Swift progress in gene editing technology is bringing hope to people who suffer from incurable genetic diseases. Staying on course, however, can be tricky. It is all too easy to run afoul of intellectual property (IP) rights, which have constrained some companies’ freedom to operate. To steer clear of such hazards, companies are developing novel CRISPR compositions that can deliver higher specificity, efficiency, and fidelity.
Besides functioning as a research tool, gene editing is generating remedies that are pressing forward to the clinic. Many potential therapies are already in the pipeline to treat a variety of diseases, and many more therapies are entering the pipeline as component suppliers expand their product offerings.
In vivo and ex vivo approaches
“It is an exciting time for medicine as we progress toward developing transformative, durable medicines for people living with serious diseases of unmet need,” says Kate Zhang, PhD, vice president, biological development, Editas Medicine.
The company’s lead pipeline candidate is EDIT-101, which is currently in clinical development for the treatment of Leber congenital amaurosis 10 (LCA10), a rare inherited retinal degenerative disorder that causes blindness or severe visual impairment at birth or during the first months of life.
According to Zhang, Editas plans to file an investigational new drug (IND) application by the end of the year for another therapy, EDIT-301, which has shown promise in recent preclinical studies. With EDIT-301, cells are collected from a patient, genetically edited, and then finally returned to the patient to treat sickle-cell disease. The cells are CD34+ hematopoietic stem cells, and they are edited at the HBG1/2 promoter in the β-globin locus using CRISPR-Cas12a (also known as Cpf1) to induce fetal hemoglobin (HbF) in naturally occurring mutations.
IND-enabling studies also have been initiated for EDIT-201, an allogeneic natural killer (NK)–cell medicine for the treatment of solid tumor cancers.
Access to both Cas9 and Cas12a allows Editas to use whichever nuclease will execute the best edit, explains Zhang. CRISPR-Cas12a is structurally distinct, and its technology has independent IP. The potential benefits include an increased number of editable sites due to distinct protospacer adjacent motifs, and increased efficiency and accuracy for some forms of gene repair due to staggered DNA cuts. This natural system is simpler to manufacture and deliver, requiring only a short, single CRISPR guide RNA (gRNA) and no trans-activating CRISPR RNA (tracrRNA).
During the development of EDIT-301, Editas compared SpCas9 and Cas12a. The latter demonstrated a superior editing profile at the HBG distal CCAAT box region for persistent and high HbF expression. The preclinical in vivo data demonstrated potentially therapeutically beneficial and robust high levels of HbF expression with long-term durability and pancellular distribution.
Capable of highly efficient knockout and transgene knockin across various cell types including induced pluripotent stem cells (iPSCs), Cas12a not only has a favorable editing profile, it can greatly increase the chances of generating clones with constructs that can drive powerful allogeneic cell therapies.
Allele-specific editing
Emendo Biotherapeutics discovers and develops novel optimized nuclease-gRNA compositions that enable allele-specific editing. The compositions aim to treat autosomal dominant disorders, which represent most untreatable genetic diseases. The autosomal dominant disorder known as severe congenital neutropenia (SCN) is of particular interest to the company. In SCN, mutations in the ELANE gene, which encodes neutrophil elastase, prevent the maturation of neutrophils and compromise the immune system’s ability to fight infections. More than 200 SCN-associated mutations in ELANE have been characterized. Targeting each one to knockout the mutated allele is impractical.
“Instead, we identified single nucleotide polymorphisms (SNPs) through analyses of databases of healthy cohorts to enable higher coverage of the patient population with a few nuclease-gRNA combinations,” relates Rafi Emmanuel, PhD, vice president of R&D at Emendo. “We identified three SNPs for ELANE that cover about 80% of the population. In collaboration with the University of Washington in Seattle, we used patients’ sequence data to show the same prevalence of the SNPs in the patients’ cohort. We rely on SNPs to knockout only the mutated allele, a very challenging discrimination level.”
Due to the limitations of spCas9, Emendo is now focusing on discovering new nucleases, termed OMNI Nucleases. Using several proprietary libraries, Emendo performs an initial SNP-based selection in bacteria to obtain variants that discriminate between wild-type and mutated alleles.
