Regulatory T cells (Treg) are key players of the suppressive tumour microenvironment (TME). Their presence is correlated with a bad prognosis in multiple cancers, while a greater ratio of effector T cells (Teff) to Treg is associated with improved outcome. Studies demonstrating high expression of CD25 on Tregs but not Teff in human tumors have underscored its relevance as a target for Treg depletion. However, clinical trials conducted to target Treg in cancer patients with existing anti-CD25 (aCD25) antibodies have shown contradicting results, and no aCD25 antibody is in clinical development for this application. Importantly, aCD25 antibodies tested to date block IL2 signalling via CD25. We demonstrated (Solomon et al. AACR2018) that such blocking drastically reduces the therapeutic activity of aCD25 antibodies and that a single administration of a depleting antibody targeting CD25 but not blocking the IL2 signalling effectively depletes Treg, increases Teff activity and promotes eradication of established tumors in several mouse models of cancer.

With this rationale in mind, we have generated a panel of fully human aCD25 IgG1 antibodies, which we characterized for their binding to human and cynomolgus monkey CD25. The antibodies were then screened for their impact on IL2 binding, using a sandwich binding assay on Octet, and on IL2 signalling, using a STAT5 phosphorylation assay. Antibodies were also tested for their ability to deplete CD25 positive cell lines and in vitro-derived human Treg in ADCC and ADCP assays. Finally, the impact of selected antibodies on Treg within the TME and on Teff responses was evaluated in human samples.

Among the antibodies binding to human and cynomolgus CD25, we have selected a panel interfering with neither IL2 binding to CD25 nor IL2 signalling. Interestingly, epitope binning assays demonstrated that none of the selected antibodies bind to epitopes overlapping with those of the existing clinical antibodies, Daclizumab and Basiliximab. Among the non-IL2 blocking antibodies, those showing the maximum target cell lysis via ADCC and highest phagocytosis via ADCP represent potential lead clinical candidates.

Most interestingly, contrary to the existing clinical aCD25 antibodies, which block IL2 signalling, our candidates do not inhibit Teff proliferation and function, confirming the specificity of action toward Treg and the importance of preserving IL2 signalling. Finally, we have shown that our clinical candidates efficiently deplete Treg in the TME.

We present the first anti-human CD25 antibodies selected for their capacity to deplete human Treg while preserving IL2 signalling and activity of Teff. These antibodies provide a novel therapeutic approach to alleviate immune suppression in the TME, which would provide an ideal combination partner for existing standard of care and IO treatments, and could potentially be used as a monotherapy.

Citation Format: Josephine C. Salimu, Mark Brown, Pascal Merchiers, Beatriz Goyenechea, Kevin Moulder, Robert Dejonge, Aghiles Boughetane, Isabelle Solomon, Frederick Arce Vargas, Karl S. Peggs, Anne Goubier, Sergio A. Quezada. Generation of first-in-class anti-CD25 antibodies depleting Treg without interfering with IL2 signalling for cancer therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2787.