Metformin

Metformin (1,1-dimethylbiguanide) is an oral drug used mainly to treat type 2 diabetes by lowering blood glucose. It is a synthetic biguanide, a molecule built by joining two guanidine units, and it has become the most widely prescribed glucose-lowering agent and a standard first-line choice in many clinical guidelines. Its origins trace back to the medieval herbal use of Galega officinalis (French lilac), whose active guanidine constituents led to the synthetic compounds from which metformin was selected for clinical development in the 1950s.

Metformin matters for three reasons. It has a long clinical record, it is inexpensive and widely available, and it appears on the WHO model list of essential medicines. Its mechanism is also a live research question: the drug acts largely on energy metabolism in the liver and other tissues, centred on mitochondrial function and activation of the energy sensor AMPK, yet the complete molecular picture is still debated. Interest in its possible uses beyond diabetes, including cancer, cardiovascular disease and ageing, has grown, but these uses remain under study.

This entry first covers the history of the drug and its chemical and pharmacokinetic properties, then the mechanisms of action in the liver and peripheral tissues, along with the unresolved debates about them. It then describes clinical use in type 2 diabetes, safety and adverse effects, the state of research into other applications, and the drug's manufacturing and global use.

Chemistry and physicochemical properties

Structure and its relation to guanidine

Metformin is N,N-dimethylbiguanide, with the molecular formula C4H11N5 and a systematic name built around the two dimethylated amidine-type nitrogens. The "bi" in biguanide reflects the joining of two guanide units. The name was given by Rathke in 1879, after he obtained a new compound from a condensation reaction involving guanidine-related precursors. The core of the molecule is therefore a guanidine derivative. One review describes the guanidine structure as a nitrogenous analog of carbonic acid, and notes that metformin carries two methyl groups where its relatives phenformin and buformin carry aromatic rings or alkyl chains. The compound's lineage traces back to the plant alkaloid galegine (isoamylene guanidine), isolated from French lilac or goat's rue (Galega officinalis). Chemically, metformin is synthesized from dimethylamine hydrochloride and 2-cyanoguanidine. Crystallographic work in 2005 led to a reassignment of the preferred tautomeric representation of biguanides, which had been inconsistently drawn in the literature (a tautomer).

Basicity and charge at physiological pH

Reported acid dissociation constants for metformin differ between sources. One review lists values of 2.8 and 11.5 and attributes basic behaviour to the higher figure, which makes metformin a strong base that is highly soluble in water. Another source gives a single pKa of 12.4, which places the compound in cationic form across the pH range of 5 to 9 . Either way, the molecule is protonated under physiological conditions. Its charge is +1 at neutral pH in blood and cells, and its calculated logP of about −1.45 indicates high hydrophilicity and poor membrane permeability. Because the ionised form predominates, passive diffusion across lipid bilayers is limited, so cellular entry relies on carrier proteins (discussed in the transport section). The same positive charge explains why cationic compounds tend to concentrate in mitochondria, whose inner membrane maintains a strongly negative potential; this is a general principle of cation distribution rather than a measured figure specific to metformin.

Salt form, solubility and tablet formulation

The drug is usually marketed in the form of its hydrochloride salt. The salt is a white, hygroscopic crystalline powder with a bitter taste, and it is very highly water-soluble, above 300 mg/mL at 25° C. That solubility, together with a low-permeability profile in the lower gut, means that absorption occurs almost exclusively in the upper part of the gastrointestinal tract, and bioavailability falls within a range that decreases as the dose rises. The short plasma half-life, reported as 1.5–4.5 hours, is why conventional immediate-release tablets must be taken repeatedly . Formulators face the opposite problem when designing extended-release tablets: high solubility makes it difficult to provide a slow release rate and control the initial burst of drug, and high unit doses compound the problem. Poor compressibility adds further difficulty. Extended-release products generally use hydrophilic matrices such as HPMC, aiming to maintain plasma levels for 8–12 hours. Because food affects absorption, there is also a concern that extended-release tablets could release their contents too quickly, a phenomenon called dose dumping.

Comparison with phenformin and buformin

Phenformin was introduced in the U.S. in 1957 to treat non-insulin-dependent diabetes mellitus, and buformin was developed alongside it. The three biguanides differ mainly in their side groups, and this structural difference is linked to their toxicity profiles. Phenformin was withdrawn on November 15, 1978 following FDA proceedings, and buformin was withdrawn in many countries. The lactic acidosis risk differed sharply: one comparison estimated one in 4,000 patients taking phenformin, compared with one in 40,000 to 80,000 in patients taking metformin or buformin. Among 330 reported cases of biguanide-induced lactic acidosis in one literature review, phenformin accounted for the large majority and metformin for a small minority . Metformin survived for several reasons: its risk was lower, but it was mostly rejected for a long period before research in the 1980s and early 1990s demonstrated a uniqueness and utility of metformin that fostered its rescue. It was introduced to the U.S. market in 1995. Lactic acidosis has not been eliminated as a concern, since metformin can also cause lactic acidosis, although less frequently than phenformin.

Discovery and historical development

Metformin's ancestry runs through a plant. Goat's rue (Galega officinalis), also called French lilac, appears in European herbal practice, where extracts were given for thirst and frequent urination, symptoms later recognised as features of diabetes. The herb was described in the 1700s as a traditional treatment for thirst and frequent urination. Accounts differ on how old this use is. One review dates the herb's application to 1772 (Galega officinalis used to treat symptoms of diabetes (Hill)), while another places folk use of extracts in the seventeenth century (Southern and Eastern European folk medicine knowledge dating back until the 17th Century). The plant was also valued for raising milk output in cattle.

Chemical analysis identified guanidine as a major constituent of the herb. Guanidine was found to reduce blood glucose when injected into rabbits in 1918. Guanidine itself could not be used clinically because of its toxicity, so attention turned to related compounds. The alkaloid galegine was obtained by the French pharmacist Georges Tanret in 1914, and in the 1920s the synthalins, which contain two guanidine units, were also tried. Their therapeutic effects could not be separated from their toxic effects, and both families were largely abandoned. By the 1930s, they had been eclipsed by insulin, which had been discovered in 1922 and was starting to be produced on a larger scale.

Chemists then turned to biguanides, compounds formed by joining two guanidine units. Two Dublin chemists, Emil Werner and James Bell, prepared N,N-dimethylbiguanide in 1922 (Metformin itself was first prepared in 1922 by the Dublin chemists Emil A. Werner and James Bell.), although another account describes it as a side product from 1921. Its hypoglycaemic properties were demonstrated by two German teams in 1929. Within the series, metformin had been shown to lower blood glucose in rabbits and dogs and was the least toxic of the methyl biguanides then tested, which is why it was later chosen over its relatives. In the 1940s it was reexamined during a search for antimalarials, and in clinical tests it appeared useful against influenza when it sometimes lowered blood glucose (metformin was rediscovered in the search for antimalarial agents in the 1940s).

