

Roche just licensed a preclinical trispecific antibody from China's Simcere Zaiming for up to $1.53 billion, and the drug hasn't even been tested in humans yet. The deal reveals how far big pharma will go to win the next-generation blood cancer arms race.
Most drugs that attract billion-dollar deals have at least been tested in humans. SIM-0660 hasn't. It's still in the preclinical stage, meaning it hasn't entered a single clinical trial. And yet Roche just wrote a check to lock it up.
The Swiss pharma giant licensed the trispecific antibody from China's Simcere Zaiming Pharmaceutical in a deal worth up to $1.53 billion. That's a massive bet on a molecule that, at this point, has only been tested in labs and animal models. So what exactly did Roche see that made it open the vault?
To understand the hype, you need to understand what SIM-0660 actually does. And to understand that, think of it like a bouncer at a club who can check three different IDs simultaneously.
Most cancer-fighting antibodies in this class are "bispecific," meaning they grab onto two things at once: a marker on the cancer cell and a marker on an immune cell called a T cell. It's like introducing a predator to its prey. The T cell shows up, recognizes the cancer cell, and kills it.
SIM-0660 is trispecific. It grabs onto three things: CD19 and CD79a on B cells (a type of immune cell that goes rogue in blood cancers), plus CD3 on T cells. By latching onto two different markers on the cancer cell instead of one, the drug makes it much harder for cancer to slip away. Cancer cells are notorious escape artists; they can stop expressing a single marker to dodge treatment. Hitting two markers at once is like locking both the front door and the back door.
The other piece Simcere is touting: the molecule was designed to limit cytokine release, a dangerous inflammatory overreaction that plagues many T-cell engagers and can land patients in the ICU. If that holds up in humans, it would be a meaningful safety edge.
The deal structure tells its own story. Roche is paying $75 million upfront, with the remaining ~$1.46 billion tied to development, regulatory, and commercial milestones. Roche also owes tiered royalties described as percentages on future sales.

AstraZeneca is reportedly negotiating a deal worth up to $15 billion with Summit Therapeutics for ivonescimab, a bispecific antibody that beat Keytruda in a head-to-head lung cancer trial. If it closes, it would be one of the largest licensing deals in biopharma history.


Join thousands of biotech professionals who start their day with our free, daily briefing.
In pharma licensing, the upfront payment is the part the buyer actually commits to. The headline number ($1.53 billion) represents what Simcere could earn if everything goes perfectly: successful trials, regulatory approvals, blockbuster sales. Think of it like a real estate deal where you put down a deposit, and the rest depends on the house passing inspection, getting rezoned, and appreciating in value.
That $75 million upfront is modest by big pharma standards, but for a preclinical asset, it's a strong signal of conviction. It tells you Roche believes the science is real enough to justify locking in global rights before anyone else can.
Roche isn't starting from scratch in B-cell cancers. Far from it. The company already sells Columvi (glofitamab) and Lunsumio (mosunetuzumab), both bispecific antibodies targeting CD20 and CD3. Columvi recently won European approval for relapsed or refractory diffuse large B-cell lymphoma, and Roche is pushing it into earlier treatment lines and new combinations.
But both of those drugs are bispecifics. They hit one tumor target (CD20) and one T-cell target (CD3). SIM-0660 represents a generational leap: a trispecific that could address the antigen-escape problem that limits current bispecifics. It also targets different B-cell markers (CD19 and CD79a instead of CD20), which could open the door to patients who've already failed CD20-directed therapies.
Roche is clearly building a multi-asset hematology platform. It has Columvi and Lunsumio for today, cevostamab (an FcRH5×CD3 T-cell engager) and allogeneic CAR-T programs in development, and now SIM-0660 for the next generation. The strategy resembles a baseball team stacking its farm system: you need prospects at every level to stay competitive.
Roche isn't the only one shopping for trispecifics. The deal lands in the middle of a full-blown arms race among big pharma companies racing to lock up next-generation multispecific antibodies.
AbbVie paid $700 million upfront for ABBV-2001, a BCMA×CD38×CD3 trispecific for multiple myeloma, from Ichnos/IGM. Johnson & Johnson is developing its own trispecific (ramantamig) targeting BCMA, GPRC5D, and CD3. GSK inked a deal worth up to $750 million with Chimagen for a next-gen T-cell engager in myeloma.
Most of these programs are focused on multiple myeloma. What makes the Roche/Simcere deal interesting is that SIM-0660 targets B-cell malignancies more broadly, not just myeloma. That's a different competitive lane, and it explains why Roche sees differentiation rather than a crowded field.
Simcere Zaiming is the oncology-focused subsidiary of Simcere Pharmaceutical Group (HKEX: 2096), a Chinese pharma company that traces back to 1995. The subsidiary was formally established in the early 2020s to develop cancer therapies for global markets.
Its secret weapon is a polyspecific antibody platform that generates trispecific molecules with what the company describes as a low-affinity, target-activated CD3 arm. Translation: the T-cell engagement only kicks into high gear when the molecule is already bound to a cancer cell, which should reduce the kind of off-target immune activation that causes serious side effects.
The Roche deal isn't Simcere Zaiming's first Western partnership. The company has also announced collaborations with AbbVie and NextCure, suggesting its platform has broad appeal. Simcere's stock reportedly rose after the announcement, as investors priced in both the deal economics and the validation of the underlying technology.
The real test is whether SIM-0660 can do in humans what it does in the lab. Preclinical promise is cheap; clinical success is rare. Roughly 90% of drugs that enter clinical trials never make it to approval, and that stat is even grimmer for drugs that haven't entered trials yet.
But Roche structured the deal to manage that risk. The heavy milestone-loaded structure means most of the money only flows if the drug actually works. And Roche's deep expertise in B-cell biology (from decades of Rituxan, Gazyva, Columvi, and Lunsumio experience) gives it a better-than-average shot at shepherding a tricky molecule through development.
For Roche, this is a calculated bet that trispecific antibodies are the future of blood cancer treatment. For Simcere Zaiming, it's proof that a Chinese biotech subsidiary can build platform technology that the world's biggest pharma companies will pay a premium to access. For patients with B-cell cancers, it's a reason to watch this space closely.
The bouncer just got a third ID to check. Now we wait to see if it actually works at the door.
AstraZeneca's oral breast cancer pill camizestrant missed its primary endpoint in the pivotal SERENA-4 trial, failing to prove a statistically significant benefit in first-line advanced breast cancer. The setback narrows the drug's commercial ceiling just as Eli Lilly's competing therapy racks up its second FDA approval.