In the past 2 decades, metformin has become a mainstay of type 2 diabetes management and is now the recommended first-line drug for treating the disease in the United States and worldwide.
Available in the United States since 1995, metformin is an attractive therapy for clinicians and patients alike. Studies have found the agent to be safe and effective, and at about $4 for a 1-month supply of the generic, that option is affordable at a time when many prescription drugs are being priced out of reach for some patients.
“Metformin is the first drug of choice, by all standards,” Oluwaranti Akiyode, PharmD, RPh, BCPS, CDE, professor and clinical pharmacist at Howard University School of Pharmacy, told Endocrine Today. “It’s a rarity that all experts agree on something. It is time-tested, proven, has good efficacy, a good safety profile and it’s cheap. Metformin has been around long before it came to the United States. That’s why I find it amazing that we only have one drug in that class.”
New research is suggesting that metformin may hold promise in treating or preventing a whole host of conditions in patients with and without type 2 diabetes. Studies show metformin may be cardioprotective in patients with diabetes and beneficial in the presence of stable congestive heart failure. The agent also may help to increase pregnancy rate in polycystic ovary syndrome, provide breast and prostate cancer benefits, and offer neuroprotection that may reduce dementia and stroke risk, Akiyode said.
Nir Barzilai, MD, an endocrinologist and director of the Institute for Aging Research at the Albert Einstein College of Medicine, said he hopes to work with the FDA to conduct an NIH/American Federation for Aging Research metformin trial later this year — Targeting Aging with Metformin (TAME) — that will demonstrate the agent’s ability to delay the onset of comorbidities related to aging, thereby reducing the period of morbidity at the end of life.
“If metformin can target and delay aging, its administration should be associated with fewer age-related diseases in general, rather than merely the decreased incidence of a single disease,” Barzilai and colleagues wrote in a study published in the June 2016 issue of Cell Metabolism. “Data from several randomized clinical trials and multiple observational studies provide evidence for such an effect, which would not be expected from glucose lowering alone.”
Endocrine Today spoke with experts who discussed the latest research demonstrating the nondiabetic benefits of metformin and the theories behind possible mechanisms.
Metformin’s history, mechanism
The use of what would eventually become metformin dates to the Middle Ages, when herbalists first derived extracts from Galega officinais, a plant known alternatively as French lilac, false indigo and Spanish sainfoin, to treat frequent urination. The plant’s active ingredient — guanidine — was first isolated in the 1800s and later synthesized as the less-toxic metformin. The drug was approved in France and the United Kingdom in 1957 and 1958, respectively, but it would be another 37 years before the FDA would approve the agent for use in the United States.
Despite the initial resistance to the drug, the FDA has been more open recently to considering studies that could examine other indications for metformin, through trials like TAME.
“The FDA has shown increasing flexibility in recent years with regard to metformin, surprisingly so, considering its original position back in the 1980s when it vowed never to approve the drug,” Alan J. Garber, MD, PhD, FACE, chief medical editor of Endocrine Today, said in an interview. “It was a surprisingly firm position, not necessarily grounded in observational science, but was related to the rather startling toxicity of [the drug’s] cousin, phenformin.” Phenformin was withdrawn from most markets in the late 1970s because of a high risk for lactic acidosis in patients.
Studies show that metformin decreases hepatic glucose production, decreases intestinal absorption of glucose and improves insulin sensitivity by increasing peripheral glucose uptake and utilization. Unlike sulfonylureas, metformin does not cause hypoglycemia or hyperinsulinemia; insulin secretion remains unchanged while fasting insulin levels and day-long plasma insulin response may decrease.
Metformin is most commonly used to treat type 2 diabetes, either alone or combined with other agents, but is also used off-label as a treatment for prediabetes, gestational diabetes and PCOS.
“There are some other off-label uses for metformin that we don’t have strong evidence for, but where there may be potential, such as HIV-related lipodystrophy,” Akiyode said. “For patients on atypical antipsychotic medications, they may experience some side effects related to insulin resistance. For that reason, those patients may consider using metformin. It may also have a role in metabolic syndrome, and it’s been studied intensely in nonalcoholic fatty liver disease, although pioglitazone may have a stronger role there.”
Exactly how metformin works to target these other conditions remains unclear, Akiyode said, and not every patient will respond to the drug in the same way.
Decades after its approval, Garber said, the mechanism remains poorly understood.
“No one knows how metformin works,” he said. “There, you have a Nobel Prize finding, if someone comes up with it.”
“Metformin, like any medicine, results in a spectrum of response,” Garber said. “A quarter of the patient population does extremely well. Half the population does rather well, and a quarter of the patient population doesn’t do well at all.”
An obesity option
Sriram Machineni, MD, of Massachusetts General Hospital in Boston, recalled a recent patient with obesity who struggled for years to lose weight, including attempting lifestyle modification and multiple weight-loss medications. An antipsychotic medication he took further contributed to his weight gain, Machineni said.
