AUTHOR=Kanikarla Marie Preeti , Sorokin Alexey V. , Bitner Lea A. , Aden Rebecca , Lam Michael , Manyam Ganiraju , Woods Melanie N. , Anderson Amanda , Capasso Anna , Fowlkes Natalie , Overman Michael J. , Menter David G. , Kopetz Scott TITLE=Autologous humanized mouse models to study combination and single-agent immunotherapy for colorectal cancer patient-derived xenografts JOURNAL=Frontiers in Oncology VOLUME=12 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.994333 DOI=10.3389/fonc.2022.994333 ISSN=2234-943X ABSTRACT=

Designing studies of immunotherapy is limited due to a lack of pre-clinical models that reliably predict effective immunotherapy responses. To address this gap, we developed humanized mouse models of colorectal cancer (CRC) incorporating patient-derived xenografts (PDX) with human peripheral blood mononuclear cells (PBMC). Humanized mice with CRC PDXs were generated via engraftment of autologous (isolated from the same patients as the PDXs) or allogeneic (isolated from healthy donors) PBMCs. Human T cells were detected in mouse blood, tissues, and infiltrated the implanted PDXs. The inclusion of anti-PD-1 therapy revealed that tumor responses in autologous but not allogeneic models were more comparable to that of patients. An overall non-specific graft-vs-tumor effect occurred in allogeneic models and negatively correlated with that seen in patients. In contrast, autologous humanized mice more accurately correlated with treatment outcomes by engaging pre-existing tumor specific T-cell populations. As autologous T cells appear to be the major drivers of tumor response thus, autologous humanized mice may serve as models at predicting treatment outcomes in pre-clinical settings for therapies reliant on pre-existing tumor specific T-cell populations.