The clinical turning point came in France. In 1956 Aron Laboratories encouraged the physicians Jean Sterne and Denise Duval to study guanidine-based glucose-lowering agents (Jan Aron encourages Jean Sterne and Denise Duval to study guanidine-based glucose-lowering agents). Sterne had been prompted by a report on flumamine, a related compound used against influenza. In 1957, Sterne carried out the first clinical trials on metformin as an oral antidiabetic agent in humans. He judged that, of the biguanides he tested, metformin offered the best balance of benefit and risk. He also coined the name Glucophage, meaning "glucose eater" (He coined the name "Glucophage" (glucose eater) for the drug and published his results in 1957.). Sources differ on the launch date: one states that metformin was sold for the first time in France by Aron Laboratories in 1959, under the evocative brand name Glucophage.

Adoption outside France was gradual. Phenformin and buformin, the more potent biguanides, were withdrawn in most markets in the late 1970s because of lactic acidosis, and its association with the more toxic phenformin and buformin tarnished the reputation of the entire biguanide class. In the United States, the FDA granted approval in December 1994 (metformin was approved by US FDA in December 1994.), while the brand Glucophage reached the market in 1995. Its label carried a boxed warning about lactic acidosis. The long-term evidence base was reinforced by the UK Prospective Diabetes Study. Published in 1998, its results, which included cardiovascular benefit, confirmed metformin's cardiovascular benefits, cementing its status as a first-line therapy. Overweight patients in the trial who received metformin also had fewer diabetes-related endpoints and lower all-cause mortality than those on diet alone.

Pharmacokinetics and drug transport

Metformin is given by mouth, but its fate in the body depends mainly on membrane transporters rather than on simple diffusion across membranes. Its fractional oral bioavailability averages about 55%, with considerable spread between people. The mean ± SD fractional oral bioavailability (F) of metformin is 55 ± 16%. The figure also falls as the dose rises: across doses of 250 to 2000 mg, bioavailability has been reported to drop from 86% to 42%. Absorption in the gut is incomplete, and the degree of absorption differs from person to person. Estimates of how the absorbed fraction crosses the enterocyte differ between models. One widely cited schematic suggests that about half of an oral dose enters by passive diffusion, while the remainder is carried by PMAT and OCT1 in the intestinal epithelium.

The parent molecule is not broken down in the body. Because hepatic enzymes do not transform it, the drug is recovered essentially as the original compound, and urine is its main exit route. Its apparent elimination half-life is reported at roughly five to seven hours, depending on the study.

Hepatic uptake through OCT1

The liver is the main site of metformin's glucose-lowering action, and entry into hepatocytes depends largely on organic cation transporter 1 (OCT1), which sits on the sinusoidal (basolateral) membrane. Experiments in knockout mice illustrate its importance. In primary hepatocytes from Oct1-deficient mice, metformin uptake was 3.4-fold lower than in cells with a functional Oct1 allele. The same reduction in liver drug levels was accompanied by weaker activation of AMPK. One estimate attributes about 76% of hepatic transport to OCT1, although OCT3 is also expressed in hepatocytes and contributes to uptake. The affinity of OCT1 for metformin is moderate, with a reported Km of about 1.47 mM.

Because OCT1 transports in both directions and the drug is not metabolised, the concentration inside hepatocytes is expected to approach the concentration in plasma. Efflux from liver cells is handled in part by MATE1, which is expressed in hepatocytes and contributes to excretion of the unchanged drug into bile or back into the blood. Reported liver concentrations are, however, several-fold higher than systemic plasma levels, because the drug arrives first through the portal circulation.

Renal clearance and transporter control

Renal elimination is the principal route of clearance. OCT2 is located on the basolateral membrane of proximal tubule cells and mediates entry of the drug from blood into the tubular epithelium. MATE1 and MATE2-K then move metformin across the luminal membrane into urine. Metformin is considered a better substrate for OCT2 than for OCT1, which is why renal OCT2 is thought to dominate its disposition. Renal clearance in healthy adults has been reported at about 507 ± 129 mL/min, well above the glomerular filtration rate, which points to active tubular secretion. Tubular secretion is thought to account for roughly three-quarters to four-fifths of total renal clearance.

Because clearance depends on these transporters, drugs that inhibit them can change metformin levels. Cimetidine reduces renal clearance mainly by inhibiting MATE proteins rather than OCT2 alone, and pyrimethamine, which selectively inhibits MATEs, also lowers it. When kidney function declines, the drug accumulates in plasma; this is why reduced renal function is a recognised risk factor for lactic acidosis, a point taken up in the safety section.

Intestinal accumulation and portal delivery

The gut wall holds a large share of the drug. In mice given a single oral dose, the small intestine showed the highest tissue accumulation of any organ. Intestinal concentrations can reach about 1.3 mM, while portal vein levels are around 0.075 mM, and the intestinal mucosa may contain concentrations up to 300 times those in the circulation. In people, a 1 g oral dose gives plasma levels of roughly 20–30 μM, and portal vein levels are estimated to be about threefold higher, which implies hepatic exposure of roughly 60–90 μM. Whether this steep gut-to-portal gradient contributes to the antidiabetic effect remains unclear, and the gut's role in action is discussed further in the mechanism section. Variants in the OCT1 gene that reduce transporter function are associated with higher plasma levels of the drug, a point that connects pharmacokinetics to inter-individual variability.

Molecular mechanism in the liver

Complex I inhibition and cellular energy charge

The best-supported starting point is the mitochondrial respiratory chain. Metformin mildly and selectively inhibits respiratory-chain complex I in hepatocytes, and this lowers the cell's mitochondrial output of ATP (its primary site of action is the mitochondria via its mild and specific inhibition of the respiratory-chain complex 1, thereby lowering energy charge). Because ATP is consumed faster than it is replaced, the cell's AMP-to-ATP and ADP-to-ATP ratios rise (AMPK is activated by increases in AMP:ATP and ADP:ATP ratios, which indicate that cellular energy balance is compromised). The mechanism is nonetheless contested on dose grounds: complex I inhibition leading to AMPK activation has been observed mainly at supra-pharmacological concentrations above 1 mM, which are not reached in clinical use.

AMPK activation and downstream lipid effects

The rise in AMP relative to ATP is the signal that switches on AMP-activated protein kinase (AMPK), a sensor of cellular energy status. Once active, AMPK pushes the cell toward ATP-generating catabolism while restraining ATP-consuming processes. In the liver, this includes the phosphorylation of acetyl-CoA carboxylase, which slows fatty acid synthesis. In one mouse study, animals lacking the AMPK-dependent phosphorylation of this enzyme no longer showed metformin's lipid-lowering and insulin-sensitising effects, which suggests that some of the drug's metabolic benefit runs through this route (metformin-induced reduction in blood glucose levels depends on its ability to lower cellular fatty acid levels through the AMPK-dependent phosphorylation of ACC). AMPK's role in the glucose-lowering effect itself is disputed, since the metabolic effect of the drug is preserved in liver-specific AMPK-deficient mice.

