From mono- to multi-cellular in vitro models: reconstructing colorectal cancer complexity for translational and personalized applications

Colorectal cancer (CRC) is a heterogeneous disease shaped by diverse molecular subtypes, tumour microenvironmental context, and rapid adaptation under therapeutic pressure. Treatment outcomes remain highly variable, despite advances in screening, molecular profiling, and targeted and immune-based therapies. This is particularly significant in microsatellite-stable disease, where immune exclusion and resistance mechanisms limit therapeutic efficacy. Disease complexity is only partially captured by current biomarkers, contributing to late-stage failure in drug development and suboptimal patient stratification in the clinic. This Mini Review discusses how advanced <i>in vitro</i> models (three-dimensional patient-derived organoids, heterotypic co-cultures, organ-on-chip platforms, and <i>ex vivo</i> tissue models) accurately reflect CRC pathophysiology and support decision-making in both translational research and precision oncology. These approaches preserve key features of CRC (including tumour architecture, clonal diversity, stromal and immune crosstalk, diffusion barriers, and exposure dynamics), facilitating accurate mechanistic studies, rational combination testing, and functional drug-response profiling. Integrating fit-for-purpose <i>in vitro</i> systems into translational workflows can help de-risk therapeutic development and support more personalized, effective treatment strategies for CRC patients.