Dogs as a high fidelity, high throughput model to evaluate CAR-T cell function and dysfunction

Project Summary
This proposal aims to train a dual-degree, DVM-PhD student for a career as a lab animal veterinarian and independent
scientist. The research outlined in this proposal will develop a pre-clinical platform to evaluate CAR-T therapy in a canine
model. Chimeric antigen receptor (CAR)-T cells have induced up to 90% remission rates for treatment
relapsed/refractory B cell malignancies. While mice have been instrumental to CAR-T progress, CAR-T therapy for solid
tumors have been hampered by this inbred, immunodeficient model. Pet dogs are a higher fidelity translational model
due to their outbred genetics, intact immune system, high incidence of cancer, and similar cancer biology.
A CAR is a fusion protein comprised of a T cell receptor signaling domain, costimulatory domain, and an antibody based
binding domain. CARs are introduced to patients’ T cells ex vivo, enabling the T cells to directly recognize tumor antigen.
CAR-T cells are a “living therapy” wherein the efficacy of the treatment relies not only on the design of the CAR, but also
how the CAR-T cells are able to home to the tumor and elicit anti-tumor effects. The CAR helps T cells to “recognize” the
tumor, but the trafficking, persistence, and effector function of these cells relies heavily on intrinsic T cell biology. To
adequately assess CAR-T cell function in vivo, design of the CAR (Aim 1) and patient T cell biology (Aim 2) will be
evaluated.
Aim 1 – Determine optimal CAR design for targeting tumor associated antigens GD2, FolR1, and CD20. CAR constructs
will be designed for the tumor associated antigens GD2, FolR1, and CD20. These CARs will be introduced to primary
canine T cells via a lentiviral vector. The CAR-T cells will be evaluated for efficacy against antigen positive tumors by IFNγ
ELISA, IL-2 ELISA, and Incucyte live cell videomicroscopy. Each of these constructs will be tested with CAR costimulatory
domains 4-1BB and CD28. The CAR constructs with the strongest reactivity will be further evaluated with an NOD scid
gamma (NSG) mouse xenograft model, measuring tumor growth inhibition and CAR-T expansion in vivo.
Aim 2 – Determine which subset of CAR-T cells preferentially traffic and persist in the tumor in vivo. To evaluate the
respective contribution of CAR-T cell subsets to anti-tumor efficacy in vivo, semi-random nucleotide barcodes will be
added to the CAR constructs allowing for the tracking of clonal lineage during CAR-T production, infusion, and post-
engraftment in mice. Using single cell sequencing, clonal diversity of the CAR-T infusion product will be compared to
clonal diversity intratumorally. Subsets of CAR-T that preferentially home to and expand in the tumor will be identified.
Together, these aims will set the basis for future studies of CAR-T therapy in a canine model. Aim 1 will provide a
candidate CAR construct to be evaluated in a canine model. Aim 2 will provide a method for understanding how CAR-T
cells traffic to and persist within a tumor in vivo. This platform will be used to screen and refine novel approaches to
CAR-T therapy in a high-fidelity, high-throughput animal model.

Grant Number: 5F31CA265165-05
NIH Institute/Center: NIH
Principal Investigator: Samuel Brill