The Anti-Aging Pathways of AMPK (Part 2)

Today, we look at the anti-aging properties of the diabetes drug, Metformin.

Might the diabetes drug, Metformin, have an anti-aging effect? Check out this three-video series to see what the science says.

The noxious weed goat’s rue was widely used as herbal medicine back in the Middle Ages for thirst and frequent urination, what we now know as a cardinal symptom and sign of diabetes. The plant produces a mitochondrial poison known as galegine. One can imagine how this could help discourage grazing herbivores, but why would crippling mitochondrial function help with diabetes? If you got the dose just right, you could impair energy production just enough to spring AMPK into action. AMPK would then dial down the liver’s energy-intensive process of sugar production, and blood sugars would fall.

Galegine proved too toxic for use in humans (and presumably goats), but was tweaked into a milder form in the 1950s. That’s the origin story of metformin––now, even 60+ years later, the most widely prescribed diabetes drug in the world. Sold originally as Glucophage (meaning “sugar eater”), metformin is now prescribed annually more than 90 million times in the U.S. alone. Despite all the strides in biotechnology, Big Pharma has yet to come up with a safer, more effective first-line treatment for type 2 diabetes than a drug that retails at pennies per pill.

Ramping up AMPK isn’t all that metformin does (for example, it also appears to foster the growth of good bacteria in the gut). But, the AMPK boost alone would be expected to have wide-ranging health effects. Remember, AMPK is activated when resources are scarce. So, it causes the body to hunker down into energy-conservation mode, and put the brakes on new construction projects. So, no more making excess blood sugar or cholesterol. Time to start tapping into the larder. And, indeed, metformin doesn’t just lower sugars in the blood, but cholesterol and body fat. And, shutting down the excess synthesis of proteins can also have health benefits.

Fibrosis is the accumulation of scar tissue in response to an inflammatory insult. As a transient, localized phenomenon, fibrosis helps heal wounds and repair damaged tissue. But in response to the systemic inflammation of aging, surplus scar tissue can build up in our heart, liver, kidneys, and lungs. Excess fibrosis is estimated to contribute to 45 percent of all deaths in the United States. But AMPK switches the body into miser mode, shrinking the energy budget for such superfluous activities and ratchets down fibrotic processes.

The AMPK priorities shift from growth to preservation might also be expected to slow cancer progression. Based on dozens of observational studies involving tens of thousands of diabetic cancer patients, those on metformin did indeed experience a significant survival advantage, correlating with a 26 percent decreased risk of dying from cancer. No wonder there are more than 100 ongoing studies to put metformin to the test against cancer. As the director of Harvard’s Cancer Center put it, “Metformin may have already saved more people from cancer deaths than any drug in history.”

If AMPK plays such a key role in the aging process, and metformin boosts AMPK, can metformin slow aging and lengthen lifespans? It certainly can in some species. C. elegans worms fed metformin maintain a youthful state and can exhibit a nearly 40 percent increase in average survival. Interfere with the AMPK pathway, though, and this lifespan extension disappears, confirming the role of AMPK in metformin’s anti-aging effects. Metformin also improves the healthspans and lifespans of mice but failed to extend the lives of rats, calling the pro-longevity effects of metformin into question.

The dose they used (15 times that used in humans) may have been too high, however, and the F344 rats they used are an inbred strain (named after the 344th brother–sister mating) that are resistant to the health benefits of calorie restriction. Since metformin acts through AMPK as a calorie restriction mimetic, it stands to reason that these animals may be less prone to metformin’s effects. What about humans?

Unlike other potential anti-aging drugs in the pipeline, we have the benefit of data on more than a million study subjects who have been tracked in metformin trials. And, indeed, compared to diabetics who take drugs other than metformin, those on metformin appear protected from age-related susceptibilities like fractures or serious infection, glaucoma, and cognitive impairment.

Most importantly, diabetics who take metformin live significantly longer than those taking different sugar-lowering drugs. Doesn’t that prove metformin is a pro-longevity drug? No. For example, maybe metformin just controlled their diabetes better. Researchers still found a survival benefit for diabetics on metformin compared to those as well-managed on other drugs, but maybe the other drugs were shortening lifespans, and so, metformin just looked good in comparison. How could you try to tease out the effects?

What if you compared the lifespans of diabetics on metformin to nondiabetics? Now, that’s not really a fair comparison. Any benefits metformin has could easily be outweighed and masked by the life-limiting nature of diabetes—unless, that is, the diabetics on metformin actually lived longer. Longer than nondiabetics. And, are you sitting down? That’s exactly what the research shows. Diabetics placed on metformin went on to live longer lives than those who never got diabetes in the first place. From a longevity standpoint, it’s as if they got lucky to be diagnosed with diabetes, because then, they had access to this lifespan-enhancing drug. So, wait. If metformin is so powerful as to more than offset such a dreaded diagnosis, should everyone be taking metformin? What are the downsides? That’s exactly what I’m going to be covering next.

