LILRB4 sounds like a bad rapper name.
It’s also a protein that is heavily expressed on monocytic AML blasts, on chronic myelomonocytic leukemia cells, and on certain multiple myeloma populations. It’s minimally expressed on normal tissue. And it actively suppresses T cell function, helping these cancers hide from the immune system.
Hit it right and you do two things at once: kill the tumor and remove an immune brake.
The problem?
Nobody had been able to file an IND on a LILRB4 T cell engager. Until now.
China’s NMPA just accepted an IND application for Mabwell’s 6MW5311, a LILRB4/CD3 bispecific T cell engager. Mabwell is calling it a global first, and the regulatory record backs that up. There are LILRB4 antibodies in development (Immune-Onc has IO-202 in Phase 1), but a bispecific TCE format hitting both LILRB4 and CD3 hasn’t been here before.
6MW5311 mechanism: 2+1 asymmetric design
Tumor cell
LILRB4 binding (×2)
T cell
CD3 binding (×1)
Steric hindrance design lowers off-target T cell activation
The “2+1” design is doing real work. Two LILRB4 binding arms increase tumor avidity, while a single CD3 binder reduces the chance of T cells getting activated where they shouldn’t be. The steric hindrance built into the format is meant to keep cytokine release syndrome in check, the classic problem that has dogged TCEs since blinatumomab.
This kind of asymmetric design is becoming the standard playbook for next-gen TCEs. It’s the same direction Deck Bio is pushing with its multi-target intracellular TCE platform, and it’s what Janux Therapeutics built its entire masked-TCE business on.
The targets Mabwell is going after, AML, CMML, and multiple myeloma, are all populations where LILRB4 expression is high and treatment options are thin. Relapsed/refractory AML especially is a graveyard. Median OS measured in months.
The few approved options (venetoclax + azacitidine, gilteritinib for FLT3-mutated) help some patients but most still die of disease.
The Biotech Voyager
Early-stage biotech signals, personalized.
The signals that matter to you, contextualized and written directly to you, so you cut through the noise and immediately understand why it matters.
Mabwell is also planning a U.S. IND for 6MW5311 in the second half of 2026. That timing matters. China-first IND filings used to mean “we’ll get there eventually” for U.S. patients. The pace has shifted. Companies like Akeso and Innovent have shown that Chinese-originated assets can move into U.S. clinical development quickly when the data are clean.
The competitive setup in LILRB4 is going to be interesting to watch.
Immune-Onc has been developing LILRB-targeted antibodies for years, with IO-202 (anti-LILRB4 monoclonal) in clinical trials and a partnership with AbbVie behind it.
BioAtla has a CAB-LILRB4 program. So Mabwell isn’t alone in the target, but they’re first to a TCE format.
The early Phase 1 data, when it comes, will tell us whether the steric hindrance design actually delivers a clean safety profile. If it does, this could become a template for other “difficult” leukemia targets where naked antibodies haven’t been enough.
For now: first IND in the world for a LILRB4/CD3 bispecific. Worth tracking.
