Adoptive T cell immunotherapy consists of ex vivo genetic modification of a patient’s own T cells to target and eradicate cancer cells expressing specific antigens. Pre-clinical experiments are usually performed in mice models to validate the efficiency, specificity and safety of these modified T cells in vivo. However, to avoid graft-versus-host-disease (GVHD, which happens when donor T cells recognize and kill the host cells) or rejection by the mouse immune system, validation studies have been performed using immunodeficient mice.
Drs. Kevin Haworth and Hans-Peter Kiem (Clinical Research Division) developed a humanized mouse model more relevant for pre-clinical evaluation of adoptive T cells. The results of this study were recently published in the journal Molecular Therapy: Methods & Clinical Development.
In this mouse model system, irradiated mice are infused with human CD34+ hematopoietic stem and progenitor cells (HSPCs). This allows the development of a human immune system in an in vivo mouse environment. From this model, the researchers isolated and reinfused modified T cells following ex vivo culture. This mouse model was named MATCH: in vivo murine-matured autologous CD3+ T cells from humans. As explained by Dr. Haworth, “this MATCH mouse model enables the use of engineered receptors designed to eliminate specific target cells such as tumor cells or virally infected cells by testing their functionality in the background of a human immune system. Previously, such studies have been performed by simple co-injection of target and effector cells into an immunodeficient mouse”.
The extent of “humanization” was assessed by quantification of the percentage of circulating human versus mouse CD45+ cells, a pan-hematopoietic marker (includes all blood lineages), in the blood over time but also engrafted in tissues at the end of the experiment. Further evaluation of the human cell lineages composition was performed by CD3, CD4, CD8 (T cells markers), CD20 (B cell marker) and CD14 (monocytes/macrophages markers) quantification. The average observed engraftment was 10-20% of human CD45+ cells persisting for up to 34 weeks, and included T cells, B cells as well as monocytes and macrophages.
To determine whether the in vivo matured human cells would tolerate autologous (same donor) CD3 T cells, the humanized mice were infused with CD3+ T cells that were previously isolated from the same human donor. To facilitate in vivo monitoring, these T cells were previously modified ex vivo using lentiviruses to express green fluorescent protein (GFP), a fluorochrome allowing tracking and quantification of the T cells by flow cytometry. However, the mice quickly showed signs of GVHD and the CD3+GFP+ cells did not persist.
The experiment was then replicated using in vivo murine-matured CD3 T cells isolated from the humanized mice instead. As such, in vivo matured CD3 T cells should recognize mice cells as self, eluding GVHD. Comparing MATCH cells to autologous human T cells, the authors did not observe any difference in transduction efficiency or proliferation. MATCH cells secreted interferon gamma cytokine and underwent TCR rearrangement upon ex vivo stimulation similarly to autologous human T cells validating their functionality. The naïve, memory and effector T cells composition was also similar.
Infused MATCH cells were 66% to 73% GFP positive on the day of the infusion. Peripheral blood composition indicated that the GFP+MATCH modified cells represented between 1.1% and 7.4% of the CD3+ cells population and persisted for more than 7 weeks. Tissue composition was analyzed at 8 weeks and indicated that the GFP+MATCH cells efficiently migrated to other tissues such as bone marrow, spleen and lungs.
The clonal diversity of the persistently engrafted MATCH was assessed by identification of unique integration sites in the lentivirally-modified T cells allowing for the identification and tracking of unique clones in vivo. Interestingly, the autologous human T cells underwent significantly higher clonal expansion than the in vivo matured human CD3 T cells, probably associated with the GVHD onset observed in these mice.
Development of the MATCH model is a stepping-stone for pre-clinical platforms with the potential to support further evaluation of new adoptive T cell immunotherapies and mirror the existing clinical protocols. Dr. Haworth further developed, “we are currently in the process of validating several engineered receptors against different types of lymphomas in collaboration with other Hutch researchers. This work will then provide a platform to compare new therapeutic products in the hopes of identifying which ones are strong candidates to advance to the clinic. We eventually would like to branch out and use the model for other types of diffuse blood cancers, and perhaps even some solid tumor models as well”.
This study was supported by the National Institutes of Health.
Haworth KG,Ironside C,Norgaard ZK,Obenza WM,Adair JE,Kiem H-P. In Vivo Murine-Matured Human CD3+ Cells as a Preclinical Model for T Cell-Based Immunotherapies. 2017. Molecular Therapy: Methods & Clinical Development, 6, 17-30.