The apparent reductions in mortality from all causes put together, and diseases of aging associated with metformin use, suggest that the drug metformin could be extending life and healthspans by acting as a geroprotective, or anti-aging, agent. What might the downsides be? Gastrointestinal side effects, such as flatulence, indigestion, and abdominal discomfort are relatively common. More than half starting metformin report diarrhea, for example, and a quarter experience nausea or vomiting. About 10 percent unknowingly taking the placebo report similar symptoms, though. So, the actual proportion of people suffering such symptoms due to the drug is probably more like one in three. Most of these effects tend to be transient, though, and can be minimized by taking metformin with food, and starting at a suboptimal dose, increasing gradually to target over a period of weeks. Once-a-day extended-release versions also tend to be more tolerable.

The most dreaded side effect, though, is lactic acidosis––the potentially fatal buildup of acid in the blood. I mean, poisoning of the mitochondria is kind of the whole point. That’s what activates AMPK––but it’s supposed to be a mild mitochondrial poison. But sometimes it evidently poisons a bit too much. Thankfully, metformin-induced lactic acidosis is extremely rare. It has a fatality rate of 50 percent, but only occurs at a rate of about 0.03 cases per 1,000 patient years. That means taking metformin for a year only has about a one in 67,000 chance of killing you.

Lactic acidosis is primary caused by insufficient metformin clearance by the kidneys. So, risk can be minimized by, for example, temporarily stopping the drug when undergoing procedures that put the kidneys at risk––such as major surgery, or intravenous contrast dyes used for certain imaging procedures. Those with mild to moderate kidney disease can still benefit from metformin at an adjusted dose. Elderly individuals should have their kidney function tested before starting metformin, and regularly monitored. Excess alcohol consumption is also believed to increase the risk of fatal metformin-induced lactic acidosis.

One of the more insidious side effects is B12 deficiency, affecting as many as one in three long-term metformin users, but can start as early as a few months. The mechanism is still unclear, but routine monitoring of B12 status has been recommended, and B12 supplementation has been suggested.

Another adverse consequence of metformin is less side effect than main effect. Remember how metformin is a mild mitochondrial poison? The way metformin boosts AMPK is by impairing our body’s ability to produce energy. What might that mean for athletic performance? Healthy men and women were found to suffer a significant decline in peak aerobic capacity after only about a week on metformin. The average hit was only 3 percent, though, which could be meaningful for an elite athlete, but would be unlikely to impede recreational physical activity. However, some individuals lost as much as 10 percent, which could be noticeably fatiguing. Note these detriments were only found at or near maximal workloads, and so, most active individuals would be unlikely to experience an impairment of routine activity. Over the long term, though, metformin may undercut physical fitness.

Older adults were randomized to 12 weeks of supervised aerobic exercise on or off metformin. They trained for 45 minutes three times a week on a treadmill, bike, or elliptical, ramping up to a vigorous target of 85 percent of their maximal heart rate. Unfortunately, metformin attenuated the improvement in physiological function after exercise training. Cardiovascular fitness is so important; yet, after three months, metformin cut their improvement in cardiorespiratory fitness in half compared to placebo––though this didn’t reach statistical significance. Metformin did, however, significantly blunt their improvement in whole-body insulin sensitivity, which normally improves with physical activity.

Aerobic exercise (or “cardio”) is critical for maintaining cardiovascular fitness, whereas resistance exercise (also known as strength training, for example, with weights or bands) is essential for maintaining muscle mass as we age. In 2019, the so-called MASTERS trial was published, investigating what metformin might do to muscle gains, and the title of the paper gives it away: “Metformin blunts muscle hypertrophy in response to progressive resistance exercise training in older adults.” After 14 weeks of progressive resistance training, men and women in the placebo group accrued significant lean body mass, but those randomized to the metformin group did not. For example, about a four percent gain in thigh muscle mass in the placebo group, versus virtually none in the metformin group. The differences in strength, however, did not reach statistical significance.

So, might the benefits of metformin outweigh the risks for expanding the healthspan and lifespan of even nondiabetics? The investigators reviewing the evidence that diabetics taking metformin live longer, on average, than nondiabetics, concluded their “results suggest that metformin could be an effective intervention to extend the lifespans of people who do not have diabetes.” You can’t know, though, until you put it to the test, which I’ll explore, next.

