The Diabetes Drug That's Turning Medicine Upside Down
Semaglutide was approved to lower blood sugar. Then it dramatically cut heart attacks. Then kidney failure. Then it showed signs of slowing Parkinson's disease. Scientists are still figuring out why a gut hormone seems to protect nearly everything.
Contents 8 sections
In 2017, the U.S. Food and Drug Administration approved semaglutide — a once-weekly injection called Ozempic — to help control blood sugar in people with type 2 diabetes. It mimics a gut hormone called glucagon-like peptide-1, or GLP-1, that the body naturally releases after eating. It tells the pancreas to make insulin. It slows the emptying of the stomach. It signals the brain to feel full.
That was the plan. What happened next was something nobody planned.
The Drug That Rewrote Obesity Medicine
The first surprise came when doctors noticed their patients on Ozempic losing substantial weight — far more than any prior medication had achieved. Novo Nordisk ran a dedicated trial. The results, published in the New England Journal of Medicine in 2021, were striking.
In a double-blind trial of 1,961 adults with obesity, participants who took semaglutide at a 2.4 mg weekly dose lost an average of 14.9% of their body weight over 68 weeks. The placebo group lost 2.4%. More than half the people taking semaglutide lost at least 15% of their body weight — something that had previously required bariatric surgery.
The FDA approved a higher-dose version called Wegovy for chronic weight management in 2021. The drug was hailed as a breakthrough. Then came the second surprise.
It’s Not About Weight
The cardiovascular results arrived in late 2023, and they changed the conversation.
The SELECT trial enrolled 17,604 people who had preexisting cardiovascular disease — prior heart attacks, strokes, or arterial disease — and obesity, but crucially, no diabetes. These were patients whose primary risk factor was weight, not blood sugar. They were randomized to semaglutide or placebo for nearly four years.
The results, published in the New England Journal of Medicine in December 2023, were unambiguous: semaglutide reduced the combined risk of cardiovascular death, nonfatal heart attack, or nonfatal stroke by 20% (hazard ratio 0.80; p<0.001).
That’s a hazard ratio comparable to statins — one of the most impactful drug discoveries of the 20th century. And it happened in people without diabetes, which meant the benefit wasn’t simply about improved blood sugar control. Something else was happening.
Researchers still debate exactly what. The weight loss itself helps — even modest reductions in body mass improve heart function, blood pressure, and inflammation. But the magnitude of benefit appears to exceed what weight loss alone would predict. GLP-1 receptors are present in the heart muscle and in blood vessel walls. The drug may be acting on cardiovascular tissue directly.
Whatever the mechanism, the clinical implication was clear: semaglutide wasn’t a diabetes drug that happened to help the heart. It was a cardiovascular drug that also helped with diabetes and obesity.
Then the Kidneys
The FLOW trial was still running when SELECT published its results. Its target was different — patients with type 2 diabetes and chronic kidney disease, a combination that typically leads relentlessly toward dialysis.
The trial was stopped early at a prespecified interim analysis. The data were already conclusive.
Published in the New England Journal of Medicine in July 2024, the FLOW results showed that semaglutide reduced the risk of major kidney disease events — kidney failure, needing dialysis, or a 50% drop in kidney function — by 24% (hazard ratio 0.76; p=0.0003). Death from cardiovascular causes fell by 29%. Death from any cause fell by 20%.
All-cause mortality down by a fifth. In a drug whose original FDA approval said nothing about mortality at all.
The FLOW trial was stopped because it was deemed unethical to continue giving placebo-arm patients a placebo when the drug’s benefit was already so clear.
A Strange Hint From Parkinson’s Research
In April 2024, a very different paper appeared in the New England Journal of Medicine. This one came from neurology.
The LIXIPARK trial tested lixisenatide — another GLP-1 receptor agonist used for diabetes — in 156 people with early Parkinson’s disease. Parkinson’s is a neurodegenerative disease caused by the progressive death of dopamine-producing neurons in the brain. There is no treatment that slows that progression. All current therapies manage symptoms.
After 12 months, the story was different in the two groups. Patients taking placebo showed the expected worsening on the Movement Disorder Society’s motor disability scale: their scores deteriorated by about 3 points. Patients taking lixisenatide had essentially stable scores — a change of essentially zero.
After a two-month washout period — when both groups had stopped taking the drug — the gap persisted. This is what researchers call a disease-modifying signal: if you’re just masking symptoms, the benefit disappears when the drug does. Here it didn’t.
