Targeted Integration of a CAR at a Novel Genomic Safe Harbor Directs Potent Therapeutic Outcomes

Chimeric receptor antigen (CAR)-T cell therapy using CARs specific for CD19 have been remarkably successful for treating chemo-refractory/relapsed B cell malignancies. These successes not withstanding, therapeutic outcomes between patients are variable and occasional cases of clonal expansion of the transduced T cells have been observed, albeit without leukemic transformation. These issues are cause for concern and need to be addressed to achieve better and safer therapeutic outcomes. Clonal expansion due to insertional mutagenesis and variegated transgene expression due to position effects are well established to be the consequence of the semi-random integration pattern afforded by gamma-retroviral and lentiviral vectors. We previously established that integration of a CAR cDNA in the TCR alpha locus (TRAC) provides consistent, regulated expression of CD19 CARs and superior CAR T cell efficacy in a mouse model of B-cell acute lymphoblastic leukemia (B-ALL). Here, we identify a novel extragenic site devoid of any known function and remote from endogenous genes, i.e. a ‘genomic safe harbor’ (GSH), that can be efficiently targeted in human T cells and drives potent CAR T cell therapy in the B-ALL mouse model.

To identify GSHs that could be efficiently targeted in T cells by CRISPR-Cas9 and that could also support durable transgene expression, we screened for genomic regions meeting both, GSH criteria and high chromatin accessibility in T cells as measured by ATAC-seq. In human primary T cells, we identified 379 such sites. Ten of the highest accessible sites were investigated. All showed high (>90%) cleavage efficiency and allowed for CAR cDNA targeted integration and expression which also translated into effective cytolytic activity of the CARs within a few days after transduction. However, thereafter CAR expression diminished over the course of a week at most but not all of these sites. In order to prevent possible heterochromatinization, we incorporated chromatin insulator elements with barrier activity flanking the CAR transcription unit. Incorporation of the chromatin insulator element dramatically improved CAR expression and functionality at one site, whereas 3 other GSHs tested were not affected. One of the 10 GSHs maintained long-term CAR expression without requiring an insulator and directed potent anti-leukemic CAR T cell efficacy in a B-ALL ‘CAR stress test’ mouse model, matching the T cell potency afforded by integrating the CAR cDNA at the TRAC locus. This finding highlights the major effect of the integration site on transgene expression and ensuing therapeutic efficacy.

We identified an extragenic GSH site that can be used for effective T cell engineering and sustained expression of a CAR. Through this study, we provide a platform for identifying GSHs that could be reliably targeted for safe and predictable expression of CARs or other immunomodulatory transgenes to potentiate adoptive immunotherapy.