Given the evidence that diabetics taking the drug metformin may live longer than people who don’t even have diabetes suggests metformin has an anti-aging effect, slowing the diseases of aging and, in doing so, extending both health and lifespans. But you don’t know until you put it to the test. There are a few small, brief trials of metformin on healthy volunteers, for example, randomizing a dozen or so individuals to metformin or placebo for a few weeks, comparing the effects on muscle and fat biopsies. Beneficial changes in gene expression in various aging pathways were noted, along with objective improvements in risk factors, such as LDL cholesterol and triglycerides. But what we’d really like are hard endpoints, like disease outcomes and death. Enter the TAME trial, Targeting Aging with Metformin.

Spearheaded by the nonprofit American Federation for Aging Research, TAME has the ambitious goal of randomizing 3,000 nondiabetic individuals aged 65-79 to metformin or placebo over a period of six years to assess the effect on age-related conditions, such as heart attacks, stroke, cancer, dementia, and death. Current treatments “just exchange one disease for another,” railed the principal investigator. Someone cured of cancer may just drop dead the next year from a heart attack. “What we want to show is that if we delay aging, that’s the best way to delay disease.”

Sadly, though the TAME trial received FDA approval in 2015, it has yet to get off the ground. You can guess why. As the editor-in-chief of an anti-aging journal put it, “There is no way in hell to make money out of it.” Metformin’s patent expired decades ago, and generics retail at pennies a pill. It’s considered a “genuinely good use of 75 million dollars,” but the researchers have only evidently been able to raise $11 million from private donors.

Some in Silicon Valley tech circles aren’t waiting for the results. Apparently, some techies have started taking metformin in a bid to live longer, though it’s only officially available by prescription for diabetes. Metformin is cheap, relatively safe, and one of the most promising drugs to combat age-related diseases, but there are reasons to temper our expectations. Though it’s been referred to as a “magic drug,” the longevity benefits in mice were small and dose-dependent. At best, it may increase the average lifespan of certain mice by five percent, but at a higher dose actually shortens lifespan. Much of the enthusiasm for the TAME trial rests on it acting as a test case to successfully get regulatory authorities, like the FDA, to formally recognize any treatment for the aging process itself, in hopes this will arouse Big Pharma to invest large-scale R&D to advance the field.

Further reservations about the panacean prospects placed on metformin arise from the landmark Diabetes Prevention Program study, published in the New England Journal of Medicine. Thousands of prediabetic individuals were randomized to metformin or placebo for years to see if the drug could prevent their prediabetes from turning into full-blown diabetes. It worked, reducing overall diabetes incidence by 31 percent compared to placebo. But it didn’t for everyone. For example, it didn’t appear to significantly benefit those 60 years or older, and only seemed to help those most at risk—those with class II obesity and the highest blood sugars. Those with lower (but still elevated) blood sugars, and those under a BMI of 35 did not have their diabetes risk significantly dampened by the drug. One small study even found that, while metformin alleviated the insulin resistance of diabetics and those with a family history of diabetes, metformin actually worsened the insulin resistance (the cause of prediabetes and type 2 diabetes) of nondiabetic obese individuals that didn’t have a genetic predisposition. If you have diabetes, it can make things better; if you don’t, it may make things worse. So, healthier individuals may not reap the benefits of metformin that some try to extrapolate from those longevity studies on diabetics. In fact, even those researchers are explicit that they don’t think metformin is ready for prime time as an anti-aging intervention by the general population.

What’s safer and works even better? There was a third arm of the Diabetes Prevention Program study. Thousands of prediabetics were randomized to metformin or placebo or lifestyle changes—a cholesterol-lowering low-fat diet and exercise, at least about 20 minutes a day of moderate intensity exercise, such as brisk walking. Compared to the placebo group, those taking metformin were 31 percent less likely to develop diabetes. The lifestyle group was 58 percent less likely. Making simple lifestyle changes was nearly twice as effective as metformin (and with fewer side effects—for example, six times fewer cases of gastrointestinal distress).

Other randomized controlled trials have found the same thing—lifestyle approaches superior to drug-based approaches for diabetes prevention. And, that 60 or so percent drop in risk was not among those that actually made the lifestyle changes––just those instructed to do so. Another diabetes prevention trial published in the New England Journal of Medicine that found the same 58 percent reduction in diabetes risk among those asked to make lifestyle changes drilled down to see what the benefit was for those who fully complied. Only a fraction of the lifestyle intervention group met the modest goals. Only about a third increased their exercise, for example. Only about a quarter met the minimum recommended intake of fiber (found concentrated in whole plant foods), or cut down on enough saturated fat (mostly dairy, dessert, chicken, and pork). But, they did better than the control group. And fewer of them developed diabetes because of it. But, what if you looked just at the folks that actually made all (or even just four out of five) of the recommended lifestyle changes? They had zero diabetes—none of them got diabetes over the course of the study, meaning effectively a 100 percent drop in risk.