In 2004, a baby was born in Germany who was, by any normal measure, impossible. He was a few days old, and he already had visible muscles — his thighs and upper arms looked like a tiny bodybuilder. Doctors checked him for every disease that could explain it, and found nothing wrong. He wasn't sick. He was just extraordinarily, effortlessly strong. [1]
When they looked at his genes, they found the answer. He'd been born with both copies of a single gene switched off — a gene that makes a protein called myostatin. And myostatin, it turns out, has one job in your body: to put the brakes on muscle growth. It's one of the reasons you don't keep adding muscle forever. This baby's brakes were simply gone. [1]
Scientists had found the myostatin gene just seven years earlier, in 1997. And the moment they understood what it did, the implications were staggering. If you could switch myostatin off in anyone, on purpose, you might be able to beat the diseases that waste muscle away — like muscular dystrophy — and even sarcopenia.
There was just one problem. For nearly thirty years, nobody could switch the myostatin gene off safely. But the breakthrough — which has just been published — arrived in time to address a growing problem that didn't even exist in 1997.
Table of Contents
- The problem that didn't exist in 1997
- The thirty-year quest to switch off myostatin
- The two drugs out in front
- EMBRAZE: putting the clean drug to the test
- What the internet will try to sell you
- References
The problem that didn't exist in 1997
That problem was an unintended consequence of what's happened to weight loss in just the last few years. We have become astonishingly good at melting fat. A decade ago, the best drugs took about eight percent off your body weight, and with the burden of daily injections, most people gave up. [2][3]
Then came semaglutide — Ozempic — and that increased weight loss to around fifteen percent. [4] Then tirzepatide pushed past twenty percent. [5] And the newest one in line, a triple-target drug called retatrutide, took people down by about twenty-eight percent of their body weight in its big trial. For someone starting at a hundred and ten kilos, that is more than thirty kilos gone, from a weekly injection. That is bariatric-surgery territory, without the surgery. [6]
But here is the catch, the one all over the internet. When you lose that much weight, some of what you lose isn't fat. It's muscle. And you've probably seen what that can look like — someone drops a huge amount of weight and ends up smaller, but hollowed out and gaunt. People started to panic that these drugs were melting people's muscle away.
But we need some crucial context here, because the truth is more boring and more reassuring than the panic. When researchers pooled twenty-two trials of these drugs and added up exactly where the lost weight came from, about a quarter of it was muscle, and three quarters was fat. [7]
Losing Weight Without Losing Muscle?
Get an evidence-based, personalized health plan — including how to protect your muscle and strength while you lose fat.
Get Your Personalized Health PlanAnd here's what people miss. That 25% lean mass loss is not unique to Ozempic. A quarter is roughly what you lose when you diet the old-fashioned way. It's roughly what you lose after weight-loss surgery. Doctors have a decades-old rule of thumb that about a quarter of any weight you lose is lean tissue. [8] So these drugs aren't doing something sinister to your body. They're doing what losing weight has always done.
And when one group went further and measured the muscle directly with MRI scans — comparing people on tirzepatide against thousands who lost weight naturally — the muscle loss was, for the same amount of weight, basically identical. And the fat marbled inside the muscle, the stuff that makes older muscle weak, actually went down. [9]
Now, we can counter this muscle loss somewhat with good protein intake and resistance exercise. But the dream result is that we only lose fat — not lean mass. We want to protect our muscles during the weight-loss journey. And the key to that holy grail of losing fat and not muscle brings us back to myostatin.
The thirty-year quest to switch off myostatin
In 1997, researchers bred mice with the myostatin gene deleted — and the mice grew enormous. They looked exactly like the "double-muscled" cattle farmers had bred for centuries — bulging, blocky, with twice the muscle of a normal animal, from a single switch turned off. [10][11]
The dream was obvious. If turning the myostatin gene off doubles muscle in a mouse, then a drug that turns it off in a person could rebuild the wasted muscle of dystrophy or old age. And then, in 2004, came the proof it might be safe in humans — that German baby. He was a natural human myostatin knockout, walking around extraordinarily strong, and completely healthy. Turn this gene off, it seemed, and nothing breaks. [1]
So the target was obvious, the safety looked reassuring, and the prize was enormous. And then, for nearly thirty years, almost everyone who tried to hit it failed. Why? The tool you'd use is a monoclonal antibody — think of it as a guided missile scientists design to lock onto one specific molecular shape and neutralise it. The problem is that myostatin, once it's switched on and active, looks almost identical to a whole family of related proteins your body needs for other jobs. So a missile aimed at active myostatin tends to hit its cousins too — and you get side effects, or you blunt things you never meant to touch. Decades of attempts, and the verdict in the field became almost a punchline: great expectations, limited success. [12]
The breakthrough, when it came, was about being more precise — about selectivity. The newest drug, apitegromab, doesn't go after active myostatin, where all the look-alike cousins are. It grabs myostatin earlier — in its inactive, folded-up precursor form, before it's ever switched on. And that precursor shape is unique. Nothing else in the family looks like it. So the missile finally locks onto myostatin and only myostatin, and leaves everything else alone. That selectivity — hitting one target cleanly — is the exact thing thirty years of attempts had been missing. [13]
And before anyone ever pointed this drug at weight loss, it earned its stripes somewhere far more demanding. Apitegromab was first tested in spinal muscular atrophy — a brutal genetic disease that destroys the muscle of children. In trials there, it improved motor function in kids who were losing it, and it did so safely. [14] So the tool had finally matured — a clean, selective, human-proven way to take the brake off muscle formation. And it matured at the exact moment the world handed it a brand-new use. Because just as apitegromab was proving itself, the new weight-loss drugs were driving faster fat loss — and faster muscle loss — than medicine has ever seen.