Ultimately, a combination of metformin and lorcaserin (Belviq, Eisai) worked for the patient, who has since maintained a 50-lb weight loss.
These results, Machineni said, were patient-specific as metformin is effective in counteracting weight gain from antipsychotic agents. Still, weight-loss results with metformin are sometimes a welcome surprise.
“What is interesting, is when I use metformin in a group of 10 patients, about seven of them either lose no weight or lose marginal amounts of weight,” he told Endocrine Today. “But the three that do respond, respond very dramatically. They’ll say, ‘The day I started taking metformin, I felt different.’ It may have something to do with central effects of metformin when it is able to get across the blood–brain barrier via the organic cation transporter (OCT2).
“We don’t really know how or why it works in these people,” Machineni said. “We have to start looking for markers to better identify the responders. These people respond dramatically, and they respond as they would to any other weight-loss drug on the market.”
Metformin is not indicated for weight loss. In the national Diabetes Prevention Program study conducted between 1996 and 1999, participants with prediabetes who were assigned metformin therapy lost some weight (mean, 2.1 kg), but less than those assigned to lifestyle intervention (mean, 5.6 kg) and far short of the 5% mean weight loss typically required for a drug to be indicated for weight loss.
“In my experience, with the patient population that I see, they don’t lose weight on it,” Akiyode said. “But the literature supports that metformin is weight neutral. Some people may experience a little weight loss, but it’s not enough to tout it as an obesity drug.”
Although the group of responders is disproportionately small, Machineni said, using metformin as an initial, first-line choice for obesity is worthwhile, either alone or combined with other agents, in part due to the low cost and safety of the drug.
“I’m upfront with patients and I tell them the numbers,” Machineni said. “Sometimes they say, ‘I don’t want to go through this, I want something more potent right up front.’ So, we have that discussion and make the best decision together, but metformin turns out to be the first drug to use in a lot of people, depending on what other medical conditions they have.”
Other emerging evidence suggests that metformin may preserve cognitive function and reduce mortality, Barzilai said.
In an analysis of 365 adults aged at least 55 years participating in the Singapore Longitudinal Aging Studies, Tze Pin Ng, MD, MBBS, of the gerontological research program at the National University of Singapore, and colleagues assessed the association of metformin use vs. nonuse with cognitive impairment (determined by Mini-Mental State Examination score) and stratified by duration of use ( 6 years, n = 114; > 6 years, n = 90). The results, published in the January 2014 issue of Journal of Alzheimer’s Disease, showed that metformin use was inversely associated with cognitive impairment in longitudinal analysis (OR = 0.49; 95% CI, 0.25-0.95) and that the lowest risk for cognitive impairment was associated with use of the drug for more than 6 years in cross-sectional analysis (OR = 0.3; 95% CI, 0.11-0.8) and longitudinal analysis (OR = 0.27; 95% CI, 0.12-0.6).
In a November 2014 observational study published in Diabetes, Obesity and Metabolism, Christian A. Bannister, MSc, of the Cochrane Institute of Primary Care and Public Health at Cardiff University, studied the progression to all-cause mortality among 78,241 patients with type 2 diabetes treated with metformin, 12,222 treated with sulfonylureas and 90,463 controls without diabetes, using data from the U.K. Clinical Practice Research Datalink.
Researchers found that patients treated with metformin monotherapy had longer survival vs. matched, nondiabetic controls (unadjusted mortality rates of 14.4 and 15.2, respectively, per 1,000 person-years), whereas those treated with sulfonylureas had reduced survival vs. matched controls (unadjusted mortality rates of 50.9 and 28.7 deaths per 1,000 person-years, respectively).
“This supports the position of metformin as first-line therapy and implies that metformin may confer benefit in nondiabetes,” the researchers wrote.
Barzilai said he hopes to explore similar findings in TAME. The researchers, supported by the American Federation for Aging Research, plan to recruit 3,000 adults aged 65 to 79 years in 14 centers across the United States, measuring the time to a new occurrence of a composite outcome that includes CV events, cancer, dementia and mortality over 5 years.
“The real risk for diabetes is aging,” Barzilai said. “Aging is a thousand-fold risk, while obesity is an eightfold risk. There is no comparison. From my point of view, if you have any improvement in one disease, you’re just exchanging one disease for another, unless you target aging. Aging is a target for diabetes just as much as specific drugs are.
“I consider metformin a tool,” he said. “Metformin will be a first-line choice, and then there will be other drugs that, in combination with metformin, will get better and better, just as diabetic drugs got better and hypertension and cholesterol drugs got better. You need pharmaceuticals to jump in [with anti aging therapies], and they’re not in the game now.”
