Non-viral is better

EuroBiotech_Just recently, co-infections with AAV2 vectors were identified in studies by UK scientists as the cause of unusually clustered infantile hepatitis this summer. What are the advantages of tailored vector-free transfection systems for primary cells in gene and cell therapies, and what is the current state of development in this area?

Wagner_Virus-free gene transfer has been of interest for many groups for a couple of decades, because it would be much cheaper than conventional replication-deficient retroviruses, which are complex biological organisms and require extensive purification and quality control. I believe that the first alternative technology promoted to replace retroviruses in T cell therapies were transposon-transposase systems such as PiggyBac or Sleeping Beauty. However, they have not been broadly adopted, because many groups struggled to establish stable manufacturing processes in the early days. First, the enzymes were not efficient enough, then there were the toxicities associated with conventional DNA plasmids. With the advent of mRNA technologies and improved plasmid manufacturing, transposases have seen somewhat of a revival, but recent reports of malignant transformation show that random transgene integrations may actually be a safety risk in differentiated cell types, such as T cells. Previously, this was something we had only observed in clinical trials with hematopoietic stem cells. Therefore, gene editing and its targeted mode of action may have a major advantage moving forward: By combining a programmable nuclease and template DNA, we can use homology-directed repair to integrate our therapeutic transgenes in a precise location. This can be achieved with viruses that deliver the template, but also in a completely non-viral fashion using synthetic DNA templates. In my opinion, this is one of the most promising virus-free platforms to engineer immune cell therapies at the moment.

EuroBiotech_What transfection efficiencies and immunogenicities do DNAs generated by genome editing show in comparison to lentiviral and adenoviral gene shuttles, and what about the translation of such platforms by partners with complementary know-how into medical practice?

Wagner_Viruses usually exploit membrane receptors on cells and endocytosis-mediated uptake, which can be very efficient in certain cell types. In fact, viruses are evolutionarily selected to do this very well. On the other side, human cells have evolved strategies to detect such viral components as well using pathogen associated pattern receptors. For example, recognizing foreign non-self-motifs in viral RNA or DNA. Non-viral approaches usually involve transfection with electroporation, chemicals or nanoparticles. Electroporation is the current gold standard and we have very efficient protocols, which reliably deliver virus-free CRISPR-Cas components, mRNA and synthetic DNA templates in more than 90% of T cells or hematopoietic stem cells in a single electroporation. Of course, large amounts of synthetic nucleic acids can also trigger the innate immune pathways in cells, which others and we have demonstrated in T cells too. For mRNA, we can avoid much of this by making the mRNA look similar to normal human mRNA and replacing uracil with less-immunogenic versions, such as pseudouridine. DNA is a bit trickier. Flooding the cells cytosol with large amounts of double-stranded DNA can be very toxic. We tested whether blocking intracellular DNA sensors after transfection can avoid this, and there is a partial impact of individual pathways, but we cannot prevent toxicity at larger doses required for higher gene-transfer efficacy. I believe that removing impurities in synthetic DNA templates as well as reducing physical stress during the transfection by removing the DNA size or using single-stranded DNA templates are promising options to reduce toxicity and improve gene transfer efficacy using gene editing. I have recently seen other physical transfection methods that is squeezing cells through small pores in the presence of nucleic acids or use microinjections to deliver them and the needed CRISPR components. It will be exciting to see how the recent hype in mRNA therapeutics and accompanying technologies, such as lipid nanoparticles, will influence how we will transfect T cells in the future.  

EuroBiotech_Are there already examples of clinical use of the synthetic DNA platform in TCR-T or CAR-T approaches in blood cancers or beyond, or is this planned and what are the data?

