Abstract Background: The limited efficacy of immune checkpoint inhibitors (ICIs) due to resistance remains an unmet medical need, making PD-1/IL-2 a promising therapeutic approach. Many strategies focus on eliminating IL-2's interaction with IL-2Rα to avoid Treg activation. However, such "not-α" IL-2 mutein-based fusion proteins have demonstrated limited clinical success. In contrast, PD-1/IL-2α-bias approach has shown promising efficacy in patients with ICI-resistant and immunologically "cold" tumors. TGI-17a is a potential best-in-class PD-1/IL-2α-bias bispecific antibody designed to specifically activate antigen-specific CD8+T cells by synergistically blocking inhibitory PD-1 signaling while enhancing activating IL-2 signaling. It is engineered with an IL-2 variant exhibiting a stronger bias towards IL-2Rα, a strategy anticipated to enhance anti-tumor activity while minimizing IL-2Rβγ-mediated toxicity. Methods: The binding affinity of TGI-17a for PD-1, IL-2Rα, and IL-2Rβγ was evaluated by Surface Plasmon Resonance (SPR). Its in vitro activity was assessed by measuring STAT5 phosphorylation (pSTAT5) in T cells via flow cytometry. Toxicity of the IL-2 variant was evaluated by intravenous administration of IL-2 variants fused to a surrogate anti-mouse PD-1 antibody in wild-type mice. The anti-tumor efficacy of TGI-17a was investigated in MC38 and B16F10 syngeneic tumor models using humanized PD-1 transgenic mice. Results: TGI-17a exhibited intermediate affinity for IL-2Rα (KD = 19.2 nM), while its affinity for IL-2Rβγ was over 50-fold lower (KD = 1060 nM), rendering it a more profoundly IL-2Rα-biased molecule compared to the competitor. In functional assays, TGI-17a did not induce pSTAT5 in peripheral T cells even at high concentrations (up to 250 µg/mL), but potently activated PD-1+IL-2Rα+ T cells. Unlike the competitor, whose function was somehow independent of PD-1 binding, TGI-17a's activity was strictly PD-1-dependent (10% pSTAT5 induction: 1.3 µg/mL for TGI-17a vs. 82 µg/mL for IL-2v-Fc). In toxicity studies, TGI-17a was well-tolerated at doses up to 40 mg/kg (IV, twice weekly) in mice, whereas the competitor caused lethality at 5 mg/kg.In the in vivo efficacy studies, TGI-17a demonstrated superior anti-tumor activity compared to the PD-1 antibody pembrolizumab in both MC38 and B16F10 models. It showed comparable efficacy to the competitor in the MC38 model and superior efficacy in the B16F10 model. Furthermore, TGI-17a exhibited significantly better efficacy than a PD-1/VEGF bispecific antibody in the MC38 model. Conclusions: These results indicate that TGI-17a, a potential best-in-class PD-1/IL-2α-bias bispecific antibody fusion protein, possesses compelling anti-tumor efficacy and a favorable safety profile in preclinical models. This robust evidence supports its further clinical development as a promising therapeutic candidate for pan-cancer applications. Citation Format: Guoshuai Cao, Yangyang Li, Yuwei Wu, Haoyu Sun, Zhigang Tian. TGI-17a: A potential best-in-class PD-1/IL-2α-biascytokine fusion protein abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 4336.
Cao et al. (2026) studied this question.