Suppression of gluconeogenesis through cyclic AMP

Hepatic gluconeogenesis is driven largely by glucagon. Glucagon raises cyclic AMP (cAMP) in hepatocytes, which activates protein kinase A and the transcription factor CREB, and this in turn increases the expression of the gluconeogenic genes encoding PEPCK and glucose-6-phosphatase (this process upregulates the expression of rate-limiting gluconeogenic genes including Pck1 and G6pc, which then enhances hepatic gluconeogenesis). A study by Miller and colleagues found that biguanides raise intracellular AMP and related nucleotides, which in turn blunts the ability of glucagon to activate adenylyl cyclase. This results in reduced levels of cyclic AMP and protein kinase A activity, which blocks phosphorylation of substrates that are important for maintaining glucose output from hepatocytes. In this model the fall in cAMP, rather than AMPK activation, is the key step, and it is the direct action of AMP on adenylyl cyclase that links the energy change to the signalling change. Lower cAMP then reduces the expression of the gluconeogenic enzymes themselves (lowered PEPCK and G6Pase protein expression in treated mice). Other proposals, such as direct inhibition of fructose-1,6-bisphosphatase, have also been put forward (metformin reduces liver glucose production by inhibition of fructose-1-6-bisphosphatase), so the step-by-step chain from complex I to enzyme expression is not settled.

Reduction in hepatic glucose output

The net result of these hepatic actions is a fall in hepatic glucose output, which is widely regarded as the main contributor to the drug's glucose-lowering effect. Metformin exerts its antihyperglycemic action primarily through lowering hepatic glucose production. The clearest early human evidence came from a study reported in 1995 by Stumvoll et al., which demonstrated that in individuals with T2DM, metformin reduced hepatic glucose output by up to 75%, primarily through suppression of gluconeogenesis. Smaller direct measurements have given more modest figures; one study of nine adults with type 2 diabetes found an average reduction of 24% after three weeks of metformin treatment, resulting in an average lowering of the fasting plasma glucose level by 30%. The relative size of the hepatic contribution is also being reassessed, as some authors argue that the glucose-lowering effect of metformin in T2DM is mediated not only by suppression of hepatic glucose production, but also, and perhaps predominantly, through actions on the stomach and intestine.

Actions in muscle, fat, gut and other tissues

Skeletal muscle

In healthy people, most insulin-stimulated glucose disposal happens in skeletal muscle, and uptake there depends on GLUT4 moving from intracellular storage vesicles to the plasma membrane. Glucose disposal in fat tissue is relatively small compared with muscle. After insulin stimulation, GLUT4 is shifted from storage vesicles to the plasma membrane through an exocytic pathway.

Metformin appears to act on this same machinery. In cultured C2C12 muscle cells, metformin increased Rab4 expression in an AMPK-dependent manner, and it stimulated phosphorylation of AS160 and PKC-zeta. When Rab4 was knocked down, insulin-induced GLUT4 translocation was suppressed. These findings come from cell culture, so they show a plausible route rather than a demonstrated mechanism in patients. The debate over how much AMPK contributes is taken up in a later section.

Adipose tissue

The evidence for fat cells is thinner. In human adipocytes, metformin has been reported to stimulate glucose uptake through AMPK-dependent GLUT4 translocation, without affecting insulin signaling. Metformin is also described as improving insulin sensitivity, but the effects on circulating free fatty acids and on lipolysis are less consistent across studies, and the human data I reviewed do not settle the direction of these changes. Any claim about adipose lipid handling should therefore be treated as provisional.

Gut

The intestine is metabolically active, and the intestines are the second biggest organ that uses glucose, following the brain. In rodents, metformin raises the level of SGLT1, the main glucose transporter of the upper small intestine. Several lines of evidence indicate that metformin not only increases GLP-1 secretion, but also upregulates SGLT-1. Metformin also increases glucose uptake by intestinal tissue, which removes glucose from the blood. Whether the net effect is reduced absorption or increased local use remains disputed, and this is a point on which studies differ.

Metformin also increases secretion of GLP-1 from intestinal L cells. Metformin can increase the secretion of glucagon-like peptide-1 by L cells, improving blood glucose homeostasis and reducing lipid secretion by intestinal epithelial cells. The gut microbiota is a further link. In rodents on a high-fat diet, metformin treatment in the upper small intestine counteracts the microbiota shift and restores glucose sensing. In another study, metformin reshaped the gut microbiota to enrich beneficial taxa, such as A. muciniphila, L. reuteri and L. johnsonii, and suppressed potentially harmful species, with heterogeneous dose–response relationships. Transplant experiments are needed to show that microbial change, rather than an associated effect, drives the glucose response.

GDF15 and energy balance

Metformin also raises circulating GDF15, a hormone linked to appetite suppression. In two randomised trials, metformin increased GDF15 levels, and GDF15 has been shown to reduce food intake and lower body weight through a brain-stem-restricted receptor. In mice, oral metformin increased circulating GDF15, with GDF15 expression increasing predominantly in the distal intestine and the kidney. A rat study reported that kidney-specific knockdown of GDF15 expression, as well as AP-specific knockdown of GFRAL expression, negated the ability of metformin to lower food intake and body weight gain. The picture is not uniform, however. A later study found that metformin lowers body weight independently of GDF15 and GFRAL in obese mice, and that in humans metformin raises plasma GDF15 without a clear association with weight loss.

Unresolved questions and scientific debates

For decades the standard account has been that metformin accumulates in mitochondria, partly inhibits complex I of the respiratory chain, lowers ATP output, and thereby activates AMPK, which in turn restrains gluconeogenesis in the liver. Several links in that chain have since been questioned, and researchers still describe the mechanism as unresolved. One review notes that the field lacks agreement on the key molecular target even after two decades of study.

Effects that AMPK activation does not explain

AMPK activation is among the most frequently cited explanations, but it does not account for everything the drug does. In mice lacking AMPK in the liver, metformin still lowered blood glucose and still suppressed hepatic glucose production, which points to a route that bypasses the kinase. A 2022 study proposed a different entry point: metformin binds the protein PEN2, which then associates with ATP6AP1 and inhibits the lysosomal v-ATPase. This recruits AXIN and LKB1 to the lysosomal surface, where they help activate AMPK, without needing the adenine-nucleotide signal. That work examined low drug doses specifically. Other investigators have argued that AMPK activation in the gut, rather than in the liver, is what matters most for blood glucose. Metabolite-level studies add another layer. Under conditions of glucose or gluconeogenic substrate excess, metformin lowers hexose phosphates in ways that do not depend on AMPK, probably through allosteric activation of phosphofructokinase-1 or inhibition of fructose bisphosphatase-1.