The trial was a phase 2 study. It was designed to detect a signal, not to prove efficacy. The sample size was modest. More trials are needed. But the signal was there, and the mechanism is plausible: GLP-1 receptors are expressed in the brain, including in regions affected by Parkinson’s. Animal studies had previously shown neuroprotective effects of GLP-1 agonists in models of dopaminergic cell loss.
The implications — if the effect holds in larger trials — would be extraordinary. Parkinson’s disease affects around 10 million people worldwide, and there has been no disease-modifying treatment in the field’s entire history.
And Then: Addiction
The breadth keeps widening.
People taking GLP-1 drugs for weight loss began reporting something odd: they were drinking less alcohol. Not because they were trying to — the craving just diminished. Some reported reduced desire for cigarettes, opioids, and gambling. These were anecdotes, but they were consistent and widespread enough to prompt formal investigation.
A 2025 systematic review in Frontiers in Pharmacology, examining 41 studies — 35 preclinical and 6 clinical — found that GLP-1 receptor agonists consistently reduced substance intake, relapse-like behaviors, and cue-induced drug seeking across multiple drug classes in animal models. The clinical studies provided preliminary human support, particularly for alcohol use disorder.
The mechanistic story fits: GLP-1 receptors are present in the brain’s mesolimbic reward circuitry — the nucleus accumbens, the ventral tegmental area — the same regions that generate the drive toward addictive substances. Activating those receptors appears to dampen reward-seeking behavior broadly, not just for food.
This is still early-stage science. The clinical trials in addiction medicine are small and short. Larger randomized controlled trials are underway. But the consistency of the preclinical signal is striking.
What Is GLP-1, Really?
The GLP-1 story has forced a rethinking of what this hormone actually does.
GLP-1 is released by cells in the small intestine after a meal. It signals the pancreas to release insulin. It slows gastric emptying. It suppresses glucagon. That’s the textbook version.
But GLP-1 receptors are expressed throughout the body in ways that are still being catalogued: in the heart, in blood vessels, in the kidneys, in the brain’s reward and motor circuits, in the immune system. It may function less like a narrow metabolic switch and more like a broad physiological coordinator — something that registers “nutrients have arrived” and cascades signals through multiple systems simultaneously.
The synthetic GLP-1 agonists like semaglutide are designed to be far more potent and longer-lasting than the natural hormone. They don’t just nudge the system; they hold it in a sustained state that natural GLP-1 never maintains. That may be why the effects extend so far beyond what anyone expected from a simple post-meal signal.
What We Don’t Know Yet
The picture is not entirely rosy, and intellectual honesty requires saying so.
The drug’s most common side effects — nausea, vomiting, diarrhea — are frequently severe enough to cause discontinuation. In the SELECT trial, 16.6% of semaglutide patients discontinued permanently due to adverse events, versus 8.2% in the placebo group.
The long-term effects of sustained GLP-1 receptor activation are not fully characterized. Questions about thyroid cancer risk in people with a personal or family history of medullary thyroid carcinoma have led to a boxed warning in the U.S. The muscle loss associated with the rapid weight reduction is a concern that ongoing research is trying to address.
The neurodegenerative and addiction findings, while exciting, are preliminary. The LIXIPARK trial was phase 2. The addiction data mostly comes from animals and small human studies. These findings deserve rigorous follow-up — not dismissal, but not premature celebration either.
And the drugs are expensive. Access is severely limited in lower-income settings. The public health benefits of a drug that reduces heart attacks, kidney failure, and potentially neurodegeneration will only be realized if the drug reaches the people who need it.
A Drug That Happened
The history of medicine is full of drugs discovered by accident — penicillin, aspirin, sildenafil. Semaglutide was not an accident; it was deliberately designed. But its discovered breadth of effect was not anticipated. Nobody designed a drug to protect the heart and the kidneys and the brain and blunt addiction simultaneously.
What seems to be happening is that we stumbled onto a hormone system that does something large and important — something we didn’t fully appreciate when we first mapped it — and we built a molecule that turns it up.
The science is moving fast. There are now trials underway testing GLP-1 agonists in heart failure, liver disease, sleep apnea, Alzheimer’s disease, and multiple addiction disorders. New molecules combining GLP-1 with other gut hormones — tirzepatide combines GLP-1 and GIP, retatrutide targets three hormones simultaneously — are already showing even larger effects on weight.
We may be at the beginning of understanding something fundamental about how the body regulates itself. Or we may be at the beginning of learning where the limits are.
Either way, this is a remarkable scientific moment. A drug that started as a blood sugar medication has, in the span of seven years, accumulated evidence that it prevents heart attacks, halts kidney failure, shows early signs of slowing Parkinson’s disease, and blunts addiction.
Whatever comes next, the GLP-1 story is one of the most extraordinary in modern medicine.