enable real-time tracking of molecular alterations, supporting truly personalized therapeutic decision-making. Furthermore, prior treatment exposure and residual toxicities must be carefully considered to balance efficacy, safety, and quality of life. This review provides a comprehensive overview of the current salvage-line landscape for mCRC, discusses the clinical utility of ctDNA as a predictive and prognostic tool, and proposes integrated strategies to optimize therapeutic outcomes in the evolving era of precision medicine.

Perioperative treatment for colorectal cancer (CRC) is undergoing a paradigm shift from uniform cytotoxic regimens toward strategies guided by tumor location, microsatellite instability (MSI)/mismatch repair (MMR) status, and recurrence risk. Four clinically relevant subgroups now shape decision-making: microsatellite stable (MSS)/proficient MMR (pMMR) colon cancer, microsatellite instability-high (MSI-H)/deficient MMR (dMMR) colon cancer, MSS/pMMR rectal cancer, and MSI-H/dMMR rectal cancer. In MSS/pMMR colon cancer, adjuvant therapy is being refined through risk-adapted treatment duration and selective use of neoadjuvant chemotherapy. In MSS/pMMR rectal cancer, total neoadjuvant therapy, selective omission of pelvic radiotherapy, and watch-and-wait strategies are redefining treatment sequencing and organ preservation. In MSI-H/dMMR colon cancer, immune checkpoint inhibitor (ICI) therapy has shown marked activity in both adjuvant and neoadjuvant settings. In MSI-H/dMMR rectal cancer, neoadjuvant ICI therapy is being explored as a non-operative organ preservation strategy. In parallel, circulating tumor DNA (ctDNA)-based minimal residual disease (MRD) assessment is emerging as a tool for postoperative escalation, de-escalation, and surveillance. Across these settings, the maturity of the evidence varies widely, ranging from established phase III standards to guideline-endorsed but still single-arm approaches and investigational strategies that require prospective validation before routine adoption. This review summarizes the current evidence and remaining challenges in biology-, site-, and risk-adapted perioperative management of CRC.