Trials of metformin (MF) as an anti-cancer drug have recently begun, enhancing its reputation as a ‘wonder-drug’, effective over a spectrum of life-threatening diseases whilst being highly cost-effective. The cancer story began following reports (wildly exaggerated, in my view) that insulin glargine increased the risk of cancer but this was not seen in patients taking concomitant MF. Allied to its beneficial impact on cardiovascular (CV) outcomes, MFs first-line position in ALL diabetes guidelines seems to have been consolidated. But how strong is the evidence?
The CV benefits of MF were first demonstrated in the United Kingdom Prospective Diabetes Study (UKPDS). Although the UKPDS overall failed to show a benefit of tight glycaemic control on CV outcomes and death, in the obese subgroup treated with MF there was a significant benefit. What is often forgotten is that this subgroup was small, numbering only 753 patients, and in those patients where MF was added to a sulphonylurea (not dissimilar to the second-line recommendation in both NICE and QIPP), there was a significant increase in ‘diabetes-related death’. Comparisons between MF and newer anti-diabetic agents will not be performed (MF is always the baseline therapy) and attempts to analyse routinely collected data are thwarted by selection biases – doctors tend to avoid MF in patients with heart failure and chronic kidney disease i.e. those patients at highest risk of CV events.
If MF were an expensive, new diabetes drug, there would be a clamour to explore these issues. As it’s cheap, the pressure is a little less intense…
Professor Steve Bain