Dan Gomari, Founder, CEO of Quinn – AI Agents for drug R&D, shared on LinkedIn:
“GSK and Roche’s B7-H3 drugs each cut small-cell lung death risk 54%. Amgen, Merck and Daiichi Sankyo now race both targets into first line.
The drugs beat topotecan in 461 and 451 relapsed patients. Both carry chemotherapy straight to B7-H3 on tumor cells. BioNTech’s B7-H3 drug plus pumitamig, a PD-L1 and VEGF antibody co-developed with BMS, shrank tumors in 70.4% of 71 patients, including 7 of 8 after a DLL3 drug. That is standout activity.
Upfront, adding Amgen’s Imdelltra to AstraZeneca’s Imfinzi in DeLLphi-305 cut maintenance death risk 58%. Imdelltra steers T cells to DLL3 on tumor cells. Boehringer’s DLL3 drug is in first-line Phase 3. GSK just paired its B7-H3 drug with ivonescimab, and Merck and Daiichi Sankyo test theirs upfront.
Every small-cell lung team faces one question. Do B7-H3 and DLL3 drugs chase the same cells, or cover each other’s escape?
We gave that question to Quinn, our team of AI agents for drug R&D. It analyzed 81 tumors from a Nature cohort and reconstructed DeLLphi-305’s 563 patients. Tumor RNA is not the protein a drug binds.
DLL3 and B7-H3 moved independently. DLL3 tracked neuroendocrine genes and dropped in YAP1-type tumors, where B7-H3 rose. B7-H3 stayed on in all 21 tumors lowest in DLL3. In DeLLphi-305, saving Imdelltra for relapse fell short even with perfect second-line survival, because many declined or died first.
A second target is a partner, not a backup. As DLL3 drugs move upfront, does your drug hit the cells that escape them? It applies to Amgen, AstraZeneca, Boehringer, BMS, BioNTech, Daiichi Sankyo, GSK, Merck and Roche.
Saving a drug for relapse can mean patients never get it. Quinn can tell you whether your combination, sequence, or patient selection holds up against tumor escape and rival trials, in days, not months.”

You can also read: After TROP2 and HER2, Is B7-H3 the Next Big ADC Target?

