Colorectal cancer has long served as the model organism of cancer evolution, and a new review in Nature Reviews Gastroenterology & Hepatology argues that the model is now changing clinical practice. The authors, led by Christopher C. T. Sng, Trevor A. Graham and Andrea Sottoriva, synthesise the latest evidence on how somatic alterations accumulate in the genomes and epigenomes of colorectal cells, tracing the disease from initial transformation through metastatic dissemination to treatment resistance. Their central message is that the pattern of these alterations reveals both the dynamics and the molecular drivers of progression, and that evolutionary principles are actively redefining therapeutic strategy.
The review’s first structural claim concerns timing. Colorectal cancers display what the authors call ‘Big Bang’ evolutionary dynamics: the canonical driver mutations are acquired early during transformation, after which the tumour undergoes a functionally neutral expansion. That expansion generates extensive intratumoural genetic heterogeneity, meaning that a single biopsy samples only one branch of a sprawling phylogenetic tree. This reframes a long-standing assumption in oncology, because it implies that the genetic diversity visible across a tumour mass is largely a product of early events rather than of ongoing, targeted selection at every stage.
The next step in the sequence is equally instructive. Adenomas and carcinomas, the review notes, share a similar driver mutational burden, so the transition from benign growth to malignancy is not marked by a fresh wave of point mutations. Instead, the adenoma–carcinoma transition is characterised by substantial copy number alterations and epigenetic rearrangements. In other words, the malignant switch is written less in the sequence of the genome than in its architecture and its regulatory state, a distinction that matters for anyone hoping to predict progression from mutation panels alone.
Metastasis, by contrast, emerges as a stringent evolutionary bottleneck. The review describes how metastatic seeding is carried out by one or a few subclones rather than by the tumour as a whole. Yet the search for a metastasis-specific genetic signature has largely come up empty: few drivers show metastasis enrichment, and no universal metastatic drivers are confirmed. What appears central to dissemination instead is phenotypic plasticity, the capacity of tumour cells to switch state, which the authors place at the heart of the metastatic process.
Treatment resistance follows a similar logic of mixed mechanisms. The review states that resistance is driven by both genetic and non-genetic routes, including phenotypic switching into reversible drug-tolerant persister states. Crucially, this resistance is often polyclonal, existing across distinct populations within one tumour simultaneously. That finding has direct consequences for targeted therapy design, because a strategy that eliminates a single resistant clone may leave several others untouched, each capable of repopulating the disease.
The immune dimension is woven through the entire trajectory rather than added at the end. Immune selection shapes colorectal cancer from an early stage through immunoediting and immune escape, and tumours span a continuum of immunogenicity. The review identifies the immune microenvironment and clonal neoantigen burden as the factors shaping response to checkpoint inhibitors, linking evolutionary history directly to immunotherapy outcome.
From these observations the authors derive a therapeutic programme. Evolution-informed strategies, they write, target intratumoural heterogeneity and plasticity using liquid biopsy tracking, molecular hyperselection and adaptive dosing, with the explicit aim of anticipating and countering resistance rather than reacting to it after the fact. Liquid biopsy tracking would monitor the evolutionary state of a tumour in real time; adaptive dosing would adjust drug pressure to steer competition between sensitive and resistant subclones instead of maximally suppressing one and freeing the other.
The review also points to translational frontiers in evolution-informed therapy, arguing that the epigenome and cell plasticity deserve the same analytical attention as DNA sequence, given their interrelationship with genetic changes and their role in progression and resistance. For a disease that remains common and accessible for sampling, with malignant progression unfolding through clearly defined morphological stages, colorectal cancer offers an unusually complete laboratory for testing whether evolution-aware medicine can outmanoeuvre a tumour that evolves as fast as its patients’ treatments can adapt.
Subject of Research: The genomic and epigenomic evolution of colorectal cancer
Article Title: The genomic and epigenomic evolution of colorectal cancer
Article References: Sng, C. C. T., Starling, N., Graham, T. A., & Sottoriva, A. (2026). The genomic and epigenomic evolution of colorectal cancer. Nature Reviews Gastroenterology & Hepatology. https://doi.org/10.1038/s41575-026-01265-x
Image Credits: AI Generated
DOI: 10.1038/s41575-026-01265-x
Keywords: colorectal cancer, cancer evolution, epigenomics, intratumoural heterogeneity, metastasis, treatment resistance, phenotypic plasticity, immunoediting, liquid biopsy, adaptive dosing, genomic, epigenomic
News Source: Nathaniel Bowman. (November 5, 2026). Colorectal cancer evolves in bursts, and evolution is rewriting treatment. Scienmag.