“We follow several cycles of positive and negative selection,” Emmanuel details. “The SNPs can be in either the wild type or mutated allele. Therefore, we select our variants on both alleles and validate that they are specific to both according to the guide that is being used for the editing. For three SNPs, we need six compositions.”
When specific nuclease variants are sought, selectivity or activity is often compromised. A variant selected in bacteria may not be very active in mammalian cells; therefore, an additional process of variant screening in mammalian system is performed, which usually ends with a handful of variants.
The selection process is continually optimized. “We want more variants that are relevant to our needs,” Emmanual emphasizes. “[They need to be] as active as the wild type and provide higher fidelity and specificity.”
Nonviral gene editing
Cancer immunotherapies that deploy chimeric antigen receptor (CAR) T cells are starting to reach the clinic. For example, the FDA has approved the CAR T-cell therapies known as Kymriah and Yescarta, which are being commercialized by Novartis and Gilead
Sciences, respectively. Both Kymriah and Yescarta are autologous products.
Although autologous CAR T cells show promise, producing them is challenging. Large numbers of T cells need to be isolated from late-stage patients, and production and quality control need to be finished within three weeks. These difficulties can be avoided if “off the shelf” allogeneic CAR T cells are processed from healthy donors’ T cells. The allogeneic approach also promises easier manufacturing scale-up and lower treatment costs.
But the allogeneic approach poses challenges of its own. The allogeneic CAR T cells may cause life-threatening graft-versus-host disease, or they may be rapidly eliminated by the host immune system. Fortunately, as recent publications have demonstrated, these challenges can be overcome provided the appropriate strategies are implemented.1–3 One strategy involves a combination of CRISPR gene editing tools and viral vectors to accomplish multiplex genome engineering. Another strategy is the use of nonviral CRISPR editing systems.4,5
Nonviral methods for engineering CAR T cells can be safer and more cost-effective than viral methods. For example, CAR elements can be inserted more precisely with nonviral methods than with viral methods, lessening the chance of oncogenic insertion. Also, nonviral methods pose fewer immunogenicity issues that do adeno-associated virus vectors.6
Finally, DNA payload manufacturing is less expensive and faster with nonviral vectors than it is with viral vectors. Since production does not use mammalian cell lines or animal-derived products, materials require less complex quality control.
A comprehensive CRISPR product portfolio for nonviral cell engineering is offered by GenScript Biotech. “For the early discovery stage,” says Lumeng Ye, PhD, a senior scientist at the company, “CRISPR gRNA screening libraries, high-purity sgRNAs, Cas9 enzymes, and single-stranded DNA CAR/T-cell receptor knockin templates for high-efficiency gene engineering are available.
“When moving to process development, we have high-quality Cas9 nuclease, sgRNA, and DNA payload materials that can be produced at large scale. We are also expanding our capability to GMP manufacturing.”
Developed and optimized for decades, viral vectors will remain popular gene editing tools even after nonviral vectors become more efficient and achieve widespread adoption. What distinguishes nonviral vectors, Ye suggests, is manufacturing simplicity and ease of use. These attributes will help nonviral vectors mature more quickly than viral vectors.
Clinical-grade RNPs
In 2017, Aldevron became the first company to launch GMP SpCas9 protein as a catalog product. The move reassured researchers who doubted whether they could source clinical-grade CRISPR proteins. “The field has come a long way in three years,” says Max Sellman, product manager, gene editing, Aldevron. “Now more than 60 companies use gene editing techniques to treat a broad variety of ailments.”
The vital first gene editing step is creating the active CRISPR machinery. Currently, most clinical-stage cell therapies involve “point of use” complexing of the Cas9 and sgRNA to form active ribonucleoprotein (RNP), which is immediately applied to cells. This procedure may be subjected to little or no quality testing. So, the degree to which the procedure succeeds in any given run, or demonstrates consistency from run to run, may be unclear.
Now groups are concerned with scale-up and process consistency, relates Sellman. They want assurances that their gene editing processes are fully controlled with no batch differences between RNP composition and performance.
