Clinical data demonstrates that one chimeric antigen receptor (CAR) design targeting T cells to a given tumor-associated antigen (TAA) will have varying therapeutic potential in different patients. For example, second generation CD19-specific CARs activated via chimeric CD28/CD3ζ or CD137/CD3ζ exhibit superior clinical responses when autologous genetically modified T cells are administered to patients with acute versus chronic B-lineage leukemias. Therefore, we investigated whether CAR species can be designed to improve the anti-tumor effect for a given tumor. This is particularly appealing as our clinical approach to gene therapy based on the electro-transfer of DNA plasmids from the Sleeping Beauty (SB) transposon system reduces the cost and complexity to manufacture individual CAR designs for small subsets of patients. Our approach to personalizing CAR+ T cells is based on the generation of a large number of CAR molecules that can be screened and assessed for their ability to benefit a given patient. We have developed an approach for high throughput assembly of CARs using triple site-specific recombination system termed “EZ-CAR platform”. This technology allows the rapid combination of 3 components of a prototypical CAR from (i) the single chain variable fragment (scFv) that defines specificity, (ii) the scaffold/hinge that appends the scFv from the cell surface, and (iii) one or more intracellular signaling domains. As proof-of-concept we generated a CD19-specific CAR using the EZ-CAR platform in parallel with CD19 CAR control CAR+ T cells (CG CAR). Both CD19 CAR sequences were inserted into Sleeping Beauty transposon vectors and electroporated into T cells. After electroporation the T cells were cultivated in presence of CD19+ Activating and Propagating Cells (AaPCs) for antigen specific expansion of the T cells. The expression of the CARs in the T cells surface was measured every week by flow cytometry (Fc+ expression), showing similar CAR expression in both groups. We also performed a Chromium Release Assay (CRA) to evaluate the killing function of CD19 CAR+ T cells generated by EZ-CAR platform against tumor cells. After 4 hours of incubation the percentage of specific cell lysis was 52% by the EZ- CAR T cells and 49% by the CG CAR T cells. These results showed that using this technology we were able to generate functional CAR+ T cells. Then we started a rapid production of CARs using a library of plasmids containing the three components of a CAR molecule: (i) anti-CD19 scFv (ii) 5 hinges with different sizes and (iii) different combinations of 7 signaling domains with the CD3ζ domain. The EZ-CAR platform is in a dynamic process where new CAR components are been added to our library. Transfection of HEK 293 cells with plasmid containing the CAR transgene were used to screen 30 different CAR constructs to ensure the expression of the CAR protein on the cell surface. The high throughput testing of individual CAR molecules has been undertaken using the iQue Screener (Intellicyt), a high throughput flow cytometer, where cytotoxic assays are performed using engineered target cells expressing a fluorescent granzyme B reporter. The release of IL-2, IFN-γ and TNF-α by the CAR+ T cells and the cytotoxic activity are concurrently evaluated. The CAR designs that result in superior killing will be further tested using tumor cells from patients.
Citation Format: Ana Beatriz Korngold, Brian Rabinovich, Helen Huls, Laurence Cooper. Designing chimeric antigen receptors for personalized immunotherapy: Rapid assembly of CARs from principal components using “EZ-CAR” platform. [abstract]. In: Proceedings of the CRI-CIMT-EATI-AACR Inaugural International Cancer Immunotherapy Conference: Translating Science into Survival; September 16-19, 2015; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(1 Suppl):Abstract nr B053.
- ©2016 American Association for Cancer Research.