The "Fasting Pill": How a Compound From Apple Bark Became Longevity's Most Interesting Drug

## A Failed Malaria Cure From Apple Bark In 1835, French chemists boiled the bark of apple trees hoping to find a fever remedy to rival quinine. What they isolated instead was a bitter compound called **phlorizin**. It did nothing useful for malaria. Decades later, physiologists noticed what it did do: animals given phlorizin started passing large amounts of sugar in their urine. At the time, sugar in the urine was the defining sign of diabetes. The disease's full name, diabetes mellitus, comes from the Latin for "honey-sweet," because physicians once diagnosed it by tasting urine. So for roughly a century phlorizin had exactly one job in science: it was the standard way to make a lab animal look diabetic. Researchers even called the effect "phlorizin diabetes." There was one problem with that label. In real diabetes, blood sugar goes up. In phlorizin-treated animals, blood sugar went **down**. The sugar in the urine was being pulled out of the blood. That was the opposite of the disease, and it took until the 1990s to identify the reason. The modern descendants of that apple-bark compound are now some of the most prescribed drugs in cardiology and nephrology. In the last two years they have also become the most interesting candidates for something longevity researchers have wanted for decades: a pill that reproduces part of what fasting does. This post covers the mechanism, what the human evidence actually shows, where the hype outruns it, and what you can do with the same biology without a prescription. ## The Mechanism: Your Kidneys Recycle 180 Grams of Sugar a Day Your kidneys filter roughly 180 grams of glucose out of your blood every day. Almost none of it reaches your urine, because the kidney takes it all back. The protein doing most of that work is **SGLT2** (sodium-glucose cotransporter 2), which sits in the first segment of the kidney's proximal tubule and reclaims about 90% of filtered glucose. A related transporter, SGLT1, picks up most of the rest further down the tubule. From an evolutionary standpoint this makes sense. 180 grams of glucose is about 720 calories. For most of human history, leaking that much energy every day would have been fatal, so the kidney evolved to waste none of it. The system has no off switch for a world where food is always available. Phlorizin blocks both SGLT1 and SGLT2. Modern drugs in the same family, the **SGLT2 inhibitors**, block SGLT2 selectively. The result: - Roughly 60 to 80 grams of glucose leave in the urine each day, which is somewhere around 200 to 300 calories. - Nothing is blocked from being absorbed in the gut. The glucose is digested, enters the blood, and is then discarded by the kidney. - Blood glucose and insulin both drift lower, and glucagon rises. That last point is the important one for this story. Lower insulin and higher glucagon is the hormonal pattern of a fast. ### Why the math does not deliver 20 pounds On paper, 200 calories a day adds up to about 73,000 calories a year, or roughly 20 pounds of fat. In clinical trials, typical weight loss is closer to 2 to 3 kg (about 5 to 8 pounds). The body notices the missing energy and increases appetite to compensate. SGLT2 inhibitors are not weight-loss drugs in any serious sense, and nobody should expect them to be. ## The Result Nobody Could Explain If the story ended there, this would be a footnote about a diabetes drug with a mild weight side effect. It did not end there. In 2015 the **EMPA-REG OUTCOME** trial followed about 7,000 people with type 2 diabetes and established cardiovascular disease. Those taking empagliflozin had a **38% lower rate of cardiovascular death** than those on placebo. No glucose-lowering drug had shown anything like that. Two details made the result hard to explain: - **The benefit appeared within months.** Improvements in blood sugar take years to change cardiovascular outcomes. - **It did not track with glucose.