One reason so many cancer drugs fail in late-stage evaluation is that the flat, two-dimensional cell cultures used in early testing are poor imitations of real tumours. A review by University of Manchester scientists, published in the British Journal of Cancer, assesses a fast-maturing alternative: three-dimensional cancer models built from biomaterials, in particular hydrogels, gel-like scaffolds that support cancer cells alongside the other cell types that shape a tumour's behaviour.
The review finds that these lab-grown tumour models can reproduce key features of the real disease, including cell-to-cell interactions and the drug resistance that so often blunts therapy in patients. Better models at the laboratory bench could help researchers identify promising compounds faster and discard doomed candidates earlier, while also reducing the number of animals used in cancer research, a goal the university pursues through its wider programme on replacement, reduction and refinement.
The authors are candid about limits. Hydrogel models cannot yet reproduce whole-body complexity, from immune responses to drug metabolism, that determines a treatment's full efficacy and safety profile, so animal studies will remain important for the foreseeable future. But as materials science and tumour biology converge, the review argues, engineered tumours are becoming serious scientific instruments rather than curiosities.