Roche just spent $190 million upfront on an obesity peptide that does not touch the GLP-1 receptor. On August 24, 2026, Hanmi Pharmaceutical announced an exclusive licensing agreement with Genentech for HM17321 — a urocortin-2 analog aimed at a completely different receptor, CRF2, that sits on skeletal muscle. Total deal value runs to $2.3 billion in milestones plus tiered royalties.
The reason a company with two obesity assets already in the clinic paid that for a Phase 1 molecule is the body-composition problem. Every incretin drug on the market reduces fat and lean mass together, and roughly a quarter to 40% of the weight lost on a GLP-1 is non-fat tissue depending on which cohort you read. HM17321 is the first licensing deal of this size where the pitch is not more weight loss — Hanmi's own preclinical comparison shows slightly less total weight loss than semaglutide — but better weight loss.
Research-context information only. Compounds discussed below are research peptides and supplements; some are investigational drugs not approved by the FDA. Protocols, doses, and reactions reported come from published research and self-reported community sources. Possession or use of investigational drugs outside an authorized clinical trial may be illegal in your jurisdiction. This article reports what has been documented, not what should be done. Consult a licensed physician for personal medical decisions.
What Roche actually bought
Urocortin-2 is an endogenous neuropeptide in the corticotropin-releasing factor family. It binds the CRF2 receptor, which is expressed in the central nervous system and — the part that matters commercially — in skeletal muscle. Activating CRF2 in muscle tissue is associated with increased muscle mass and increased fat oxidation. HM17321 is Hanmi's long-acting, CRF2-selective analog of that peptide, built on the company's own half-life extension platform.
This is not a new target so much as a revived one. Pfizer investigated urocortin-2 as a metabolic target years ago without producing an approved drug, and Denmark's Gubra started a Phase 1/2 for its own urocortin-2 candidate in July 2026. Hanmi's asset is roughly eight months ahead of Gubra's on the clinic timeline.
The disclosed terms:
| Term | Detail |
|---|---|
| Upfront | $190 million |
| Milestones | Up to $2.3 billion (development, regulatory, commercial) |
| Royalties | Tiered, on net sales |
| Territory | Worldwide excluding South Korea (Hanmi retains) |
| Handoff | Hanmi completes Phase 1; Genentech develops from Phase 2 |
| Announced | August 24, 2026 |
What Hanmi has reported preclinically, presented across ADA and ObesityWeek posters and repeated in the licensing announcement: weight reduction in diet-induced obese mice as monotherapy and in combination with GLP-1 drugs; in obese non-human primates, weight loss driven predominantly by fat mass; and in a head-to-head against semaglutide, slightly less total body weight loss but greater fat mass loss, with lean mass increased and intramuscular lipid content reduced. Hanmi's R&D head In-Young Choi framed the rationale plainly to BioPharma Dive: "The paradigm of obesity treatment is evolving beyond simple weight reduction toward improving body composition."
Genentech's stated interest includes fixed-dose combination products pairing HM17321 with incretin-based treatments — which is the more likely commercial shape of this. Not a replacement for a GLP-1, but a second agent in the same syringe.
Why the muscle question is worth $2.3 billion
The lean-mass problem has been accumulating evidence for eighteen months, and it is the single most consistent complaint in GLP-1 community threads that is also backed by published data.
A 670,422-person nference analysis reported tirzepatide users losing roughly 2% more lean mass at 12 months than semaglutide users, with a "depletive metabotype" subgroup at 10.3% — our breakdown is in Tirzepatide Loses More Muscle Than Semaglutide. A 24-month semaglutide cohort in adults 65 and over reported accelerated muscle loss alongside falling gait speed, covered in GLP-1s Cause Sarcopenia in Seniors. And the pharma response so far has been antibody-based: the BELIEVE trial of bimagrumab plus semaglutide reported 22.1% weight loss with 92.8% of it from fat, which we covered in BELIEVE Trial: GLP-1 Combo Keeps 92% Fat Loss.
HM17321 is the peptide-side answer to the same question, and it arrives with a different economic argument. Bimagrumab, trevogrumab and apitegromab are monoclonal antibodies — expensive to make, expensive to combine. A long-acting peptide that can be co-formulated into an existing weekly injection is a much cheaper path to a combination product, which is plausibly what $190 million upfront on a Phase 1 asset is really buying.
It also slots into a Roche obesity portfolio that is now three assets deep: petrelintide, the amylin analog partnered with Zealand, enicepatide (CT-388), the dual GLP-1/GIP agonist from the Carmot acquisition, and now a non-incretin muscle agent that could pair with either.
What it changes for anyone on a GLP-1 today
Nothing, immediately — and that is the honest answer. HM17321 is in a first-in-human Phase 1 study (NCT07219589) enrolling healthy volunteers and people with obesity. Roche does not take the wheel until Phase 2. There is no human efficacy data, no dosing schedule in the public record, and no numeric body-composition figure from any peer-reviewed publication. A registrational package is years away, and nothing under that name exists in the research-supply market.
What the deal does confirm is a direction of travel: the industry has stopped treating total weight loss as the only endpoint that matters, and body composition is now something a large pharma will underwrite at nine figures. For anyone currently running a GLP-1 protocol, the practical levers are the ones that already have documentation behind them — resistance training and protein intake first, with the growth-axis and mitochondrial compounds people layer alongside them a distant second. Our ranked breakdown of what is actually documented for each is in Preserve Muscle on Semaglutide & Tirzepatide.
Nothing here moved availability or pricing. Current per-vendor $/mg and COA coverage are on Best Semaglutide Vendors, Best Tirzepatide Vendors and Best Tesamorelin Vendors, with active codes on the deals page.
