

Genentech just dropped $190 million upfront (and up to $2.3 billion total) on an obesity drug that doesn't work like Ozempic. HM17321 takes a completely different approach: burning fat while actually building muscle, a trick no GLP-1 drug can pull off.
Every obesity drug on the market has the same dirty secret: when you lose weight, you don't just lose fat. You lose muscle too. Sometimes a lot of it. It's like renovating your house by tearing down load-bearing walls; sure, the place looks different, but the structural integrity takes a hit.
Genentech thinks it found a fix. And it's willing to pay up to $2.3 billion to prove it.
The Roche subsidiary just locked up exclusive worldwide rights (outside South Korea) to HM17321, a first-in-class obesity molecule from South Korea's Hanmi Pharmaceuticals. The deal includes $190 million upfront, with the rest tied to development, regulatory, and commercial milestones, plus tiered royalties on future sales.
What makes this molecule special? It's not another GLP-1 drug. It doesn't work like Ozempic or Zepbound. Instead, HM17321 takes a completely different approach to weight loss, one that could reshape how we think about the entire obesity market.
HM17321 is a long-acting UCN2 analog, which is a fancy way of saying it activates a specific receptor in your body called CRF2 (CRFR2). If GLP-1 drugs are the appetite police, telling your brain you're full, HM17321 is more like a body recomposition coach. It works directly on fat and muscle tissue.
In preclinical studies (animal testing, not yet humans), the drug did two things simultaneously. It burned fat by cranking up lipolysis (fat breakdown) and dialing down lipogenesis (fat creation). And it preserved or even grew lean muscle mass by activating protein synthesis pathways, including mTOR signaling, which is essentially the molecular on-switch for muscle building.
The results showed up in both obese mice and obese primates. In monkeys, the drug reproducibly drove fat loss while maintaining or increasing lean mass. That's not just "less muscle loss." That's actual muscle gain during weight loss, a combination that current obesity drugs simply can't deliver.

Ascendis Pharma and BioMarin just ended an 18-month patent war spanning three continents with a royalty deal that has BioMarin collecting 20% of YUVIWEL's U.S. sales. The settlement removes a major legal cloud, but the competitive chess match in achondroplasia is just getting started.


Join thousands of biotech professionals who start their day with our free, daily briefing.
Hanmi has been emphatic about this distinction. The company says HM17321 doesn't merely prevent muscle wasting; it actively increases lean mass. If that holds up in humans, it would be a genuine game-changer.
This deal doesn't exist in a vacuum. Genentech has been on an obesity shopping spree, assembling a multi-asset portfolio with the stated goal of becoming a top-three obesity company by 2030.
The foundation was laid in 2023 with the $2.7 billion acquisition of Carmot Therapeutics, which brought in CT-388, a dual GLP-1/GIP agonist now heading into Phase III trials. Roche has also partnered with Zealand Pharma on petrelintide for obesity, and it's planning a combination study of CT-388 plus petrelintide later in 2026. Add CT-996 (an oral GLP-1 candidate) to the mix, and you start to see the strategy.
Genentech isn't trying to build one drug. It's building a franchise. CT-388 handles the incretin-based weight loss. Petrelintide adds an amylin angle. And now HM17321 brings something nobody else in the top tier has: a non-GLP-1 mechanism designed to protect muscle.
Think of it like assembling a basketball roster. You need your scoring guard (CT-388), your stretch forward (petrelintide), and your defensive anchor (HM17321). Each one does something the others can't, and together, they cover the full court.
The obesity drug market is a monster, dominated by Novo Nordisk and Eli Lilly. Their drugs work. Semaglutide and tirzepatide deliver impressive weight loss. But industry analysts have been saying for months that the next competitive battleground isn't just how much weight patients lose. It's what kind of weight they lose.
When patients on GLP-1 drugs shed 15-20% of their body weight, a significant portion comes from lean mass. For older patients or those already at risk for sarcopenia (age-related muscle loss), that's a real clinical concern. Lose enough muscle, and you trade one health problem for another.
This is exactly why analysts viewed the Hanmi deal as validation of a broader market shift. South Korean brokerage Kiwoom Securities raised Hanmi's target price by 25.9% to 730,000 won after the announcement, arguing that the upfront payment reflected a strong assessment of HM17321's differentiated mechanism. Hanmi's stock jumped on the news, with investors treating the deal as proof that the pipeline has real, externally validated value.
The fact that Genentech paid $190 million upfront for a Phase I asset (Hanmi is still completing the first human study) tells you something important about the premium the market places on body-composition science. This isn't a late-stage, de-risked molecule. It's early. And someone still wrote a very large check.
Genentech isn't the only company chasing next-generation obesity approaches. Amgen's MariTide is advancing through clinical development, with Phase 3 results expected in 2027. Viking Therapeutics reported a 12.2% weight-loss result at 13 weeks for its oral VK2735 program. Lilly is pushing retatrutide, a triple agonist targeting GLP-1, GIP, and glucagon receptors. Novo has CagriSema and amycretin in its pipeline.
But none of these are UCN2-based. None are specifically designed to build muscle while burning fat. If HM17321's preclinical profile translates into human data, it could occupy a category of one, either as a standalone therapy or (more likely) as a combination partner layered on top of existing GLP-1 drugs.
That combination angle matters. Kiwoom Securities specifically flagged HM17321's potential for combination therapy as a key part of its value. Imagine a world where patients take a GLP-1 for appetite suppression and add HM17321 to protect their muscles. It's the kind of one-two punch that could redefine treatment protocols.
Under the deal terms, Hanmi will finish the ongoing Phase I trial. Then Genentech takes over, running Phase II and everything after. The timeline from here is long; this drug won't be on pharmacy shelves tomorrow, or even in 2028. But the strategic implications are immediate.
Roche/Genentech has now committed billions of dollars in deal value across its obesity build-out (Carmot at $2.7 billion, HM17321 at up to $2.3 billion, plus the Zealand partnership). That's not dabbling. That's declaring war on Novo and Lilly.
The question is whether HM17321's animal data will hold up in people. Plenty of preclinical superstars have flamed out in human trials. But if this one works, if it really can help patients lose fat and gain muscle at the same time, Genentech won't just have a competitive obesity portfolio. It'll have the one thing every other company in the space is missing.
For nearly 30 years, patients with essential thrombocythemia had zero new FDA-approved treatments. PharmaEssentia's BESREMi just broke that streak with trial results that crushed the competition seven to one.