Abstract
Despite advances in screening and therapy, colorectal cancer (CRC) remains a leading cause of cancer-related mortality worldwide, underscoring the need for early detection and for predicting treatment efficacy. This review highlights circulating cell-free DNA (cfDNA) fragmentomics as a promising non-invasive approach for tumor detection and disease monitoring. We focus on fragmentomic features - such as fragment size distributions, fragment-end motifs, and epigenetic signals - which, when integrated into machine-learning models, have shown strong performance in distinguishing patients with CRC from healthy controls. Emerging evidence indicates that, these signatures may support early-stage detection, track disease progression, and predict pathologic complete response (pCR), thereby enabling more personalized treatment strategies. We also discuss the potential role of fragmentomics in non-operative management, including "watch-and-wait" approaches. However, important gaps remain in clinical translation; prospective trials and standardized assays/analysis pipelines are required to validate these findings and define their real-world utility.