The discovery of metformin began with the synthesis of galegine-like compounds derived from Gallega officinalis, a plant traditionally employed in Europe as a drug for diabetes treatment for centuries.
In 1950, Stern et al discovered the clinical usefulness of metformin while working in Paris. They observed that the dose–response of metformin was related to its glucose lowering capacity and that metformin toxicity also displayed a wide security margin.
Metformin acts primarily at the liver by reducing glucose output and, secondarily, by augmenting glucose uptake in the peripheral tissues, chiefly muscle. These effects are mediated by the activation of an upstream kinase, liver kinase B1, which in turn regulates the downstream kinase adenosine monophosphatase protein kinase (AMPK).
AMPK phosphorylates a transcriptional co-activator, transducer of regulated CREB protein 2 (TORC2), resulting in its inactivation, which consequently downregulates transcriptional events that promote synthesis of gluconeogenic enzymes. Inhibition of mitochondrial respiration has also been proposed to contribute to the reduction of gluconeogenesis since it reduces the energy supply required for this process.
Metformin’s efficacy, security profile, benefic cardiovascular (CV) and metabolic effects, and its capacity to be associated with other antidiabetic agents makes this drug the first glucose lowering agent of choice when treating patients with type 2 diabetes mellitus (TDM2).
Metformin and pre-diabetes
In 2000, an estimated 171 million people in the world had diabetes, and the numbers are projected to double by 2030. Interventions to prevent T2DM diabetes, therefore, have an important role in future health policies. Developing countries are expected to shoulder the majority of the burden of diabetes. One of the main contributing factors to this burden is the Western lifestyle, which promotes obesity and sedentarism.
Impaired glucose tolerance (IGT) and impaired fasting glucose (IFG) statuses are associated with increased and varying risk of developing T2DM. IGT has been associated with an increased risk of CV events and may determine an increased mortality risk. The association of IFG with CV events, however, has not been well established.
When lifestyle interventions fail or are not feasible, pharmacological therapy may be an important resource to prevent T2DM. Several different drug classes have been studied for this purpose.
The best evidence for a potential role for metformin in the prevention of T2DM comes from The Diabetes Prevention Program trial. Lifestyle intervention and metformin reduced diabetes incidence by 58% and 31%, respectively, when compared with placebo.
These data suggest that, at least in the short-term, metformin may help delay the onset of diabetes. The benefits of metformin were primarily observed in patients <60 years old and in patients with a body mass index (BMI) greater than 35 kg/m2.
The prevalence of pre-diabetes as well as the progression rate to diabetes may differ between different populations, making the application of results from certain studies of different ethnical groups inappropriate. IGT is highly prevalent in Asian and Indian population groups.
Indians have several unique features such as a young age of diabetes onset and lower BMI along with high rates of insulin resistance and lower thresholds for diabetic risk factors.
In a meta-analysis of randomised controlled trials, Salpeter et al reported a reduction of 40% in the incidence of new-onset diabetes with an absolute risk reduction of 6% during a mean trial duration of 1.8 years.
Lily and Godwin reported a decreased rate of conversion from pre-diabetes to diabetes in individuals with IGT or IFG in their systematic review and meta-analysis of randomized controlled trials. This effect was seen at both a higher metformin dosage (850 mg twice daily) and lower metformin dosage (250 mg twice or three times daily) in people of varied ethnicity.
Metformin in the management of adult diabetic patients
Current guidelines from the American Diabetes Association/European Association for the Study of Diabetes (ADA/EASD) and the American Association of Clinical Endocrinologists/American College of Endocrinology (AACE/ACE) recommend early initiation of metformin as a first-line drug for monotherapy and combination therapy for patients with T2DM. This recommendation is based primarily on metformin’s glucose-lowering effects, relatively low cost, and generally low level of side effects, including the absence of weight gain.
Metformin’s first-line position was strengthened by the United Kingdom Prospective Diabetes Study (UKPDS) observation that the metformin-treated group had risk reductions of 32% for any diabetes-related endpoint, 42% for diabetes-related death and 36% for all-cause mortality compared with the control group.
