Search

Adoptive cellular therapy using genetically engineered immune cells holds tremendous promise for the treatment of advanced cancers. While the number of available receptors targeting tumor specific antigens continues to grow, the current reliance on viral vectors for clinical production of engineered immune cells remains a significant bottleneck limiting translation of promising new therapies. Here, we describe an optimized methodology for efficient CRISPR-Cas9 based, non-viral engineering of primary human T cells that overcomes key limitations of previous approaches. By synergizing temporal optimization of reagent delivery, reagent composition, and integration mechanism, we achieve targeted integration of large DNA cargo at efficiencies nearing those of viral vector platforms with minimal toxicity. CAR-T cells generated using our approach are highly functional and elicit potent anti-tumor cytotoxicity in vitro and in vivo. Importantly, our method is readily adaptable to cGMP compliant manufacturing and clinical scale-up, offering a near-term alternative to the use of viral vectors for production of genetically engineered T cells for cancer immunotherapy.

Cancer immunotherapy using T cells and NK cells modified with viral vectors to express a chimeric antigen receptor (CAR) has shown remarkable efficacy in treating hematological malignancies in clinical trials. However, viral vectors are limited in their cargo size capacity, and large-scale manufacturing for clinical use remains complex and cost prohibitive. As an alternative, CAR delivery via DNA transposon engineering is a superior and cost-effective production method. Engineering via transposition is accomplished using a two-component system: a plasmid containing a gene expression cassette flanked by transposon inverted terminal repeats (ITRs) paired with a transposase enzyme that binds to the ITRs, excises the transposon from the plasmid, and stably integrates the transposon into the genome.Here, we used the newly developed hyperactiveTc Buster(Bio-Techne) transposon system to deliver a transposon containing a multicistronic expression cassette (CD19-CAR, mutant DHFR, and EGFP) to primary human peripheral blood (PB) NK cells and T cells. We optimized methods to avoid DNA toxicity and maximize efficiency. Our cargo contained a mutant dihydrofolate reductase (DHFR) which allowed us to enrich for stable transposon integration using methotrexate (MTX) selection. We then tested CAR-NK and CAR-T cells in functional assays against CD19-expressing Raji cells. CAR-expressing NK and T cells produced significantly more cytokines than CAR-negative controls and efficiently killed target cells. We recognize that cryopreservation manufactured CAR-expressing cells will be necessary for clinical translation. We observed reduced cytotoxicity of CAR-NK cells immediately after thaw, but increasing the NK dose overcame this loss of function.Our work provides a platform for robust delivery of multicistronic, large cargo via transposition to primary human NK and T cells. We demonstrate that CAR-expressing cells can be enriched using MTX selection, while maintaining high viability and function. This non-viral approach represents a versatile, safe, and cost-effective option for the manufacture of CAR-NK and CAR-T cells compared to viral delivery.

Background Engineered immune cells hold tremendous promise for the treatment of advanced cancers. As the scale and complexity of engineered cell therapies increase, reliance on viral vectors for clinical production limits translation of promising new therapies. Here, we present an optimized platform for CRISPR/Cas9-targeted, non-viral engineering of primary human T cells that overcomes key limitations of previous approaches, namely DNA-induced toxicity and low efficiency integration of large genetic cargos. Methods A systematic optimization of nucleic acid delivery, editing reagent composition, and culture protocol was performed to overcome DNA toxicity. Targeted knockin (KI) at AAVS1 and TRAC was compared across multiple vector configurations with genetic cargos ranging from 1 to 3 kilobases (kb) in size. Integration efficiency was measured by flow cytometry and sequencing. Off-target editing and integration were evaluated using GUIDE-seq and targeted locus amplification (TLA), respectively. Phenotype and function of non-virally and lentivirus engineered CAR-T cells was compared using flow cytometry, cytokine profiling and cytotoxicity assays. Results We identified a temporal window following T cell activation where transfection efficiency, cell-cycle-status, and cytosolic DNA sensor expression were optimal for targeted DNA integration and reduced toxicity. Within this window, we targeted a 1kb GFP reporter to the AAVS1 locus with an efficiency of ~45% using homologous recombination (HR). Efficiency was reduced to ~11% with a larger ~3kb TCR cassette targeted to the TRAC locus, consistent with previous reports.1–3 To improve large cargo integration we employed homology mediated end-joining (HMEJ) and short homology design (48bp vs. ~1kb for traditional HR).4 Using HMEJ, knockin of the 1kb GFP cassette at AAVS1 reached ~70%. Strikingly, integration of the 3kb TCR at TRAC reached ~50% using HMEJ. Additional optimization of the culture protocol doubled post-engineering survival and proliferation (up to ~35-fold expansion in 7 days). Non-virally engineered TRAC KI CAR-T cells were phenotypically and functionally equivalent to lentivirally engineered T cells in vitro. In vivo assays in xenograft models are underway and results will be presented. Conclusions Comprehensive, orthogonal optimization of parameters impacting nucleic acid delivery and DNA-toxicity in combination with novel modalities for integration achieved knockin of TCR and CAR cargo at efficiencies equivalent to that of current viral vector platforms without compromising expansion or function. Our protocol is suitable for clinical scale production under GMP conditions and offers an improved methodology over previous methods for non-viral engineering of human T cells. References Roth TL, Puig-Saus C, Yu R, Shifrut E, Carnevale J, Li PJ, Hiatt J, Saco J, Krystofinski P, Li H, Tobin V, Nguyen DN, Lee MR, Putnam AL, Ferris AL, Chen JW, Schickel J-N, Pellerin L, Carmody D, Alkorta-Aranburu G, Del Gaudio D, Matsumoto H, Morell M, Mao Y, Cho M, Quadros M, Gurumurthy CB, Smith, B, Haugwitz M, Hughes SH, Weissman JS, Schumann K., Esensten JH., May AP, Ashworth A., Kupfer G. M., Atma S., Greeley W. & Marson A. Reprogramming human T cell function and specificity with non-viral genome targeting. Naturedoi:10.1038/s41586-018-0326-5 Parker Autoimmune SN, Zuckerberg Biohub C, Francisco S. & Helen U. Polymer-stabilized Cas9 nanoparticles and modified repair templates increase genome editing efficiency. Nat. Biotechnol. doi:10.1038/s41587-019-0325-6 Schober K, Muller TR, Gokmen F, Grassmann S, Effenberger M, Poltorak M, Stemberger C, Schumann K, Roth TL, Marson A. & Busch DH. Orthotopic replacement of T-cell receptor α- and β-chains with preservation of near-physiological T-cell function. Nature Biomedical Engineering 3, 974–984 ( 2019). Wierson WA, Welker JM, Almeida MP, Mann CM, Webster DA, Torrie ME, Weiss TJ, Kambakam S, Vollbrecht MK, Lan M, McKeighan KC, Levey J, Ming Z, Wehmeier A, Mikelson CS, Haltom JA, Kwan KM, Chien C-B, Balciunas D, Ekker SC, Clark KJ, Webber, BR, Moriarity BS, Solin SL, Carlson DF, Dobbs DL, McGrail M & Essner J. Efficient targeted integration directed by short homology in zebrafish and mammalian cells. Elife9, ( 2020).