Complex I or another target?

Advocates of the complex I model still regard it as the most coherent explanation. A 2025 commentary in Science Advances argued that the evidence convincingly shows metformin lowers glucose by inhibiting complex I, although the same article acknowledges that the role of the liver and of hepatic AMPK has been challenged. A rival model emerged in 2014, when a study reported that metformin non-competitively inhibits mitochondrial glycerophosphate dehydrogenase (mGPD), a redox-shuttle enzyme. On this view, the drug raises the cytosolic NADH/NAD⁺ ratio, which reduces the conversion of lactate and glycerol into glucose. A 2020 preprint re-examined this claim using four enzyme assays in insulin-secreting and liver cells and could not demonstrate direct inhibition. Its authors also noted that blocking mGPD in beta cells would be expected to impair insulin secretion, which would sit awkwardly with the drug's clinical benefit. Subsequent work has blurred the picture further. A 2022 study in PNAS proposed that guanides and biguanides act on complex IV, which indirectly reduces GPD2 activity. A Journal of Biological Chemistry study found that low concentrations of metformin oxidised the hepatocyte NADH/NAD redox state yet still inhibited gluconeogenesis, which argues for a redox-independent mechanism. The debate is therefore between overlapping, partly incompatible models rather than between a settled answer and a single rival.

The problem of concentration

A central source of disagreement is dose. A pharmacokinetic review estimated that doses of 250 mg/kg or more expose the liver to concentrations above 1 mM, and it argued that complex I inhibition and related effects occur mainly at supra-pharmacological levels that are not reached in clinical use. Cell culture experiments commonly use millimolar concentrations, so findings from them may not transfer to patients. Defenders of the complex I model respond that low-dose changes in adenine nucleotides may be real but lie near the limit of assay detection. Inconsistent dosing across studies is itself a reason why results diverge, and this makes it hard to assign a single molecular effect to a particular clinical outcome.

Why responses differ between individuals

Responses to metformin vary considerably between people, and transporter genes are one candidate explanation. Because the drug is a cation at physiological pH, its movement into cells depends on transporters rather than simple diffusion. OCT1, encoded by SLC22A1 and expressed mainly in the liver, carries the drug into hepatocytes. In knockout mice lacking Oct1, the glucose-lowering effect of metformin was abolished. Variants such as R61C, Met420del and G465R reduce transport function, and some have been associated with poorer glycaemic response or with gastrointestinal intolerance. In a 2022 cohort of 81 patients taking 500 mg twice daily, steady-state drug levels varied widely, and the authors examined SLC22A1 and SLC47A1 (MATE1) variants as contributors. Evidence is inconsistent, however. One study found that response in responders and non-responders was independent of OCT1 isoform variation and mRNA expression, while an SLC47A1 variant has been linked to differences in HbA1c reduction. Other transporter variants, including those in SLC22A3 and SLC47A2, showed no significant association in one cohort. Ethnicity and the underlying disease are likely to confound these associations, so the practical use of transporter genotyping in prescribing remains under investigation.

Clinical use in type 2 diabetes

In the United Kingdom, metformin has been treated as the preferred first-line drug for type 2 diabetes since the National Institute for Health and Care Excellence (NICE) guidance of 2015, a position that rests largely on the UK Prospective Diabetes Study (UKPDS). In practice it is usually the starting point of treatment, and it is also the most common base for combination regimens, to which other insulin or non-insulin agents are added when glycaemic control is insufficient.

Glycaemic control and dosing

Metformin lowers blood glucose without the weight gain associated with some older agents. Most studies show modest weight loss of 2 to 3 kg during the first six months of treatment, and weight gain does not occur in patients with type 2 diabetes who receive it.

Doses in adults are typically between 500 and 3,000 mg per day, and the drug is generally taken before meals, with the dose raised slowly from once daily, by about 500 mg per day each week, to reduce gastrointestinal effects. Common regimens are 500 mg three times daily or 850 mg twice daily. For the prolonged-release formulation, the maximum daily dose is 2,000 mg, a ceiling that reflects its better tolerability relative to the immediate-release tablet.

Prediabetes

Metformin has also been studied for delaying the onset of type 2 diabetes in people with prediabetes. In the US Diabetes Prevention Program (DPP), the metformin group developed diabetes at a rate of 7.8 cases per 100 subject-years, compared with 11.0 in the placebo group, corresponding to a relative risk reduction of about 31%. Intensive lifestyle intervention performed better in the same trial, with a 58% reduction. After a washout period, the protective effect fell, and analyses suggest that only a portion of the early benefit was attributable to the drug's immediate pharmacological action, while a persistent reduction remained. In the DPP Outcomes Study, the reduction in risk was still present at 10, 15 and 22 years of follow-up, though it diminished over time.

Benefit was not uniform. Metformin was most effective in younger, heavier participants, and a greater effect was seen in women with prior gestational diabetes. Based on these data, the American Diabetes Association advises considering metformin for people under 60 with a body mass index above 35 or a history of gestational diabetes. Use in prediabetes remains an off-label use in many settings, and improvements in long-term clinical outcomes in people with prediabetes have yet to be demonstrated.

Long-term outcome evidence

The UKPDS 34 study, published in 1998, randomised 1,704 patients who weighed more than 120 percent of their ideal body weight, of whom 342 were assigned to metformin. Compared with conventionally treated patients, metformin recipients had risk reductions of 32% for any diabetes-related endpoint, 42% for diabetes-related death, and 36% for all-cause mortality. Among patients on intensive glucose control, metformin also showed a greater effect than chlorpropamide, glibenclamide or insulin on several endpoints, including all-cause mortality and stroke. The benefit on cardiovascular events contrasted with the findings for sulfonylureas and insulin, and the authors suggested that the absence of weight gain might partly explain it.

These findings have limits. No significant reduction in microvascular complications was observed with intensive metformin therapy in obese participants, and crossovers between treatment groups added uncertainty. The UKPDS also reported a signal in non-overweight patients: those taking metformin plus a sulfonylurea had a higher risk of death than those on sulfonylurea alone, a result the trial researchers considered non-significant and which was largely overlooked. Later reanalyses have debated how far the overall mortality benefit should be generalised, and the cardiovascular evidence from subsequent trials is discussed elsewhere in this entry.

