SAN DIEGO — Use of metformin may offer some atherosclerosis protection for patients with type 1 diabetes but does not appear to improve glycemic control, new data suggest.

Findings from the double-blind, placebo-controlled Reducing with Metformin Vascular Adverse Lesions (REMOVAL) trial were presented June 11 here at the annual the American Diabetes Association (ADA) 2017 Scientific Sessions and were also simultaneously published in Lancet Diabetes & Endocrinology.

The multicenter trial randomized a total of 428 adults aged 40 and older with type 1 diabetes and at increased cardiovascular risk to daily metformin 1000 mg twice daily or placebo as adjuncts to insulin. The primary end point — significant reduction in mean far-wall carotid artery intima-medial thickness (cIMT) at 3 years — was not met, although metformin did reduce maximal far-wall cIMT (which includes plaque).

And it did significantly reduce HbA1c, but only for the first 3 months of the 3-year study.


But metformin did lower body weight, LDL cholesterol, and insulin dose, suggesting that it might provide additional benefit for type 1 diabetes patients, according to lead investigator John R Petrie, MD, PhD, professor of diabetic medicine at the University of Glasgow, Scotland.   

This "changes the way we think about metformin. The guidelines suggest you can add it in obese type 1 diabetes patients to reduce glucose, but we didn't see much evidence of glucose reduction beyond 3 months. Obviously, don't use it for glucose lowering. But consider using it for reducing weight and LDL cholesterol and possibly atherosclerosis prevention."

Indeed, Partha Kar, MD, FRCP, clinical director of diabetes at Portsmouth Hospitals NHS Trust, United Kingdom, told Medscape Medical News that the findings essentially reinforce the way he's already using metformin in type 1 patients to reduce body weight in those with increased body mass index (BMI) and insulin resistance.

But, given that REMOVAL didn't meet its primary cIMT end point — a surrogate marker for atherosclerosis — Dr Kar said the findings don't convince him to broaden his use of metformin in type 1 diabetes patients in hopes of reducing their cardiovascular risk.

"It's not going to ramp up my prescriptions. I will still continue to use [metformin] in patients I think are appropriate," he said, also noting, "It's used off-label in type 1 diabetes, and there are no data here for any regulatory body to license it."

And in an editorial accompanying Dr Petrie's paper, Eberhard Standl of Munich Diabetes Group eV at Helmholz Center, Neuherberg, Germany, had this to say: "The findings from REMOVAL do not necessarily mean removal of metformin from use in type 1 diabetes, but the primary core question of whether metformin has a practically relevant role in the treatment of (specific) patients with type 1 diabetes has yet to be substantiated."

"Nevertheless," he adds, "for patients with type 1 diabetes currently being exposed to experimental metformin therapy it is helpful that REMOVAL — among the many interesting details reported — did not show an increased risk for minor or major hypoglycemia, but rather suggested the need for regular monitoring for vitamin B12 deficiency."

REMOVAL's Mixed Results 

All of the patients in REMOVAL had at least three of 10 specified CVD risk factors, including BMI ≥ 27 kg/m2, HvA1c > 8.0%, strong family history of CVD, and/or current smoking.


They had a mean age of 55.5 years, diabetes duration of 33.8 years, HbA1c 8.05%, and BMI 28.5 kg/m2. About a third were using insulin pumps. Their mean blood pressure was 130/72 mm Hg and LDL cholesterol 2.2 mmol/L, with 73% and 82%, respectively, on antihypertensives and statins.


After a 3-month optimization phase, the patients were randomized to 1000 mg twice daily of oral metformin or placebo, in combination with titrated insulin therapy for 36 months.


The primary outcome — difference in mean within-person cIMT slopes (metformin relative to placebo) — was -0.005 mm per year, nonsignificantly in favor of metformin (= .167). Prespecified sensitivity analyses showed similar results, but a tertiary outcome — averaged maximal cIMT — did show a slower progression over time with metformin (difference in slope -0.013 mm/year, = .0093).

HbA1c levels were significantly reduced by metformin at 3 years (-0.13%, = .0060), but this was entirely due to a drop within the first 3 months (-0.24%, < .0001) and not thereafter. There was no reduction in average insulin dose requirement over the 3 years (-0.005 units/kg, = .5450).   


However, metformin did significantly reduce body weight (-1.17 kg, < 0.0001) and LDL cholesterol (-0.13 mmol/L, = .0117).


There was a significant increase in estimated glomerular filtration rate with metformin (4.00 mL/min per 1.73 m2) in the first month, but it leveled off thereafter. (The reason for this is unclear and warrants further study, the authors note in their paper.)


Discontinuations were twice as common overall among the metformin patients compared with placebo (27% vs 12%, = .0002), as were discontinuations due to gastrointestinal adverse effects (16% vs 3%). Serious adverse event rates were similar (16% metformin vs 15% placebo, = .8418).


There were no differences in major or minor hypoglycemia. Biochemical vitamin B12 deficiency was more common with metformin (12% vs 5%, = .0094), but there were no cases of lactic acidosis. There were five and two deaths in the metformin and placebo groups, respectively, but none were deemed treatment related.


What About Newer Adjunctive Agents for Type 1 Patients? 

Dr Kar said he's more encouraged by data for the adjunctive use of more potent glucose-lowering agents in type 1 diabetes patients, such as the sodium glucose cotransporter-2 (SGLT2) inhibitors, noting that "the drops in A1c are quite significant [with these agents]."


However, Dr Petrie told Medscape Medical News, "We don't know the safety of SGLT2 inhibitors. There are phase 3 trials going on [in type 1 diabetes], but there's obviously the concern about ketoacidosis."


Nonetheless, he added, "We think adjunctive therapy [in type 1 diabetes] is a field to be explored. It may be worth investigating use of [glucagonlike peptide-1 receptor agonists] as well."


He also pointed out that this is the first-ever large trial designed to investigate an intervention to try to modify cardiovascular outcomes in type 1 diabetes.


"Our big concern is people don't understand there are differences between type 1 and type 2, particularly when it comes to cardiovascular prevention, although they both carry risk."


Bottom line, Dr Petrie said, "This is the best evidence we're going to have for some time about metformin in type 1 diabetes. Some patients can't tolerate it, but about seven out of eight can. If you can tolerate, it will help your weight control, LDL, and may help prevent atherosclerosis. And it's cheap. We're not saying give it to everyone."   


REMOVAL was funded by JDRF. 


Lancet Diabetes Endocrinol. Published online June 11, 2017. AbstractEditorial


www red DiabetologNytt