Tumour reversion represents a promising field of investigation. The occurrence of cancer reversion both in vitro and in vivo has been ascertained by an increasing number of reports. The reverting process may be triggered in a wide range of different cancer types by both molecular and physical cues. This process encompasses mandatorily a change in the cell-stroma interactions, leading to profound modification in tissue architecture. Indeed, cancer reversion may be obtained by only resetting the overall burden of biophysical cues acting on the cell-stroma system, thus indicating that conformational changes induced by cell shape and cytoskeleton remodelling trigger downstream the cascade of molecular events required for phenotypic reversion. Ultimately, epigenetic regulation of gene expression (chiefly involving presenilin-1 and translationally controlled tumour protein) and modulation of a few critical biochemical pathways trigger the mesenchymal-epithelial transition, deemed to be a stable cancer reversion. As cancer can be successfully 'reprogrammed' by modifying the dynamical cross-talk with its microenvironment thus the cell-stroma interactions must be recognized as targets for pharmacological intervention. Yet, understanding cancer reversion remains challenging and refinement in modelling such processes in vitro as well as in vivo is urgently warranted. This new approach bears huge implications, from both a theoretical and clinical perspective, as it may facilitate the design of a novel anticancer strategy focused on mimicking or activating the tumour reversion pathway.