Special populations

Metformin's use in pregnancy has shifted over time. Evidence on its safety in gestation has led to it being recognised as a valuable tool in preventing insulin resistance-related pregnancy complications and in managing gestational diabetes. Yet the findings remain mixed. In a placebo-controlled trial of 257 pregnant women with polycystic ovary syndrome (PCOS), metformin did not reduce pregnancy complications such as gestational diabetes, pre-eclampsia or preterm delivery, whereas other prospective studies reported reductions in gestational diabetes. Current reviews indicate that continuing metformin into pregnancy does not increase the risk of miscarriage or preterm delivery, and some prescribing advice still calls for stopping the drug once pregnancy is confirmed.

In PCOS, metformin is widely used off-label, and it has been included as a management option in leading guidelines for PCOS. Its principal role is improving insulin resistance and ovulation. Meta-analyses of trials in women with PCOS found that metformin may increase the pregnancy rate and reduce the risk of ovarian hyperstimulation syndrome in women undergoing one type of assisted reproduction. Use in children and adolescents with type 2 diabetes is supported by a smaller evidence base than in adults, and treatment in younger patients is generally guided by specialist services.

Adverse effects, contraindications and safety monitoring

Gastrointestinal symptoms are the most common reason people on metformin report side effects. Nausea, diarrhoea, abdominal discomfort, bloating, flatulence and occasionally vomiting typically appear early in treatment, particularly after starting or increasing the dose. A 2022 meta-analysis of randomised trials in type 2 diabetes found that metformin was associated with more abdominal pain, diarrhoea and nausea than placebo or comparator drugs, and that bloating and diarrhoea were more frequent with immediate-release tablets than with extended-release ones. Clinicians usually manage these symptoms by starting at a low dose and titrating upward slowly, by taking the drug with meals, or by switching to an extended-release formulation. The extended-release product is designed to release the drug over a longer period, which lowers the peak concentration in the gut; this reduces, but does not abolish, GI intolerance. Some individual differences have been reported, including associations with body mass index and with variation in transporter genes, but these factors do not yet guide routine prescribing.

Vitamin B12 deficiency is the principal long-term metabolic side effect. Metformin appears to impair the absorption of vitamin B12 from the terminal ileum, and the risk rises with duration and dose. In a trial cited in clinical reviews, mean vitamin B12 levels fell by about 19% over several years, and the number needed to harm for developing deficiency was roughly 14 patients treated for four years. In 2022 the UK Medicines and Healthcare products Regulatory Agency (MHRA) advised clinicians to check B12 in patients with symptoms of deficiency and to consider periodic monitoring in those at higher risk, including older adults, people with gastric or small-bowel surgery, Crohn's disease, and autoimmune conditions. The American Diabetes Association recommends periodic testing, particularly after about four years of therapy or when risk factors are present. Co-prescription with proton pump inhibitors has been associated with a stronger signal. No universally agreed testing interval exists, and surveys suggest that uptake of monitoring in primary care is uneven. Because neurological damage can become permanent, anaemia (often with raised mean corpuscular volume) or peripheral neuropathy should prompt immediate B12 measurement. Deficiency is usually corrected with supplementation, with intramuscular injections used in severe cases.

Lactic acidosis is the most serious recognised risk, but it is rare. Estimates of its incidence in the trial and observational literature range from about 3 to 10 cases per 100,000 person-years. Its signs are often nonspecific, such as malaise, myalgia, drowsiness, breathlessness and vague abdominal distress. Lactate accumulates when metformin is cleared more slowly than usual, and the drug is excreted largely unchanged by the kidney, so reduced renal function is the main predisposing factor. Labelling also identifies sepsis, dehydration, hypoxic states, acute heart failure, hepatic impairment and excessive alcohol intake as risks, because alcohol potentiates the drug's effect on lactate metabolism. Drugs that impair renal function, cause haemodynamic change, disturb acid-base balance or inhibit renal transporters can raise metformin levels. Metformin is contraindicated in metabolic acidosis, including diabetic ketoacidosis. In US labelling revised in 2016, metformin is contraindicated when the estimated glomerular filtration rate (eGFR) is below 30 mL/min/1.73 m²; starting it is not recommended between 30 and 45; and, when eGFR falls below 45 in a patient already taking it, the benefits and risks must be reassessed. Renal function is to be checked before starting and at least annually, more often in older adults. An observational analysis published in 2018 suggested that the excess acidosis risk was confined to eGFR values below 30, which informed the revision, though the figures remain debated.

Metformin is also temporarily stopped around iodinated contrast procedures given intravascularly, during acute illness with a risk of hypoperfusion or hypoxia, and during surgery or hospitalisation. The usual approach is to hold the drug on the day of the procedure and restart only after renal function has been reassessed and found stable. Labelling also advises against use in patients with mitochondrial disorders such as MELAS and MIDD, because of a theoretical risk of lactate accumulation. Dosing is adjusted to kidney function, so eGFR is the key variable that determines whether the drug can be continued, reduced or must be withdrawn.

Research into other applications

Interest in metformin beyond glucose lowering rests mainly on observational data, and the results have been contested. A 2013 meta-analysis of 37 studies, covering about 1.5 million participants, found that metformin users had a lower summary risk of overall cancer incidence and cancer mortality than non-users. The authors concluded that metformin could reduce incidence of overall, liver, pancreatic, colorectal and breast cancer, as well as mortality from overall, liver and breast cancer. No significant association with prostate cancer incidence was found. Larger pooled analyses have produced similar headline figures, but they have also drawn criticism. One analysis of 47 studies in people with diabetes estimated a 31% lower cancer incidence, yet between-study heterogeneity was considerable. Commentary in Diabetes Care argues that many of the studies included in such meta-analyses have been shown to be affected by various time-related biases that tend to exaggerate the benefit of a drug. A 2012 meta-analysis reported that the difference in risk ratios between observational studies and randomised trials was statistically significant, which is consistent with that concern. Site-specific findings are also inconsistent: a meta-analysis of thyroid cancer found non-statistically significant negative associations, with limited evidence quality.

Cardiovascular research has followed a similar pattern. Observational studies and small trials suggested a possible cardioprotective effect beyond glycaemic control, particularly in heart failure. The large Met-HeFT trial, reported in 2026, tested this directly. Over a mean follow-up of 3.7 years, there was no significant difference between metformin and placebo for the primary endpoint of death, worsening heart failure, acute myocardial infarction or stroke (25.1% vs. 23.4%; hazard ratio 1.10). Secondary endpoints such as all-cause death, unplanned heart failure events, new-onset diabetes or changes in the heart strain marker NT-proBNP were also not significantly affected. The investigators concluded that metformin was not associated with any harm and still remains beneficial in terms of glucose lowering, but showed no significant evidence for any independent cardioprotective benefits. Studies of endothelial function and inflammatory markers have been similarly difficult to interpret, because they are often small, short, and rely on surrogate measures rather than clinical events.