** The protection was not explained by how much a patient's blood sugar improved. Later trials extended the finding to heart failure and chronic kidney disease, including in people who did not have diabetes at all. A drug designed to lower blood sugar was protecting organs through some route that had little to do with blood sugar. ## The Fasting Connection The leading explanation is that these drugs push the body into a state that resembles fasting. People taking SGLT2 inhibitors show a consistent, modest rise in blood **ketones**. Your liver makes ketones when insulin is low and it is burning fat for fuel, which normally means you have not eaten for a while. In this case, people were eating three meals a day and still producing them. In cell and animal studies, the same drug class moves the main nutrient-sensing pathways in the fasting direction: - **AMPK**, the cell's low-energy sensor, is activated. - **mTOR**, the growth-and-storage pathway, is suppressed. - **SIRT1**, a NAD+-dependent enzyme tied to stress resistance and repair, is upregulated. - **Autophagy**, the cell's recycling of damaged components, increases. If those four sound familiar, they are the same switches covered in our [autophagy explainer](/blog/autophagy-the-nobel-prize-winning-cellular-repair-system-and-why-modern-life-shuts-it-off). Compounds that trigger this pattern without actual food restriction have a name: **caloric restriction mimetics**. One caveat belongs right here. Most of the pathway evidence comes from cells and animals. The ketone rise is well documented in humans; the downstream AMPK, mTOR and autophagy changes are much harder to measure in living people and are largely inferred. ## What the Research Shows ### The telomere trial (2025) A randomized, double-blind, placebo-controlled trial published in *Cell Reports Medicine* gave the SGLT2 inhibitor **henagliflozin** or placebo to 142 adults with type 2 diabetes for 26 weeks. The primary endpoint was telomere length in white blood cells, a marker that shortens as cells divide and is often used as a rough proxy for cellular aging. Reported results: - About **90%** of the drug group showed an increase in telomere length, compared with about **66%** on placebo. Both groups improved, which is expected given that all participants received lifestyle guidance, but the drug group improved more. - **Blood ketones rose** in the drug group. - **IGFBP-3 increased.** This protein binds IGF-1 and reduces its activity. Lower IGF-1 signaling is one of the most consistent features of caloric restriction across species. - **Granzyme B**, a protein cytotoxic T cells use to kill damaged or infected cells, went up. - Uric acid, weight, BMI and blood sugar all improved more than on placebo. ### Senescent cells (2024) A paper in *Nature Aging* reported that the SGLT2 inhibitor canagliflozin reduced the burden of **senescent cells** in mice. Senescent cells are damaged cells that stop dividing but do not die, and they secrete inflammatory signals that affect surrounding tissue. The drug appeared to work by helping the immune system clear them. In a mouse model of accelerated aging, treatment also extended lifespan. ### Lifespan in normal mice The US National Institute on Aging's Interventions Testing Program, which tests compounds in genetically diverse mice at three independent sites, found that canagliflozin extended median lifespan by about 14% in male mice. It had no effect in females. ### The drug-ranking paper (2022) A 2022 review in *Aging Cell* scored FDA-approved drugs on their evidence as candidates for targeting aging, using criteria that included hallmarks of aging, animal lifespan data and human outcomes. SGLT2 inhibitors received the highest score, ahead of metformin. ## Where the Evidence Stops These results are interesting. They are also easy to overstate, so here is what they do not show. - **The telomere trial was in people with type 2 diabetes, average age around 52.** It says nothing direct about healthy people. - **It was small and short.** 142 people for six months is a signal, not a conclusion. - **Telomere length is a surrogate marker.** Measurements are noisy, and longer telomeres in blood cells have never been shown to translate into a longer life in a trial. - **Henagliflozin is approved in China, not in the US or Europe,** and the manufacturer supplied the drug for the study. - **The lifespan data are from mice,** and in the most rigorous mouse study the benefit appeared in males only. - **The fasting state is partial.** The ketone levels involved are modest, roughly comparable to the early part of a fast. It is not equivalent to a multi-day fast. For a metabolically healthy person, whether an SGLT2 inhibitor adds healthy years is an open question. No trial has tested it. ## The Risks Are Real SGLT2 inhibitors are prescription drugs. In the US the approved versions are empagliflozin (Jardiance), dapagliflozin (Farxiga) and canagliflozin (Invokana), and they are approved for type 2 diabetes, heart failure and chronic kidney disease. Using them in healthy people for longevity is off-label. The side effects follow from the mechanism: - **Genital yeast infections and urinary tract infections.