Epidemiologic studies suggest that metformin use may be associated with both reduced cancer incidence and mortality. In a September 2012 study published in the European Journal of Endocrinology, Chin-Hsiao Tseng, MD, PhD, of National Taiwan University College of Medicine, Taipei, and colleagues analyzed data from 493,704 men and 502,139 women without colon cancer covered by national health insurance from 2003 to 2005, assessing whether metformin use had a protective effect. Although patients with diabetes had a higher risk for developing colon cancer, patients assigned metformin had a 27% reduced risk, according to researchers. Comparing patients with diabetes for less than 1 year, between 1 and 2 years, and at least 3 years with patients without diabetes, adjusted RRs were 1.308 (95% CI, 1.02-1.679), 1.087 (95% CI, 0.9-1.313) and 1.185 (95% CI, 1.055-1.33), respectively. The duration of metformin use showed an inverse trend, according to researchers, with an RR of 0.643 (95% CI, 0.49-0.845) in users for at least 3 years vs. nonusers.
“Metformin clearly affects AMP-activated protein kinase, which effects mitochondrial energy generation and may deprive malignant, inefficient cells of energy and therefore reduce their potential growth rate,” Garber said. “Although some of the observations [with metformin] imply not just [preventing] a progression of cancer but an incidence of cancer, which might suggest broader effects than merely energy deprivation.”
In commentary on a phase 3 randomized controlled trial assessing metformin’s role as a protective agent, published in March 2016 in The Lancet Oncology, Andrew T. Chan, MD, MPH, associate professor of medicine at Harvard Medical School and program director of the gastroenterology training program at Massachusetts General Hospital, noted that the mechanistic rationale for an anticancer effect of metformin is compelling.
“There’s been a lot of interest in metformin for cancer prevention largely because there has been evidence in human populations that individuals who take metformin have a lower risk of developing cancer,” Chan told Endocrine Today. “In parallel with that data, there are experimental findings that support a direct effect of metformin on specific pathways that mediate or promote cancer. It’s been shown in both animal models and cell lines that metformin can inhibit some of the key steps that are important for tumors to grow, such as promotion of cell growth, motility, invasion and migration. There’s been an interest because these two streams of data have come together to support the cancer-prevention potential of metformin.”
A few mechanisms have been hypothesized to explain what may be happening, he said. The most prominent theory is the idea that metformin activates a specific protein kinase, which down regulates the mTOR signaling pathway, which is important for tumor cell metabolism.
“More broadly, there is some evidence that insulin itself — and insulin-related pathways — may promote cancers,” Chan said. “So, there is the thought that metformin does modulate insulin and sensitizes cells to insulin, and that may also have something to do with its effects. I don’t think there is a definite mechanism that has been established, but there are a few potential mechanisms that may be plausible.”
Repurposing commonly used drugs, like metformin, for cancer prevention is attractive for a few reasons, he said.
“One is we have a wealth of experience in the use of these drugs,” Chan said. “We know a lot about what they do and their side effect profile. From the standpoint of clinical practice, people are more familiar with these agents, and are more comfortable with using them. So, we’ve been looking at different possibilities based on the evidence out there, and focused on medications that seem to be associated with a low risk of cancer, like aspirin and metformin.
“We have to think about metformin as having two possible roles,” he said. “One is in prevention of cancer, and the other is in treatment of cancer. There could be overlapping mechanisms or there could be distinct mechanisms. But there is a need for studies looking at both scenarios because there is a promising role for metformin in both.” – by Regina Schaffer
- Bannister CA, et al. Diabetes Obes Metab. 2014;doi:10.1111/dom.12354.
- Barzilai N, et al. Cell Metab. 2016;doi:10.1016/j.cmet.2016.05.011.
- Chan AT. Lancet Oncol. 2016;doi:10.1016/S1470-2045(16)00006-1.
- Ng TP, et al. J Alzheimers Dis. 2014;doi:10.3233/JAD-131901.
- Tseng CH. Eur J Endocrinol. 2012;doi:10.1530/EJE-12-0369.
- For more information:
- Oluwaranti Akiyode, PharmD, RPh, BCPS, CDE, can be reached at Howard University College of Pharmacy, 2300 Fourth St., NW, Washington, D.C. 20059; email: email@example.com.
- Nir Barzilai, MD, can be reached at Albert Einstein College of Medicine, 1300 Morris Park Ave., Belfer Building, #701, Bronx, NY 10461; email: firstname.lastname@example.org.
- Andrew T. Chan, MD, MPH, can be reached at Massachusetts General Hospital, Gastrointestinal Unit, GRJ-722, 55 Fruit St., Boston, MA 02114; email: email@example.com.
- Alan J. Garber, MD, PhD, FACE, can be reached at Baylor Clinic, 6620 Main St., Suite 1100, Houston, TX 77030; email: firstname.lastname@example.org.
- Sriram Machineni, MD, can be reached at Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114; email: email@example.com.
Disclosures: Garber reports consulting for Intarcia and Novo Nordisk. Machineni reports serving as an education consultant for Ethicon, a division of Johnson & Johnson. Akiyode, Barzilai and Chan report no relevant financial disclosures.
Click here to read the , "Should metformin be the first-line therapy choice in type 2 diabetes treatment?"