The idea that metformin might slow ageing itself draws on animal and cellular work. Its effects on AMPK and mTOR signalling, autophagy and inflammation are highlighted as possible routes through which it could influence ageing biology. Human evidence is thinner. Short trials in healthy volunteers have reported beneficial changes in gene expression in various aging pathways, along with objective improvements in risk factors such as LDL cholesterol and triglycerides, but these trials cannot show effects on disease or death. The Targeting Aging with Metformin (TAME) concept addresses this gap. It is a double-blind randomized placebo-controlled clinical trial that tests whether the commonly used antidiabetic drug metformin is capable of delaying the onset of age-related diseases. The planned design enrols 3,000 nondiabetic individuals aged 65-79 for about six years. Its proposal was also notable because it would treat aging itself as the indication, and regulatory discussions with the FDA began in 2015. Reports on TAME's status conflict: one source states that as of 29 August 2026, there is no completed recruitment and no result, whereas another describes the trial as still enrolling in mid-2026. Both agree that no efficacy results have been published.

Laboratory studies have also explored antimicrobial, antiviral and anticancer effects. Metformin has been shown in several in vitro and in vivo studies to reduce cancer cell proliferation in several cancer cell types. Reports of activity against bacteria, mycobacteria and viruses have come mostly from cell culture and animal models. Concentrations used in such experiments are often higher than the drug levels reached in patients, and effects seen in a dish do not reliably predict benefit in people. Approved indications therefore remain limited to type 2 diabetes and related uses, and claims for cancer prevention, longevity or infection treatment should be read as hypotheses that await randomised clinical evidence.

Manufacturing, regulation and global use

Metformin is now sold almost entirely as generic product. Once the original patent protection ended, many manufacturers began making the hydrochloride salt, which pushed prices far below those of the original brand. Because of this low cost, it is among the cheapest diabetes medicines available in most markets. It is also the most widely prescribed drug for type 2 diabetes, and the immediate-release form is the type most commonly dispensed.

Global use is closely tied to where diabetes is growing. About 80% of the 828 million adults with diabetes live in low- and middle-income countries (LMICs), and WHO guidance recommends cost-effective glucose-lowering drugs, metformin among them, for settings with limited resources. In a survey of 32 national essential medicines lists in LMICs, every list included metformin. Access, however, remains uneven. In one multinational pharmacy survey, metformin was stocked by all pharmacies in high-income countries but by only about 65% of those in low-income countries (excluding India). Affordability showed a similar gap: roughly 27% of low-income households could not afford a monthly supply, compared with under 1% of high-income households. Quality testing of generic products in LMICs has also been carried out; a study of brands sold in Ethiopia found that all of them met WHO visual-inspection and USP specifications for weight variation, disintegration and assay.

Metformin was added to the WHO Model List of Essential Medicines in 1997, and it remains on the list as a 500 mg hydrochloride tablet. The list, first published in 1977, is designed to define the minimum medicines a basic health system should provide, and it is revised periodically. National regulators in major markets, including the US Food and Drug Administration, the European Medicines Agency and the UK's MHRA, license metformin and include it in national formularies, although the specific labelling and prescribing rules differ between jurisdictions.

Quality control became a major issue in 2020. In May of that year the FDA asked several manufacturers to voluntarily recall certain extended-release (ER) metformin tablets after testing found N-nitrosodimethylamine (NDMA), a nitrosamine impurity regarded as a probable carcinogen, above acceptable intake limits. Recalls by individual companies continued through October 2020, November 2020 and January 2021. A later analysis reported that the FDA recalled 281 ER metformin products between February 2020 and January 2022. Immediate-release products were not found to contain NDMA at levels of concern, and the FDA advised clinicians to keep prescribing metformin when clinically appropriate. It also warned patients not to stop taking their medicine without consulting a clinician, because abrupt discontinuation could itself be harmful. Nitrosamines can form during manufacturing or storage, which led regulators to add impurity testing across many drug classes.

Taken together, the combination of low price, long clinical experience and broad regulatory approval has made metformin a cornerstone of diabetes care in many health systems. Its availability and affordability, however, still depend on national procurement, supply-chain quality and household income.