The UKPDS demonstrated that metformin is as effective as sulfonylurea in controlling blood glucose levels of obese patients with T2DM. Metformin has been also been shown to be effective in normal weight patients.
Metformin in combination therapy
Although monotherapy with an oral hypoglycemic agent is often initially effective, glycaemic control deteriorates in most patients, which requires the addition of a second agent.
Currently, marketed oral therapies are associated with high secondary failure rates. Combinations of metformin and insulin secretagogue can reduce HbA1c between 1.5% to 2.2% in patients sub-optimally controlled by diet and exercise.
The optimal second-line drug when metformin monotherapy fails is not clear. All noninsulin antidiabetic drugs when added to maximal metformin therapy are associated with similar HbA1c reduction but with varying degrees of weight gain and hypoglycaemia risk.
A meta-analysis of 27 randomised trials showed that thiazolidinediones, sulfonylureas, and glinides were associated with weight gain, glucagon-like peptide-1 analogs, glucosidase inhibitors and dipeptidyl peptidase-4 inhibitors were associated with weight loss or no weight change.
Sulfonylureas and glinides were associated with higher rates of hypoglycaemia than with placebo. When combined with metformin, sulfonylureas and alpha-glucosidase inhibitors show a similar efficacy on HbA1c.
Metformin and sulfonylureas
The combination of metformin and sulfonylurea (SU) is one of the most commonly used and can attain a greater reduction in HbA1c (0.8–1.5%) than either drug alone.
The glimepiride/metformin combination results in a lower HbA1c concentration and fewer hypoglycaemic events when compared to the glibenclamide/metformin combination.
The use of metformin was associated with reduced all-cause mortality and reduced CV mortality. Metformin and SU combination therapy was also associated with reduced all-cause mortality.
Epidemiological investigations suggest that patients on SUs have a higher CV disease event rate than those on metformin. Patients who started SUs first and added metformin also had higher rates of CV disease events compared with those who started metformin first and added SUs. These investigations are potentially affected by unmeasured confounding variables.
Metformin and insulin
Metformin as added to insulin-based regimens has been shown to improve glycaemic control, limit changes in body weight, reduce hypoglycemia incidence, and to reduce insulin requirements (sparing effect), allowing a 15%-25% reduction in total insulin dosage.
The addition of metformin to insulin therapy in T1DM is also associated with reductions in insulin-dose requirement and HbA1c levels.
Metformin and thiazolinediones
The addition of rosiglitazone to metformin in a 24-week randomised, double-blind, parallel-group study significantly decreased HbA1c concentration and improved insulin sensitivity and HOMA ß cell function.
However, in spite of preventing diabetes incidence, the natural course of declining insulin resistance may not be modified by a low dose of the metformin-rosiglitazone combination.
The A Diabetes Outcome Progression Trial or ADOPT study assessed the efficacy of rosiglitazone, as compared to metformin or glibenclamide, in maintaining long-term glycaemic control in patients with recently diagnosed T2DM.
Rosiglitazone was associated with more weight gain, oedema, and greater durability of glycaemic control, metformin was associated with a higher incidence of gastrointestinal events and glibenclamide with a higher risk of hypoglycaemia.
Metformin and α glicosidase inhibitor
Acarbose reduces the bioavailability of metformin. However, it has been reported that the association of acarbose to metformin in sub-optimally controlled patients reduced HbA1c by about 0.8-1.0%.
Metformin and incretin-based therapies
Dipeptidyl Peptidase-4 prolongs the duration of active glucagon-like peptide 1 (GLP-1) by inhibiting DPP-4 peptidase, an enzyme, which cleaves the active form of the peptide. This action results in an improvement of insulin secretion as a physiological response to feeding.
The mechanism of DPP-4 inhibitors is complementary to that of metformin, which improves insulin sensitivity and reduces hepatic glucose production, making this combination very useful for achieving adequate glycaemic control.
Metformin has also been found to increase plasma GLP-1 levels, probably by either direct inhibition of DPP-4 or by increased secretion, leading to reduced food intake and weight loss.
