According to a commentary, these findings represent the first randomised controlled trial evidence of an effect of metformin on an established surrogate endpoint for cancer. It notes that though well designed, this study has a few limitations, such as the inclusion of a small number of selected patients at high risk of adenoma recurrence and absence of data about safety or efficacy of other conventional doses of metformin. In addition, patients did not undergo a clearing colonoscopy to confirm absence of polyps before randomisation and were assessed for incident recurrent polyps after only 1 year, thus, some of the polyps found might have included prevalent lesions missed at the baseline examination rather than incident tumours. Nevertheless, the commentator believes these exciting findings should sustain optimism that metformin might have a role in cancer prevention, thereby encouraging and informing the development of more definitive randomised controlled trials.