Background Immune checkpoint inhibitors (ICIs) such as anti-PD-1/PD-L1 and anti-CTLA-4 antibodies have achieved enormous success in cancer treatment. However, only a subset of patients shows clinical responses. The combination of CTLA-4 and PD-1 blockers have increased response rates in patients, however severe immune related adverse events (irAEs) have limited their use. TIGIT is another inhibitory receptor, and often co-expressed with PD-1 on tumor-infiltrating CD8+ cytotoxic T cells. Several anti-TIGIT blocking antibodies have shown encouraging anti-tumor activity in clinical studies when combined with PD-1/PD-L1 blockers. To improve clinical efficacy and more profoundly promote anti-tumor immunity, we have generated a novel tri-specific antibody, GBD209, targeting PD-1/CTLA-4/TIGIT and by fine tuning the potency of each arm.

Methods Specific nanobodies against PD-1, CTLA-4, and TIGIT were generated by immunization of alpaca. The tri-specific antibody was designed with a small size of ~125KD for good penetration in solid tumors. Various of formats were screened by in vitro and in vivo assays to balance the binding/blocking activity of the three targets. Fc functions were silenced by mutations in Fc region (L235A/L236A/G238A). In vitro characterization included affinity, binding/ligand blocking, cell reporter activity, functional activity in mixed lymphocyte reaction (MLR). In vivo efficacy was evaluated in PBMC-humanized CDX models.

Results After multiple rounds of lead screening and construction of the tri-specific antibody, the final molecule, GBD209, exhibited potent binding to PD-1- or TIGIT-expressing cells and fully blocked PD-1 and TIGIT pathways in PD-1/PD-L1 or TIGIT/CD155 reporter systems. To reduce the toxicity associated with CTLA-4 inhibition, a unique CTLA-4 partial blocker was used for GBD209. The anti-CTLA-4 arm only partially blocks the binding of CTLA-4 to its ligands, resulting in better in vivo tolerability than ipilimumab. The potency of CTLA-4 inhibition was further moderated by the structural design of the tri-specific antibody. GBD209 bound strongly to PD-1⁺TIGIT⁺CTLA-4⁺ cells (EC50 = 4nM), whereas the binding of GBD209 to CTLA-4 single positive cells was much weaker (EC50=78 nM). The different binding property is intended to reduce the impacts on peripheral CTLA-4⁺ T cells. By in vivo studies using PBMC-humanized A375 model, GBD209 showed superior anti-tumor efficacy compared to bsAb benchmarks that target PD-1/CTLA-4.

Conclusions GBD209 is a first-in-class anti-PD-1/CTLA-4/TIGIT tri-specific antibody that blocks all three checkpoints. GBD209 may profoundly provoke anti-tumor immunity and overcome PD-1 resistance in patients. The tri-specific modality and unique design of GBD209 may represent a novel and effective therapeutics for cancer treatment.