Authors:

Di Sun¹, Huanhuan Guo¹, Hongjun Li¹, Xinlai Yao¹, Yiqing Xu1, Shenjie Zhang1, Shijie Zhou1, Xiaoli Zi1, Lisa Zheng1, Xueming Qian1, Yi Gu1

¹Suzhou Transcenta Therapeutics Co., Limited, Suzhou

Abstract:

Regulatory T (Treg) cells are a distinctive lineage of CD4+ T cells characterized by the expression of FoxP3 and CD25 markers. Treg cells are involved in tumor development and progression by inhibiting anti-tumor immunity. Accumulation of Treg cells in the tumor microenvironment (TME) is often associated with an adverse prognosis in multiple cancers and also has implication for predicting response to immunotherapy[1]. Treg cell depletion strategy has been explored in preclinical and clinical settings to induce effective anti-tumor immune responses in TME. CD25, a high-affinity binding subunit alpha of the IL-2 receptor, is constitutively and highly expressed on Treg cells but also transiently upregulated on effector T cells. The high level of CD25 expression on Treg cells could deprive of IL-2 for effector T cells and inhibit their proliferation. Thus, an effective approach to selectively suppress Treg cell function is to deplete Treg cells while avoid blocking IL-2 binding to CD25 for allowing T effect cell proliferation and activation.

Using our immune tolerance breaking antibody platform, we generated a humanized IgG1 subtype anti-CD25 monoclonal antibody with reduced fucose (TST010). TST010 bound to human CD25 with high affinity and selectivity but did not block IL-2 binding to CD25. Non-blocking activity to IL-2 signaling on human PBMC was further confirmed by the phosphorylation of STAT5 using FACS analysis. By reducing fucosylation during cell culture process, TST010 mAb gained the enhanced ADCC activity. In an ADCC reporter assay, TST010 showed its sub-nanomolar ADCC activity against CD25-overexpression SU-DHL-1 cell line and induced human Treg cells, but much less to activated human CD8+ T effect cells due to lower CD25 expression. Similarly, in a NK cytotoxicity assay, TST010 selectively killed induced human Treg cells but not activated CD8+ T cells. In vivo anti-tumor activity of TST010 was evaluated in the syngeneic mouse tumor models. The mouse colon cancer MC38 model was developed in the human CD25 knock-in mice. In this model, TST010 markedly reduced the Treg (FoxP3+CD4+) population in both peripheral blood and TME. Importantly, the CD8+T/Treg ratio in tumor was increased starting at 2 weeks post TST010 treatment, and it resulted in a significant single agent tumor growth inhibition at 10 mg/kg. Furthermore, in the checkpoint inhibitor refractory LLC tumor model, the anti-CD25 mAb showed a good combination effect with a PD-(L)1 inhibitor (data not shown).

In summary, we have discovered a novel therapeutic anti-CD25 mAb that can deplete Treg cells without blocking IL-2 signaling. As demonstrated in our preclinical tumor models, it has a good potential to induce effective anti-tumor immune responses in TME and tumor growth inhibition especially in combination with PD-1/PD-L1 inhibitor.

Conslusions:

•TST010, a humanized IgG1 anti-CD25 antibody with low fucose level, binds specifically to CD25;
•High affinity binding to CD25 through affinity maturation;
•Potent ADCC activity selectively to iTreg, but not  activated CD8+T cells in vitro;
•Good anti-tumor activity in MC38 syngeneic model;
•Increased CD8+T population and CD8+T/Treg ratio via Treg depletion in vivo;
•Potential for combination with checkpoint inhibitor