Experience of malignancies with oral glucose-lowering drugs in the randomised controlled ADOPT (A Diabetes Outcome Progression Trial) and RECORD (Rosiglitazone Evaluated for Cardiovascular Outcomes and Regulation of Glycaemia in Diabetes) clinical trials

Aims/hypothesis Observational and mechanistic studies have suggested a possible relationship between treatment with metformin and decreased incidence of cancer in participants with type 2 diabetes. We extracted data for malignancies from the ADOPT (A Diabetes Outcome Progression Trial) and RECORD (Rosiglitazone Evaluated for Cardiovascular Outcomes and Regulation of Glycaemia in Diabetes) randomised controlled clinical trials, in which the efficacy and/or safety of metformin was assessed in comparison with sulfonylureas and rosiglitazone. Methods Neoplasm occurrences were collected as adverse events in these studies. We reviewed and re-analysed the individual participant data in both studies for serious adverse events, malignancies reported as adverse events and related neoplasms of special interest. Results In ADOPT, 50 participants (3.4%) on metformin and 55 (3.8%) on each of rosiglitazone and glibenclamide (known as glyburide in the USA and Canada) developed serious adverse event malignancies (excluding non-melanoma skin cancers). This corresponds to 1.03, 1.12 and 1.31 per 100 person-years, giving hazard ratios for metformin of 0.92 (95% CI 0.63-1.35) vs rosiglitazone and 0.78 (0.53-1.14) vs glibenclamide. In RECORD, on a background of sulfonylurea, 69 (6.1%) participants developed malignant neoplasms in the metformin group, compared with 56 (5.1%) in the rosiglitazone group (HR 1.22 [0.86-1.74]). On a background of metformin, 74 (6.7%) participants in the sulfonylurea group developed malignant neoplasms, compared with 57 (5.1%) in the rosiglitazone group (HR 1.33 [0.94-1.88]). Conclusions/interpretation The malignancy rates in these two randomised controlled clinical trials do not support a view that metformin offers any particular protection against malignancy compared with rosiglitazone. However, they do not refute the possibility of a difference compared with sulfonylureas.