References

1. Metformin Chemistrynews-medical.net
2. A Biguanide Class Treatment of Type 2 Diabetesmethodist.edu
3. Metformin derivativessciencedirect.com
4. Metformin Hydrochloride - an overviewsciencedirect.com
5. Chemical structure of metformin.researchgate.net
6. Extended release pharmaceutical formulations of water-soluble active pharmaceutical ingredients and methods for making the sameimage-ppubs.uspto.gov
7. Formulation and Evaluation of Extended-Release Solid Dispersion of Metformin Hydrochloride - PMCpmc.ncbi.nlm.nih.gov
8. (PDF) Formulation and evaluation of extended release tablets containing metformin HClresearchgate.net
9. Phenformin and lactic acidosis: a case report and review - ScienceDirectsciencedirect.com
10. Phenforminen.wikipedia.org
11. Buformin - an overviewsciencedirect.com
12. 50 Hong Kong Med J Vol 14 No 1 # February 2008 # www.hkmj.org Introductionhkmj.org
13. Metformin: historical overviewlink.springer.com
14. (PDF) Formulation of Extended-Release Metformin Hydrochloride Matrix Tabletsresearchgate.net
15. Evaluation of Food Effect on the Pharmacokinetics of Sustained Release Metformin in Healthy Indian Volunteersclinicaltrials.gov
16. History of Metformin: Origins, Development, and Uses Today - GoodRxgoodrx.com
17. Metformin: Therapeutic profile in the treatment of type 2 diabetes - Bailey - 2024 - Diabetes, Obesity and Metabolism - Wiley Online Librarydom-pubs.onlinelibrary.wiley.com
18. COMMENTARY Metformin: historical overview Clifford J. Bailey1link.springer.com
19. Galegine - an overviewsciencedirect.com
20. Metformin: History and mechanism of actionlgcstandards.com
21. New hopes for metformin - Towards a better understanding of its mechanisms of actionmedecinesciences.org
22. Metformin (Glucophage) - eMedExpert.comemedexpert.com
23. (PDF) Metformin: historical overviewresearchgate.net
24. Who Originally Made Metformin? The History and Key Figureswellzrx.com
25. CENTER FOR DRUG EVALUATION AND RESEARCH APPLICATION NUMBER: 212595Orig1s000accessdata.fda.gov
26. Metformin: its botanical background - Bailey - 2004 - Practical Diabetes International - Wiley Online Libraryonlinelibrary.wiley.com
27. Metformin History & Development: From French Lilac to First-Line Diabetes Drughealthrx.com
28. Metformin Approval History - Consensus Academic Search Engineconsensus.app
29. (PDF) Clinical Pharmacokinetics of Metforminresearchgate.net
30. JCI - Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin actionjci.org
31. Metformin and Systemic Metabolism - PMC - NIHpmc.ncbi.nlm.nih.gov
32. In Vivo Imaging of Human 11C-Metformin in Peripheral Organs: Dosimetry, Biodistribution, and Kinetic Analysesjnm.snmjournals.org
33. Quantitative contributions of hepatic and renal organic cation transporters to the clinical pharmacokinetic cimetidine-metformin interactionbiorxiv.org
34. Interindividual variability in metformin pharmacokinetics in pediatric patientsnature.com
35. Transporters Involved in Metformin Pharmacokinetics and Treatment Response - Journal of Pharmaceutical Sciencesjpharmsci.org
36. Metformin is a Superior Substrate for Renal Organic Cation Transporter OCT2 rather than Hepatic OCT1researchgate.net
37. Mini Review Transporters Involved in Metformin Pharmacokinetics and Treatmentjpharmsci.org
38. OCT2 - Transporters - Solvo Biotechnologysolvobiotech.com
39. A PET Tracer for Renal Organic Cation Transporters, 11C-Metformin: Radiosynthesis and Preclinical Proof-of-Concept Studiesjnm.snmjournals.org
40. Metformin in therapeutic applications in human diseases: its mechanism of action and clinical studylink.springer.com
41. Cellular and Molecular Mechanisms of Metformin Actionacademic.oup.com
42. The coordinated actions of metformin in the intestine and liver improves hyperglycemia - ScienceDirectsciencedirect.com
43. Accumulation of metformin by tissues of the normal and diabetic mouse - PubMedpubmed.ncbi.nlm.nih.gov
44. pmc.ncbi.nlm.nih.govpmc.ncbi.nlm.nih.gov
45. Metformin: From Mechanisms of Action to Therapies - ScienceDirectsciencedirect.com
46. The mechanisms of action of metforminlink.springer.com
47. Cellular and molecular mechanisms of metformin: an overviewportlandpress.com
48. Linderane Suppresses Hepatic Gluconeogenesis by Inhibiting the cAMP/PKA/CREB Pathway Through Indirect Activation of PDE 3 via ERK/STAT3ncbi.nlm.nih.gov
49. Biguanides suppress hepatic glucagon signalling by decreasing production of cyclic AMPnature.com
50. Metformin—mechanisms of its glycemia-reducing effect - ScienceDirectsciencedirect.com
51. Metformin reduces liver glucose production by inhibition of fructose-1-6-bisphosphatasenature.com
52. Reframing metformin as a gut-targeted glucose-lowering therapy: Mechanistic insights and translational relevance - ScienceDirectsciencedirect.com
53. Patient-specific effects of metformin on the hepatic metabolism in adolescents with metabolic dysfunction-associated steatotic liver disease (MASLD)link.springer.com
54. Molecular Mechanisms for the Regulation of Insulin-Stimulated Glucose Uptake by Small Guanosine Triphosphatases in Skeletal Muscle and Adipocytesncbi.nlm.nih.gov
55. Metformin induces Rab4 through AMPK and modulates GLUT4 translocation in skeletal muscle cells - PubMedpubmed.ncbi.nlm.nih.gov
56. (PDF) Metformin regulates glucose transporter 4 (GLUT4) translocation through AMP-activated protein kinase (AMPK)-Mediated Cbl/CAP Signaling in 3T3-L1 Preadipocyte cells.researchgate.net
57. Frontiersfrontiersin.org
58. The Gut Microbiota-Related Antihyperglycemic Effect of Metformin - PMCpmc.ncbi.nlm.nih.gov
59. REVIEW articlefrontiersin.org
60. Metformin Alters Upper Small Intestinal Microbiota that Impact a Glucose-SGLT1-Sensing Glucoregulatory Pathway: Cell Metabolismcell.com
61. Reframing metformin as a gut-targeted glucose-lowering therapy: Mechanistic insights and translational relevancencbi.nlm.nih.gov
62. GDF15 mediates the effects of metformin on body weight and energy balance - PubMedpubmed.ncbi.nlm.nih.gov
63. Metformin triggers a kidney GDF15-dependent area postrema axis to regulate food intake and body weight - ScienceDirectsciencedirect.com
64. The GDF15-GFRAL pathway is dispensable for the effects of metformin on energy balance - ScienceDirectsciencedirect.com
65. Understanding the action mechanisms of metformin in the gastrointestinal tractfrontiersin.org
66. Cellular and Molecular Mechanisms of Metformin Action - PMCpmc.ncbi.nlm.nih.gov
67. Meeting metformin again for the first timescience.org
68. Low-dose metformin targets the lysosomal AMPK pathway through PEN2nature.com
69. The Metformin Mechanism on Gluconeogenesis and AMPK Activation: The Metabolite Perspectivemdpi.com
70. Metformin suppresses gluconeogenesis by inhibiting mitochondrial glycerophosphate dehydrogenasenature.com
71. If Metformin Inhibited the Mitochondrial Glycerol Phosphate Dehydrogenase It Might Not Benefit Diabetesbiorxiv.org
72. Metformin, phenformin, and galegine inhibit complex IV activity and reduce glycerol-derived gluconeogenesispnas.org
73. Low metformin causes a more oxidized mitochondrial NADH/NAD redox state in hepatocytes and inhibits gluconeogenesis by a redox-independent mechanism - Journal of Biological Chemistryjbc.org
74. Editorial: Metformin: Beyond Diabetesncbi.nlm.nih.gov
75. SLC22A1 - an overviewsciencedirect.com
76. The influence of metformin transporter gene SLC22A1 and SLC47A1 variants on steady-state pharmacokinetics and glycemic response - PMCpmc.ncbi.nlm.nih.gov
