The study was conducted as a cancer centre in America between 2004 and 2008. The study included 973 patients with pancreatic adenocarcinoma and 863 controls. Medical histories were collected by personal interview and frequency of use of antidiabetic therapies was compared between cases and controls.
In patients with diabetes, metformin was found to significantly reduce the risk of pancreatic cancer. Insulin and insulin secretagogues (sulphonylureas and metaglinides) significantly increased the risk while glitazones increased the risk but not significantly.
- Metformin - Odds ratio (OR) 0.38 [95% confidence interval (CI) 0.22 - 0.69, p=0.001]
- Insulin - OR 4.99 [95% CI 2.59 - 9.61, p<0.001]
- Insulin secretagogues - OR 2.52 [95% CI 1.32 - 4.84, p=0.005]
- Glitazones - OR 1.55 [95% CI 0.78 - 3.07, p=0.213]
These differences remained in never-users of insulin but the risk associated with insulin secretagogues became non-significant in patients who had diabetes for longer than 2 years.
The authors of this study highlight several potential limitations including the design of the study being open to reporting bias, the study having been conducted in one location which may mean the results are not representative of a wider population and the sub group analyses only used small numbers of patients. They recommend that these observations should be confirmed in larger studies.
Action: Metformin is already recommended as first line treatment in type 2 diabetes. An accompanying editorial notes that these findings may "provide an additional incentive for patients and physicians to follow this recommendation".