Genentech Just Paid $190M for a Drug That's Never Been in a Human

Why Genentech Is Writing Big Checks

This deal doesn't exist in a vacuum. Genentech (and parent company Roche) has been on an absolute shopping spree in obesity, assembling a portfolio that looks less like a pipeline and more like a starting lineup.

The roster so far:

• CT-388: a once-weekly dual GLP-1/GIP shot, acquired through the Carmot deal, now their lead clinical obesity asset with Phase 3 work underway.
• Petrelintide: an amylin analog licensed from Zealand Pharma in a deal worth over $5.3 billion, designed to work as a standalone therapy or in combination with CT-388.
• CT-996: an oral GLP-1 option, because not everyone wants to inject themselves weekly.
• HM17321: the new muscle-sparing wildcard from Hanmi.

The strategy is obvious once you see the pieces together. Roche isn't trying to beat Novo Nordisk and Eli Lilly with one drug. It's building a full menu of options: injectable, oral, incretin-based, non-incretin, standalone, and combination. Different mechanisms for different patients, different stages of treatment, different competitive angles.

The Zealand deal alone involved roughly $1.65 billion upfront. Adding $190 million for Hanmi's early-stage asset shows Roche is willing to pay premium prices across the board, from late-stage to very early-stage, to fill every slot.

The Billion-Dollar Bet on Body Composition

The weight-loss drug market has a dirty secret: patients on GLP-1 drugs lose a significant amount of lean muscle mass along with fat. For older patients or those with limited mobility, that's a real problem. Losing 30 pounds sounds great until a quarter of it is muscle you needed.

This is the gap HM17321 is designed to fill. If it works in humans the way it works in animals (a very large "if" that deserves bold: if), it could be combined with GLP-1 drugs to preserve or even increase muscle while the incretin does the heavy lifting on appetite and weight.

Analysts are already modeling the upside. KB Securities estimated HM17321 could hit $1.75 billion in annual sales by year five after launch. Shinhan Securities valued the asset at 1.7 trillion won and raised Hanmi's total new-drug portfolio value to 4.9 trillion won on the back of this deal. Korean financial press called it Hanmi's largest-ever single-asset licensing agreement.

Preclinical Deals Are Getting Absurd (or Are They?)

Let's put this in context. A $2.3 billion deal for an early-stage asset sounds bananas, and it kind of is. But it's also increasingly normal in the obesity space.

The record-setter in 2025-2026 was AstraZeneca's deal with CSPC Pharma, valued at up to $18.5 billion with a $1.2 billion upfront payment, covering multiple obesity and diabetes programs. Pfizer's arrangement with Metsera was reportedly worth around $10 billion. Novo Nordisk, Regeneron, and others have all written nine- and ten-figure checks for early-stage metabolic assets.

Tracker data shows that upfront payments for outbound licensing deals roughly doubled between 2025 and 2026. Early-stage deals are now averaging upfronts of about 15% of total deal value, a number that keeps climbing. Nearly three-quarters of recent Chinese-originated licensing pacts involved preclinical or Phase 1 assets.

The market is pricing these deals like franchises, not lottery tickets. Companies are paying for the option to compete in a therapeutic category that could eventually be worth hundreds of billions of dollars.

Hanmi's Hot Streak

HM17321 isn't Hanmi's only big win this year. In June 2026, the company signed a separate licensing deal with Eli Lilly for sonefpeglutide, a GLP-2 analogue. Two major out-licensing agreements with two of the world's biggest pharma companies in the span of a few months is remarkable for a Korean drugmaker that many Western investors still haven't heard of.

Hanmi has been running what it calls H.O.P. (Hanmi Obesity Pipeline) since 2023, focused on three priorities: improving the quality of weight loss, making dosing easier, and managing patients after they've lost weight. The Genentech and Lilly deals suggest that strategy is paying off in a big way.

The Bottom Line

Genentech paid $190 million for a drug that has zero human data. That's not reckless; it's strategic. The obesity market is shaping up to be the biggest therapeutic gold rush since oncology, and Roche clearly believes that showing up with only one mechanism of action is like bringing a knife to a gunfight.

HM17321's promise of fat loss plus muscle preservation could be genuinely game-changing. But "could be" is doing a lot of work in that sentence. This molecule still needs to prove it works in people, survive years of clinical trials, and navigate a regulatory path that doesn't yet exist for its mechanism.

The check has been written. Now comes the hard part: cashing it.

ALS Just Broke Another Heart. This Time It Was Novartis.

Novartis killed its ALS drug lifonebart after a Phase 2 trial missed every endpoint, adding to a decades-long streak of industry failures against one of neuroscience's most stubborn diseases. But the company's next move tells a different story.