Does Metformin Therapy Impact CVD in T2D? It’s Complicated

MADRID — Metformin is taken every day by more than 200 million people worldwide. Its value as a glucose-lowering agent has been established for decades. But does metformin hold similar promise for reducing rates of cardiovascular disease (CVD)? 

This question was the focus of a thought-provoking presentation by Simon Griffin, MD, leader of the Prevention of Diabetes and Related Metabolic Disorders in High Risk Groups program at Cambridge University, Cambridge, England, at the European Association for the Study of Diabetes (EASD) 2024 Annual Meeting.

Griffin began by noting that trials such the Steno-2 trial have shown that lifestyle interventions alone in patients with type 2 diabetes (T2D) reduce the risk for cardiovascular (CV) events and death. Similarly, the 30-year follow-up of the Da Qing study found there was a significant reduction in the risk for CV events associated with a lifestyle intervention for people with impaired glucose tolerance.

However, using rosiglitazone to prevent diabetes does not have a beneficial role in reducing the risk for CV events, as the DREAM study showed.

“The moment you stop the tablet that lowers glucose, the incidence of diabetes is the same as if you hadn’t given participants the tablet in the first place,” Griffin said. “The more difficult conclusion to swallow is that rosiglitazone actually increases the risk of CV events.”

Metformin’s Indisputable Role in Diabetes

Metformin is a resounding success reducing the risk for diabetes, said Griffin.

“In the Indian Diabetes Prevention Program me [native Asian Indians], 250 mg twice a day of metformin reduced diabetes incidence by 26% in the 531 people with impaired glucose tolerance, despite of a relatively small change in weight.”

And, in the US Diabetes Prevention Program Research Group trial, where people were randomized to receive placebo, 850 mg of metformin twice a day, or a lifestyle-modification program, there were “dramatic reductions” in the incidence of T2D: 31% with metformin and 58% with the lifestyle modification program. Furthermore, metformin was cost-saving, Griffin said.

“Metformin is good if you’ve got diabetes and it’s cost saving if you’ve got so-called pre-diabetes. And unlike rosiglitazone, metformin appears to have a legacy effect; the moment you stop the tablet, the benefits don’t appear to disappear.”

Does Metformin Reduce CVD in Diabetes? 

To answer the question posed by his presentation, Griffin turned to the results of several meta-analyses, with varying inclusion criteria.

“The trials are very different to current trials, and frequently CV events weren’t the primary outcome; they were merely adverse events documented somewhere in the text of the paper,” he noted. In addition, “most of the CV endpoint trials of metformin didn’t feature a placebo group, and unsurprisingly, the meta-analyses are dominated by the UK Prospective Diabetes Study (UKPDS).”
 

They are informative, nonetheless. He highlighted the following studies in particular:

• Selvin reported a significant reduction in CV mortality of 26% with metformin, but a nonsignificant 15% reduction in CV morbidity.
• Lamanna reported a nonsignificant 6.3% reduction in CV events and a 10% reduction in the risk for myocardial infarction.
• Griffin’s group reported an 11% reduction in the risk for myocardial infarction.
• In an update of Lamanna’s review, Minami reported a 48% reduction in CV events and a non-significant 20% reduction in-all cause mortality.
• The UKPDS reported a consistent 31% reduction in the risk for CV events over time associated with metformin.

Should Metformin Be Added to the Water Supply?

Griffin then asked, does metformin reduce CV events among people with impaired glucose tolerance or those with normoglycemia? 

“Unfortunately, after 3 years of 850 mg twice a day of metformin vs placebo in the US Diabetes Prevention Program, there was no effect on CV events.”

However, an important consideration in assessing metformin trials is that they’re mostly old and different from more recent CV endpoint trials, he noted. Because the drug is off patent and a first-line treatment for diabetes and prediabetes in many countries, it’s difficult to do a CV endpoint trial with it.

“What you might do instead is a metformin trial in people who haven’t got diabetes, but with intermediate endpoints,” Griffin said.

This is exactly what researchers did in the CAMERA trial, observing that despite the effect on weight and A1c, there was no impact on carotid intima-media thickness or carotid plaque score.

By contrast, a recent study found that disrupting metformin-targeted genes was associated with a 37.8% lowering of the risk for CVD in the general population.

To reduce the burden of CVD related to hyperglycemia, a complimentary approach is needed to shift the population distribution of underlying determinants, per Rose’s prevention paradox, Griffin said. The paradox shows a population-based preventive health measure brings large public health benefits, although it offers little benefit to each individual participant. This means a study to lower the CVD burden of hyperglycemia would need an individualized component as well as a public health component.

Asked during the Q&A whether he could provide examples of countries where such an approach has worked, Griffin pointed first to the Västerbotten intervention.

“This was an individual-based approach to glucose tolerance tests on 40-, 50- and 60-year-olds, encouraging them to change their lifestyle,” he said.

“But it also had a public health component of trying to change the food offerings in local supermarkets from the original, very fatty northern Scandinavian food to a slightly less fatty northern Scandinavian food and that appears to have had a positive impact on risk of CVD in that population.”

Another example is The North Karelia Project, which led to significant reductions in CVD mortality and prompted an aggressive salt-reduction program and other public health measures in Finland.

“Should we give up on metformin as a preventive therapy for cardiovascular disease in people without diabetes,” another meeting attendee asked.

“No, I would say not,” Griffin responded. “If someone’s got so-called prediabetes, then you should encourage them to change their lifestyle. You should treat their blood pressure aggressively, treat their cholesterol aggressively, and offer people metformin. I might want to follow them up and remeasure some of their blood tests and possibly even suggest they take added B12 or measure their B12,” because metformin can reduce B12 levels.

“The question is, should you give metformin to people with normoglycemia? You give statins to people whose cholesterol is not very high. You give blood pressure treatment to people whose blood pressure is not very high just because they’re at high CV risk,” he pointed out. In short, “should metformin be put in the water supply or doled out like statins?”

The VA-Impact trial is looking at just that. The trial is recruiting people without diabetes but with existing CVD. 

“I think we should wait for those results in 2029,” Griffin concluded.

Griffin declared honoraria from AstraZeneca and Eli Lilly and Company for postgraduate education meetings. Over 5 years ago, he undertook an independent feasibility study for a CV endpoint trial of metformin among people with prediabetes for which Merck Serono provided the active drug and placebo.

Marilynn Larkin, MA, is an award-winning medical writer and editor whose work has appeared in numerous publications, including Medscape Medical News and its sister publication MDedge, The Lancet (where she was a contributing editor), and Reuters Health.