587 Tumor-activated Fc-engineered anti-CTLA-4 monoclonal antibody, XTX101, demonstrates tumor-selective PD and efficacy in preclinical models. (10th December 2020)

• Record Type: Journal Article
• Authors: Jenkins, Kurt; Johnson, Parker; Zhang, Minjie; Guzman, Wilson; Eskiocak, Ugur; McLaughlin, Megan; O'Toole, Caitlin; Pederzoli-Ribeil, Magali; Park, Miso; Williams, John; Karow, Margaret; O'Neil, Jennifer; Clackson, Timothy; O'Hagan, Ronan
• Is Part Of: Journal for immunotherapy of cancer. Volume 8(2020)Supplement 3
• Page Start: A351
• Page End: A351
• Publication Date: 2020-12-10
• Subjects: Cancer -- Immunotherapy -- Periodicals; Cancer -- Immunological aspects -- Periodicals; Tumors -- Immunological aspects -- Periodicals; Immunotherapy -- Periodicals; 616.99406105
• DOI: 10.1136/jitc-2020-SITC2020.0587
• Languages: English
• ISSNs: 2051-1426

Abstract

Abstract : Background: The clinical benefit of CTLA-4 blockade to cancer patients has been well established. However, the promising antitumor activity shown by anti-CTLA-4 monoclonal antibodies (mAb) has been limited by the occurrence of immune-mediated adverse reactions, especially when CTLA-4 inhibition is used in combination with anti-PD-1 therapy. These dose-limiting toxicities restrict the therapeutic use of CTLA-4 blockade. To overcome these limitations, we have developed a potent anti-CTLA-4 antibody that is selectively active in the tumor microenvironment (TME). This antibody is engineered with an Fc region for enhanced FcγR binding and peptides that mask antigen-binding regions. The masking peptides are designed to be selectively cleaved and released by proteases that are more active in the TME, resulting in restoration of full activity of the antibody in the TME. Methods: A novel, fully-humanized anti-huCTLA-4 mAb was shown to bind human CTLA-4 with improved affinity compared to ipilimumab, as measured by SPR. Engineering of the Fc region enhanced FcγR binding and ADCC function. In addition, CDR-binding peptides identified by phage display were covalently linked to the antibody using a protease-sensitive polypeptide linker. This engineered anti-CTLA-4 antibody (XTX101) showed protease-dependent binding to CTLA-4 both with recombinant and tumor tissue derived proteases. Results: XTX101 demonstrated a 100-fold reduction in binding to human CTLA-4 by ELISA, compared to the non-masked antibody. Incubation with recombinant protease led to cleavage and release of the masking peptides and restored full binding to CTLA-4. Similarly, in vitro ADCC activity was impaired by masking and restored in a protease-dependent manner. SEB-stimulated human PMBCs were minimally responsive in vitro to XTX101, whereas PBMCs treated with proteolytically-activated XTX101 exhibited robust activation of T cell function. In human CTLA-4 knock-in mice with syngeneic MB49 tumors, XTX101 treatment led to complete tumor regression, enhanced CD8+ T cell proliferation, and depletion of tumor Tregs in the TME. By contrast, XTX101 had minimal pharmacodynamic effects in the periphery. In addition, XTX101 is effectively activated in culture supernatants from human solid tumor explants obtained from a broad range of tumor types. Conclusions: XTX101 is a tumor-selective anti-CTLA-4 mAb capable of: 1) effective CTLA-4 blockade, 2) depletion of intratumoral Tregs through enhanced antibody-dependent cellular cytotoxicity (ADCC) function, 3) minimization of systemic immune cell activation, and 4) potent anti-tumor activity. These pre-clinical data support the further evaluation of XTX101 in clinical studies.