Wagner_On 31st August 2022, Jiquin Zhang and colleagues published the first clinical study using CAR T cells produced by virus-free using CRISPR-Cas9 and synthetic DNA in Nature. The first results in lymphoma patients are very encouraging, because manufacturing was feasible, and most patients benefited from the treatment. Aside from that, others and we are working very hard to bring other TCR-T or CAR-T therapies to patients using similar manufacturing approaches but other diseases. Recently, there has been a major update from one of the frontrunners around Dr Alexander Marson with an updated technology, that uses CRISPR and modified single-stranded DNA. They already demonstrate large-scale manufacturing of CAR T cells for multiple myeloma, and I assume we can expect clinical data from them or one of their associated spin-outs soon.

EuroBiotech_What about cost projections and supply constraints for the synthetic DNAs versus vector-based approaches used?

Wagner_In theory, synthetic DNA is much more flexible, less costly, very stable and thereby a serious contender for vector-based engineering. The waiting times and costs for commercial vector production are slowing down early-stage academic investigations with novel CARs or TCRs. For personalised approaches, where the transgene is customized for each patient, vector manufacturing is simply cost- and time-prohibitive today. Customised DNA synthesis has become much more affordable over the last ten years, and I am expecting that we will see more innovation and growing manufacturing capacity in the synthetic DNA/RNA space. This will likely cause competition and increase quality and price of synthetic DNA.
My dream would be a bench top-like device that can be set up in GMP labs around the globe, which would enable groups like ours to synthesise single-stranded or double-stranded synthetic DNA for diverse applications. There are the first research-grade machines to make small oligos, but it is still science fantasy for clinical applications now because it would require almost perfect fidelity and long nucleic acids are still hard to make. Until then, we will rely on DNA synthesis companies around the globe to increase their manufacturing capacity and come up with protocols to enable very small scale and very large synthesis scales to accommodate the needs of a highly innovative and changing field.

EuroBiotech_How long do you think it will take to translate this approach and where do you see opportunities and challenges?

Wagner_As Zhang and colleagues demonstrated, the future is already here. Identifying the best genomic regions that enable the generation of potent CAR or TCR T cells using non-viral knock-ins is an interesting question. More studies are needed to answer the question of whether the DNA double-strand breaks are a problem for the safety of non-viral knock-in based cell products. To date, we still need DNA double-strand breaks to trigger the repair mechanisms that integrate our transgenes at our location of choice. This will be a big topic in projects where multiple genes are targeted at once, for example for allogeneic cell therapies.
In a universal off-the-shelf CAR T cell product, we want to edit additional genes to increase the persistence or functionality of CAR and TCR T products. With conventional editors, this is very likely to lead to translocations between break points. Novel CRISPR editors such as base editors or prime editing may circumvent DNA breaks in the future, but they must also be carefully evaluated regarding undesired off-target effects on the genome.

EuroBiotech_How is your research progressing and where do you want to be in a year?

Wagner_Overall, we have started the first scaling attempts and established processes to optimize cell yields for allogeneic CAR-T cells. Unfortunately, many of our clinical programs were slowed by the pandemic and there is a backlog of cell products that await clinical testing. For example, we are finally planning to start a first trial with CRISPR-Cas9 gene-edited regulatory T cells in patients with kidney transplantation. Here, we use CRISPR-Cas9 to force a gene knock-out and induce drug resistance.
This is an important project for us and the IND process guides us in preclinical studies and may help us to finalize our process for the CAR knock-in platform using synthetic DNA. I hope that by the end of next year, we will have perfected manufacturing for our first CAR-T product candidate using non-viral knock-in. Due to the regulatory requirements in Europe, it will be important to identify partners to provide the synthetic DNA and other gene editing components made in a GMP environment.


This interview sponsored by GenScript showcases our interaction with Dr. Dimitrios L. Wagner. GenScript is a provider of CRISPR reagents for preclinical use and GMP applications. Dr. Wagner’s lab tests GenScript CRISPR reagents to translate non-viral CRISPR knock-in from the lab to GMP-compatible clinical applications of cell and gene therapies for diseases with unmet clinical need, including CAR T cell therapy for cancer applications. This interview was originally published in European Biotechnology Magazine Autumn Edition 2022.