Saxagliptin added to metformin led to clinically and statistically significant reductions in HbA1c from baseline versus metformin/placebo in a 24-week randomised, double-blind, placebo-controlled trial.
Saxagliptin at doses of 2.5mg, 5mg, and 10 mg plus metformin decreased A1 by 0.59%, 0.69%, and 0.58%, respectively, in comparison to an increase in the metformin plus placebo group for all comparisons.
A meta-analysis of 21 studies examined incretin-based therapy as an add-on to metformin in patients with T2DM for 16 – 30 weeks. Seven studies used a short-acting GLP-1 receptor agonist (exenatide BID), seven used longer acting GLP-1 receptor agonists (liraglutide or exenatide LAR), and 14 examined DPP-4 inhibitors.
Long-acting GLP-1 receptor agonists reduced HbA1c and fasting glucose levels to a greater extent than the other therapies.
Effects of metformin on vascular protection
Diabetic patients are at high risk of CV events, particularly of coronary heart disease by about three-fold. It has been stated that T2DM patients without a previous history of myocardial infarction (MI) have the same risk of coronary artery disease (CAD) as non-diabetic subjects with a history of MI.
This has led the National Cholesterol Education Program to consider diabetes as a coronary heart disease risk equivalent. Although there is no doubt that there is an increased risk of CAD events in diabetic patients, there is still some uncertainty as to whether the CV risk conferred by diabetes is truly equivalent to that of a previous MI.
In 1980, Scambato et al reported that, in a three-year observational study of 310 patients with ischaemic cardiomyopathy, patients treated with metformin had reduced rates of re-infarction, occurrence of angina pectoris, acute coronary events other than acute MI, and death in patients.
The largest effect was seen in re-infarction rates, a post hoc analysis showed that this effect was significant. After this study, the UKPDS, the largest randomised clinical trial in the newly diagnosed T2DM population largely free of prior major vascular events, randomly assigned treatment with metformin to a subgroup of overweight individuals.
In 1990, another subgroup of patients (n = 537), who were receiving the maximum allowed dosage of SU, were randomised either to continue sulfonylurea therapy or to allow an early addition of metformin.
Metformin provided greater protection against the development of macrovascular complications than would be expected from its effects upon glycaemic control alone. It had statistically significant reductions in the risk of all-cause mortality, diabetes-related mortality, and any endpoint related to diabetes, but not in MI.
The UKPDS post-trial reported significant and persistent risk reductions for any diabetes-related end point (21%), MI (33%), and death from any cause (27%).
Following UKPDS, other studies have reported significant improvement of all-cause mortality and CV mortality. A retrospective analysis of patients’ databases in Saskatchewan, Canada reported significant reductions for all-cause mortality and CV mortality of 40% and 36%, respectively. The PRESTO trial showed significant reductions of any clinical event (28%), MI (69%), and all-cause mortality (61%). The HOME trial reported a decreased risk of developing macrovascular disease.
In non-diabetic subjects with normal coronary arteriography but also with two consecutive positive (ST depression > 1mm) exercise tolerance test, an eight week period on metformin improved maximal ST-segment depression, Duke score, and chest pain incidence compared with placebo.
A recent meta-analysis suggested that the CV effects of metformin could be smaller than had been hypothesised on the basis of the UKPDS, however, its results must be interpreted with caution given the low number of randomised controlled trials included.
Effects on body weight
Metformin may have a neutral effect on body weight of patients with T2DM when compared to diet or may limit or decrease the weight gain experienced with SUs, TDZ, insulin, highly active antiretroviral therapy and antipsychotics drugs.
Modest weight loss with metformin has been observed in subjects with IGT. However, a meta-analysis of overweight and obese non-diabetic subjects, found no significant weight loss as either a primary or as secondary outcome.
The mechanisms by which metformin contributes to weight loss may be explained through the reduction in gastrointestinal absorption of carbohydrates and insulin resistance, reduction of leptin and ghrelin levels after glucose overload, and by induction of a lipolitic and anoretic effect by acting on GLP-1.
Reference: Rojas LBA, Gomes MB, Metformin: an old but still the best treatment for type 2 diabetes, Diabetology & Metabolic Syndrome, 2013.