Protocol to generate a patient derived xenograft model of acquired resistance to immunotherapy in humanized mice.

Immunotherapy has revolutionized cancer treatment, but preclinical models are required to understand immunotherapy resistance mechanisms underlying patient relapse. This protocol describes how to generate an acquired resistance humanized in vivo model to immunotherapies in patient-derived xenografts (PDX). We detail steps to inject human CD34+ cells into NSG mice, followed by generation of immunoresistant PDX in humanized mice. This approach recapitulates the human immune system, allowing investigators to generate preclinical resistance models to different immunotherapies for identifying the resistant phenotype. For complete details on the use and execution of this protocol, please refer to Martínez-Sabadell et al., 2022 and Arenas et al. (2021).

The target antigen determines the mechanism of acquired resistance to T cell-based therapies.

Despite the revolution of immunotherapy in cancer treatment, patients eventually progress due to the emergence of resistance. In this scenario, the selection of the tumor antigen can be decisive in the success of the clinical response. T cell bispecific antibodies (TCBs) are engineered molecules that include binding sites to the T cell receptor and to a tumor antigen. Using gastric CEA+/HER2+ MKN45 cells and TCBs directed against CEA or HER2, we show that the mechanism of resistance to a TCB is dependent on the tumor antigen. Acquired resistant models to a high-affinity-CEA-targeted TCB exhibit a reduction of CEA levels due to transcriptional silencing, which is reversible upon 5-AZA treatment. In contrast, a HER2-TCB resistant model maintains HER2 levels and exhibit a disruption of the interferon-gamma signaling. These results will help in the design of combinatorial strategies to increase the efficacy of cancer immunotherapies and to anticipate and overcome resistances.