77. Pharmacogenetic variation of SLC47A1 gene and metformin response in type2 diabetes patientsncbi.nlm.nih.gov
78. Role of human organic cation transporter-1 (OCT-1/SLC22A1) in modulating the response to metformin in patients with type 2 diabetesncbi.nlm.nih.gov
79. The Influence of OCT3 and MATE2 Genetic Polymorphisms in Poor Response to Metformin in Type 2 Diabetes Mellitusncbi.nlm.nih.gov
80. Studies that changed clinical practice. UKPDS: United Kingdom Prospective Diabetes Study - DiabetesontheNetdiabetesonthenet.com
81. Formulations for the prevention and treatment of insulin resistance and type 2 diabetes mellitusimage-ppubs.uspto.gov
82. Role of Metformin in Polycystic Ovary Syndrome (PCOS)-Related Infertility - PMCpmc.ncbi.nlm.nih.gov
83. Therapeutic Use of Metformin in Prediabetes and Diabetes Prevention - PMCpmc.ncbi.nlm.nih.gov
84. Full article: Metformin for diabetes prevention: update of the evidence basetandfonline.com
85. Metformin for diabetes prevention: update of the evidence base - PubMedpubmed.ncbi.nlm.nih.gov
86. Use of metformin following a population-level intervention to encourage people with pre-diabetes to enroll in the National Diabetes Prevention Program - PMCpmc.ncbi.nlm.nih.gov
87. Metformin Use in Overweight Patients with Type 2 Diabetesaafp.org
88. (PDF) Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34)researchgate.net
89. Implications of the United Kingdom Prospective Diabetes Studydiabetesjournals.org
90. Reappraisal of Metformin Efficacy in the Treatment of Type 2 Diabetes: A Meta-Analysis of Randomised Controlled Trials - PMCpmc.ncbi.nlm.nih.gov
91. The Role of Metformin in Metabolic Disturbances during Pregnancy: Polycystic Ovary Syndrome and Gestational Diabetes Mellitusncbi.nlm.nih.gov
92. Effects of metformin use in pregnant patients with polycystic ovary syndrome - PMCpmc.ncbi.nlm.nih.gov
93. Update on the therapeutic role of metformin in the management of polycystic ovary syndrome: Effects on pathophysiologic process and fertility outcomes - Kerstin MG Brand, Ulrike Gottwald-Hostalek, Aimee Andag-Silva, 2025journals.sagepub.com
94. The Effectiveness of Metformin in Diabetes Preventioncureus.com
95. The Future of Metformin for Prediabetesdiatribe.org
96. The “slower” the betterncbi.nlm.nih.gov
97. (PDF) Gastrointestinal adverse events of metformin treatment in patients with type 2 diabetes mellitus: A systematic review, meta-analysis and meta-regression of randomized controlled trialsresearchgate.net
98. Metformin use and vitamin B12 deficiency: New MHRA guidance - DiabetesontheNetdiabetesonthenet.com
99. Clinician awareness and implementation of vitamin B12 monitoring guidance in metformin users: a primary care surveyncbi.nlm.nih.gov
100. Vitamin B12 Deficiency in Patients Taking Metformin: Pathogenesis and Recommendationsncbi.nlm.nih.gov
101. What evidence is there to support monitoring a B12 level in patients who are taking metformin? How often should it be monitored?inpharmd.com
102. Vitamin B12 Deficiency Associated with Metformin and Proton Pump Inhibitors and Their Combinations: Results from a Disproportionality and Interaction Analysis - PMCpmc.ncbi.nlm.nih.gov
103. Metformin: Package Insert / Prescribing Information / MOAdrugs.com
104. Possible Side Effects of Metformin Use in Patients with Type 2 Diabetesconsultqd.clevelandclinic.org
105. Lactic Acidosis Risk in Metformin Patients With Reduced Kidney Functiondiabetesincontrol.com
106. Metformin Extended Release Versus Metformin Immediate Release in Subjects With Type 2 Diabetesclinicaltrials.gov
107. Metabolic Effects of Metformin Therapy in Obstructive Sleep Apneacdn.clinicaltrials.gov
108. Metformin as a Novel Therapy for Autosomal Dominant Polycystic Kidney Diseasecdn.clinicaltrials.gov
109. HIGHLIGHTS OF PRESCRIBING INFORMATIONaccessdata.fda.gov
110. Reference ID: 4346286 - accessdata.fda.govaccessdata.fda.gov
111. Association of metformin use with cancer incidence and mortality: a meta-analysis - PubMedpubmed.ncbi.nlm.nih.gov
112. Metformin and Cancer Risk and Mortality: A Systematic Review and Meta-analysis Taking into Account Biases and Confoundersaacrjournals.org
113. Metformin and Cancer: Solutions to a Real-World Evidence Failurediabetesjournals.org
114. Cancer Risk in Diabetic Patients Treated with Metformin: A Systematic Review and Meta-analysisjournals.plos.org
115. Metformin and thyroid carcinoma incidence and prognosis: A systematic review and meta-analysisncbi.nlm.nih.gov
116. No significant evidence for protection against cardiovascular outcomes with metformin in heart failuremedicalxpress.com
117. No significant evidence for protection against cardiovascular outcomes with metformin in heart failure: Studymedicaldialogues.in
118. The Anti-Aging Mechanism of Metformin: From Molecular Insights to Clinical Applicationsncbi.nlm.nih.gov
119. The TAME Trial: Targeting Aging with Metforminnutritionfacts.org
120. Targeting Aging with Metformin (TAME)link.springer.com
121. Increasing Healthspan: Prosper and Live Longncbi.nlm.nih.gov
122. Metformin and Longevity: Evidence, TAME and Risksprogevita.com
123. Metformin Shows No Cardiovascular Benefit in HFrEF: Met-HeFT Trial Findingshcplive.com
124. No significant evidence for protection against cardiovascular outcomes with metformin in heart failureescardio.org
125. TAME Trial 2026: Current Status, Funding Delays, and What Comes Next for Metformin Anti-Aging Researchstackingyears.com
126. Metformin for Anti-Aging & Longevity: TAME Trial, Dosage & Alternatives (2026)yourhealthier.com
127. FDA Recalls Extended-Release Metformin Due to NDMA Impuritiesmedscape.com
128. FDA announces a new recall for extended-release metformin due to contaminationcardiovascularbusiness.com
129. FDA requests voluntary recall of contaminated metformin productshealio.com
130. Extended-release metformin recalled due to impurityimmattersacp.org
131. FDA Updates and Press Announcements on NDMA in Metforminfda.gov
132. Metformin Recall Type 2 Diabetes Drug Recalled By the FDAhealthcentral.com
133. Comparative evaluation of quality attributes among six different brands of metformin hydrochloride tablets marketed in Bahir Dar city, Ethiopia - PMCpmc.ncbi.nlm.nih.gov
134. In vitro comparative quality evaluation of different brands of metformin hydrochloride tablets available in Mekelle City, Tigray Regional State, Ethiopia - PMCpmc.ncbi.nlm.nih.gov
135. s12249 023 02724 3link.springer.com
136. Availability and affordability of essential medicines for diabetes across high-income, middle-income, and low-income countries: a prospective epidemiological studyresearchgate.net
137. (PDF) Availability and affordability of essential medicines for diabetes across high-income, middle-income, and low-income countries: a prospective epidemiological study - Medicineacademia.edu
138. National evidence on glucose-lowering medication use for diabetes from 62 low- and middle-income countries - PMCpmc.ncbi.nlm.nih.gov
139. Selection of Essential Medicines for Diabetes in Low and Middle Income Countries: A Survey of 32 National Essential Medicines Listsjournals.plos.org
140. Metformin Essential medicine status General description INN Metformin ATC codeslist.essentialmeds.org
141. Diabetes and the WHO Model List of Essential Medicines - The Lancet Diabetes & Endocrinologythelancet.com
142. World Health Organization Model List of Essential Medicines 21st List 2019iccp-portal.org