** Sugar in the urine feeds yeast and bacteria. This is the most common reason people stop the drug. - **Dehydration and low blood pressure.** Glucose pulls water with it. The risk is higher in older adults and in anyone taking a diuretic. - **Euglycemic ketoacidosis.** This is the serious one. Ketones can climb to dangerous levels while blood sugar reads normal, so the usual warning sign is missing. It is rare, and the risk goes up with prolonged fasting, very low-carb diets, heavy alcohol use, acute illness and surgery. That third point matters for this audience specifically. Stacking an SGLT2 inhibitor on top of extended fasting or strict keto is exactly the combination that raises ketoacidosis risk. This is a drug to use with a physician who knows your labs, not one to source on your own. ## The Practical Protocol You do not need a prescription to act on the biology here. The drug's interesting effects come from lowering insulin, raising ketones and shifting AMPK and mTOR. Those levers are available to everyone. ### 1. Use an overnight fasting window A daily 14 to 16 hour gap between dinner and your first meal lowers insulin for long enough to start producing ketones in most people. Going to 18 hours occasionally pushes further. This is the real version of what the drug imitates. ### 2. Add a longer fast occasionally A 24-hour fast once or twice a month reaches the range the drug appears to mimic. If you have a medical condition or take glucose-lowering medication, clear this with your doctor first. ### 3. Train, including Zone 2 Exercise activates AMPK directly and empties glycogen, which makes it easier to reach ketosis during the next fasting window. Two to three hours of easy aerobic work per week plus resistance training covers most of the benefit. ### 4. Reduce refined carbohydrate, especially late in the day Lower evening glucose and insulin means you enter your overnight fast in a better starting position. Our comparison of [Mediterranean, keto and intermittent fasting](/blog/mediterranean-diet-vs-keto-vs-intermittent-fasting-what-the-evidence-actually-supports) covers which eating patterns have outcome data behind them. ### 5. Consider exogenous ketones as a tool, not a substitute [Keto BHB](/products/keto-bhb) supplies beta-hydroxybutyrate, the same ketone your liver produces during a fast. It raises blood ketone levels and many people use it to make a fasting window or a low-carb transition easier. It does not lower insulin or switch on the fasting pathways by itself, so treat it as a support for fasting and not a replacement for it. ### 6. Support the NAD+ side AMPK and SIRT1 signaling depend on NAD+, which declines with age. Fasting and exercise both raise it. If you supplement, [NAD+](/products/nad) pairs logically with a fasting routine. Our [NMN vs NR breakdown](/blog/nmn-vs-nr-which-nad-precursor-actually-works) covers the precursor evidence. ### 7. If appetite is what breaks your fasts [The Natural GLP-1 Reset](/products/the-circadian-reset-protocol) is our guide to food timing, protein and fiber strategies that improve satiety signaling, which is usually the limiting factor in sustaining a fasting practice. ### 8. If you already qualify for the drug, ask about it If you have type 2 diabetes, heart failure or chronic kidney disease, SGLT2 inhibitors have strong outcome data for those conditions. That conversation is worth having with your doctor regardless of the longevity angle. ## The Bottom Line A compound first pulled from apple bark in 1835 led to a class of drugs that makes the kidney discard sugar, lowers insulin, and produces a measurable ketone rise in people who are still eating normally. Those drugs reduce cardiovascular death in high-risk patients, clear senescent cells in mice, and in one small trial lengthened telomeres in people with diabetes. That is a strong case for more research. It is not yet a case for healthy people to take a prescription drug with real side effects. The fasting state these drugs partly imitate is something you can produce yourself, at full strength, with a clock and some discipline. *This article is for educational purposes only and is not medical advice. SGLT2 inhibitors are prescription medications. Do not start, stop or combine any medication with fasting without talking to your physician.*