Preview

PDF, English (PhD Thesis in PDF/A-1B) - main document Download (9MB) | Terms of use

Abstract

Adoptive T cell therapy (ACT) with Chimeric Antigen Receptors (CARs) revolutionized the field of immunotherapy by achieving up to 90% complete responses in the treatment of hematological malignancies. Due to their success in clinical trials, there are currently 6 FDA approved CAR T cell products for treatment against B cell malignancies. Unlike TCR T cells, antigen recognition by CAR T cells is peptide-Major Histocompatibility Complex independent. Therefore, an isolated functional CAR construct can be used to treat a cohort of patients sharing a common antigen. Unlike the success observed in the CAR T cells therapies against hematological malignancies, the treatment of solid tumors with CAR T cells showed a limited success so far. There are couple of reasons for impaired function of CAR T cells in the solid tumors. The highly immunosuppressive environment of the tumor stroma can hinder CAR T cell infiltration and activation in the tumor tissue. Moreover, there is limited number of identified CARs suitable for ACT against solid tumors. Therefore, more binders need to be screened to identify tumor specific CARs. Current golden standard for CAR identification is the use of single-chain phage display libraries. Screening of phage display libraries can yield in high affinity scFvs, but these constructs still need to be cloned into a functional CAR backbone for further characterization. It is known that not all isolated scFvs can give rise to functional CAR constructs. Moreover when cloned into a CAR context, scFvs may cause a tonic signaling effect that leads to early CAR T cell exhaustion, thus decreasing antitumor potential of T cells. Besides, one of the biggest disadvantage of phage display libraries is that a functional CAR identification can take up to several months. Thus, it is hardly possible to use phage display libraries for development of personalized CAR T cell therapies as during the CAR isolation and validation period, disease progression may render the generated CAR T cells unbeneficial to the patient. In this study, an innovative platform was generated to accelerate identification of antigen-specific CAR constructs. To do so, a cellular CAR library screening approach was developed by engrafting a plasmid CAR library into a Jurkat reporter cell line. Generation and screening of such cellular CAR libraries for antigen-specific CAR identification has not been reported yet. For this purpose, an antigen-specific stimulation sensitive single-cell derived reporter Jurkat clone was cultivated. This clone was then transfected with a plasmid CAR library that was generated by cloning a randomly assembled naïve scFv repertoire into a functional CAR backbone. Since the CAR library was engrafted into the reporter cells directly, tonic signaling CARs were easily identified as they activated the cells in the absence of antigen-specific stimulation and excluded from further analysis. Consequently, remaining library cells that express functional CARs were screened for an antigen of interest on a single-cell level by using Bruker’s Lightning™ device. Tumor antigen reactive cells were identified by induced reporter expression and exported from the Lightning™ device to have their CARs sequenced for subsequent functional validation. Above mentioned process, from generation of plasmid CAR libraries to in vitro validation, allows CAR-T cell discovery in as little as 6 weeks. Thus, cellular CAR libraries will enable the development of highly individualized CAR T cell therapies against patient derived tumor antigens in a timeframe that the patient can still benefit from.