Characterization of a novel potency endpoint for the evaluation of immune checkpoint blockade in humanized mice

Humanized mice are emerging as valuable models to experimentally evaluate the impact of different immunotherapeutics on the human immune system. These immunodeficient mice are engrafted with human cells or tissues, that then mimic the human immune system, offering an alternative and potentially more predictive preclinical model. Immunodeficient NSG mice engrafted with human CD34+ cord blood stem cells develop human T cells educated against murine MHC. However, autoimmune graft versus host disease (GvHD), mediated by T cells, typically develops 1 year post engraftment. Here, we have used the development of GvHD in NSG mice, using donors with HLA alleles predisposed to autoimmunity (psoriasis) to weight in favor of GvHD, as an endpoint to evaluate the relative potency of monoclonal and BiSpecific antibodies targeting PD-1 and CTLA-4 to break immune tolerance. We found that treatment with either a combination of anti-PD-1 & anti-CTLA-4 mAbs or a quadrivalent anti-PD-1/CTLA-4 BiSpecific (MEDI8500), had enhanced potency compared to treatment with anti-PD-1 or anti-CTLA-4 monotherapies, increasing T cell activity both and . This resulted in accelerated development of GvHD and shorter survival of the humanized mice in these treatment groups commensurate with their on target activity. Our findings demonstrate the potential of humanized mouse models for preclinical evaluation of different immunotherapies and combinations, using acceleration of GvHD development as a surrogate of aggravated antigenic T-cell response against host.