Designing the matrix: extracellular matrix-informed strategies for bioengineered cancer models

Bioengineered three-dimensional (3D) cancer models have improved the replication of tumour architecture and microenvironmental complexity, yet their predictive value and reproducibility remain inconsistent. A key reason is that extracellular matrix (ECM) context is often treated as a background culture condition rather than as an explicit experimental variable. Cancer cells respond not only to dimensionality, but also to coupled biochemical, structural, mechanical and dynamic matrix cues, including ligand presentation, fibre architecture, porosity, stiffness, viscoelasticity, degradability and remodelling. In this mini-review, we examine ECM-mimicking cancer models through a materials science-informed lens and argue that matrix selection should be guided by model purpose rather than perceived physiological complexity alone. We discuss tissue-derived matrices for preserving tissue-specific and patient-relevant cues, natural and semi-synthetic hydrogels for balancing biological context with usability, synthetic matrices for mechanistic dissection, transport-aware systems for drug-response studies and dynamic hydrogels for modelling time-dependent matrix remodelling. We propose that ECM mimicry should move from a descriptive label to a hypothesis-driven design principle. Making matrix design explicit will improve comparison across models, reduce ambiguity in interpreting cancer phenotypes and strengthen the mechanistic and translational relevance of in vitro cancer systems.