Comparative properties of experimental tumour models. In vitro assays of tumour growth and response to therapy can be conducted in two dimensions or three dimensions - the latter more closely approximating the biology of solid tumours than a simple monolayer. Cultures can be enhanced by the addition of matrix proteins and/or host cells and can be adapted to measure not only tumour cell proliferation, but also additional cancer hallmarks such as invasion. Standard in vivo assays depend upon the transplantation of established human tumour cell lines into athymic (immune-incompetent) hosts. These models are relatively simple and easy to use, but are increasingly complemented by genetically engineered mice harbouring targeted genetic mutations which render them susceptible to developing mammary cancers. The figure summarises key advantages and disadvantages of each model and means by which their clinical relevance and utility might be enhanced. Based on a figure provided courtesy of Claire Nash in Dr Valerie Speirs’ group (University of Leeds).