The two drugs out in front
Two medications now lead the race to apply this breakthrough to the problem of muscle loss during weight loss. They represent two completely different philosophies of how aggressive you should be.
The first is apitegromab — the clean one. It does exactly what we just described: it selectively blocks myostatin, the brake, so your body holds on to more muscle while you lose fat. It spares muscle.
But bimagrumab is the showstopper. Instead of just blocking myostatin, it blocks a broader docking port — the activin receptor — which is a much heavier hand on the whole system. On its own, it does something no diet drug does — people's muscle actually grew as their fat fell — though the total weight loss was modest. Bolt it onto semaglutide and you get the headline number: substantial weight loss, about 92% of it from fat, with muscle largely protected. It is, genuinely, the closest thing anyone has made to "lean and ripped from a vial." [15]
But that power comes with a serious downside. Most people on bimagrumab got muscle spasms. About a third got acne. And more seriously, it pushed their LDL cholesterol — the kind that clogs arteries — up by as much as nearly 17 percent. That is the wrong direction for your heart. [15] And then, quietly, in September 2025, the company developing one of the big bimagrumab combinations paused that program. [16]
All of this creates a lingering worry: the harder a drug pushes to actually build muscle, the louder the heart-safety question gets. Push gently, you spare muscle cleanly. Push hard enough to grow it, and your cholesterol starts heading the wrong way. Which is exactly why the gentle, clean drug — apitegromab — matters so much. The whole bet is that you can protect muscle without poking the heart. So does it work?
EMBRAZE: putting the clean drug to the test
This is the trial that inspired this article. It's called EMBRAZE, and the design is beautifully simple. They took a hundred and two adults with obesity and put every single one of them on tirzepatide. Then they split them in two. One half also got apitegromab, the muscle-protector. The other half got a dummy infusion on top. Neither the patients nor the doctors knew who got which. So the only difference between the two groups was that one molecule — anything that showed up in the results had to be the apitegromab doing it. After six months, they scanned everyone and measured precisely how much of the lost weight was fat, and how much was muscle. [13]
And it worked. The group on apitegromab lost about half as much muscle. Both groups lost the same total weight — but in the apitegromab group, a far bigger share of that weight was fat: about eighty-five percent fat, versus seventy percent in the placebo group. After thirty years of failure, the brake had finally been let off, safely, in adults losing weight. For the first time, researchers have shown that the ancient rule — lose weight, lose muscle — isn't inevitable. It's a hugely significant achievement. [13]
But we also need to be clear about what this study found, because there is a key result the headlines skipped. The apitegromab group were not measurably stronger. The trial tested it directly, using grip strength and how easily people could stand up out of a chair. On both, the muscle-protected group was no better than the dummy-infusion group. [13]
The drug kept the muscle on the scan. But that extra muscle didn't let people do anything more with their bodies. And those are two different things. Keeping muscle that shows up on a scan is the means. Being stronger, more capable, less frail — that's the actual goal. And only the first one was proven here. There's even a fair question about whether the spared "lean mass" on the scan is fully working muscle at all — the scan can't tell you that. [13]
Still, this trial represents where the future is potentially headed: a future where we can all be a healthy weight and have strong muscles.
What the internet will try to sell you
One blunt word on what the internet will try to sell you here. Search "how to keep muscle on Ozempic" and you'll fall into a world of grey-market SARMs and muscle peptides with names like YK-11. For protecting muscle during weight loss, these have essentially no proper human evidence — no well-designed, well-conducted randomized controlled trials. What they have instead is a real track record of liver damage and worse. Not worth your money, and not worth your liver. [17]
So here's what to actually do about muscle loss while on a GLP-1 medication today. The thing that works right now — with no prescription and no risk — is the unglamorous one: resistance training a couple of times a week, and enough protein. That's the part you control, and it already works. I take 1.25 milligrams of tirzepatide a week myself, and I protect my muscle exactly that way — lifting and good protein.
The one supplement worth adding is creatine, the most studied ingredient there is for muscle and strength. [18] But just because I take a supplement, that doesn't mean you have to as well.
Creatine is one of the ingredients in MicroVitamin+ Powder, alongside TMG.
Losing Weight Without Losing Muscle?
Get an evidence-based, personalized health plan — including how to protect your muscle and strength while you lose fat.
Get Your Personalized Health PlanReferences
1. https://www.nejm.org/doi/full/10.1056/NEJMoa040933
2. https://www.nejm.org/doi/full/10.1056/NEJMoa1411892
3. https://pmc.ncbi.nlm.nih.gov/articles/PMC7448157/
4. https://www.nejm.org/doi/full/10.1056/NEJMoa2032183
5. https://www.nejm.org/doi/full/10.1056/NEJMoa2206038
7. https://www.metabolismjournal.com/article/S0026-0495(24)00341-X/abstract
8. https://onlinelibrary.wiley.com/doi/10.1111/obr.12143
9. https://www.thelancet.com/journals/landia/article/PIIS2213-8587(25)00027-0/fulltext
10. https://www.nature.com/articles/387083a0
11. https://pubmed.ncbi.nlm.nih.gov/9356471/
12. https://www.mdpi.com/2073-4409/10/3/533
13. https://doi.org/10.1038/s41591-026-04440-4
14. https://www.thelancet.com/journals/laneur/article/PIIS1474-4422(25)00225-X/fulltext
15. https://doi.org/10.1038/s41591-026-04204-0
16. https://www.biopharmadive.com/news/lilly-terminate-obesity-trial-bimagrumab-muscle-diabetes/761105/
18. https://jissn.biomedcentral.com/articles/10.1186/s12970-017-0173-z