A new platform under development will deliver customized Cas9 RNP for clinical use. Aldevron will complex together GMP Cas9 plus a custom sgRNA and deliver ready-to-use RNP complex with a full certificate of analysis. A consistent RNP source with established specifications simplifies the manufacturing process for cell therapies while offering tighter quality control in a key early step.
The RNP manufacturing process is designed around off-the-shelf CRISPR proteins, especially wild-type SpCas9 and engineered SpyFi™ Cas9 nucleases, both available at GMP quality. Since the specific sequence of the sgRNA will differ, Aldevron is working with key gRNA suppliers to ensure that clients can select their preferred manufacturer in compliance with Aldevron’s quality system requirements.
“We are a custom manufacturer at heart, so we will modify our standard process to meet individual requirements,” informs Sellman. “Targeted for clinical use, our new RNP manufacturing platform is best served by cGMP or GMP-Source quality grades of service; research-grade quality may be suitable for some applications. The process is very scalable from milligram to multigram amounts.”
It all begins at the benchtop
In 1998, while working at the Carnegie Institution in Washington, DC, Craig Mello, PhD, of the University of Massachusetts Medical School (UMMS), and Andrew Fire, PhD, of Stanford University, discovered RNA interference (RNAi). Their work earned them the 2006 Nobel Prize in Physiology or Medicine.
Although the ultimate in scientific recognition took just 8 years, regulatory approval for RNAi took an additional 12 years, notes Satinder Rawat, PhD, senior licensing officer, UMMS. As part of the core team at the UMMS Office of Technology Management, Rawat understands the critical aspects of transferring academic intellectual property.
“We either use software tools or create our own benchmarks while utilizing our network during our assessments,” he details. “Companies and investors are cautious about very early stage and novel technologies. Our strategy is to create a technology bucket with a focus on novelty and quality. Some of our programs have progressed to clinical trials, much further than other universities.”
Rawat cross-pollinates new ideas by creating intra- and interuniversity collaborations. “Elements of intellectual property exist in every contract,” he explains. “Relationship building is crucial to licensing and collaborations as well as generating service contracts for the cores.”
In 2008, he took on the task of building a new gene therapy technology portfolio. Although the portfolio aroused skepticism at first, it has since figured in the launch of several startups. For example, gene delivery technology developed by Robert Kotin, PhD, of UMMS is being commercialized by Generation Bio. The company, which Kotin co-founded, packs genetic payloads into lipid nanoparticles that are preferentially taken up by the liver. “Technology takes a lot of different paths,” advises Rawat. “But it eventually gets to the right people and partners.”
A new approach the department adopted has become a constant model for novel technologies; companies invest in the research prior to licensing, putting dollars into the discovery laboratory. Not only is this model less expensive than starting a new lab, it provides the company access to deep scientific expertise. “You need investment before something becomes a product,” declares Rawat. “Our scientists can become a virtual research team while retaining IP ownership that we can license in the future. We carry the risk until the technology is proven.”
References
1. Depil S, Duchateau P, Grupp SA, et al. ‘Off-the-shelf’ allogeneic CAR T cells: development and challenges. Nat. Rev. Drug Discov. 2020; 19(3): 185–199. DOI: 10.1038/s41573-019-0051-2.
2. Quach DH, Becerra-Dominguez L, Rouce RH, Rooney CM. A strategy to protect off-the-shelf cell therapy products using virus-specific T-cells engineered to eliminate alloreactive T-cells. J. Transl. Med. 2019; 17(1): 240. DOI: 10.1186/s12967-019-1988-y.
3. Mo F, Watanabe N, McKenna MK, et al. Engineered off-the-shelf therapeutic T cells resist host immune rejection. Nat. Biotechnol. 2020; Jul 13. DOI: 10.1038/s41587-020-0601-5.
4. Stadtmauer EA, Fraietta JA, Davis MM, et al. CRISPR-engineered T cells in patients with refractory cancer. Science 2020; 367(6481): eaba7365. DOI: 10.1126/science.aba7365.
5. Roth TL, Puig-Saus C, Yu R. Reprogramming human T cell function and specificity with non-viral genome targeting. Nature 2018; 559(7714): 405–409. DOI: 10.1038/s41586-018-0326-5.
6. Verdera HC, Kuranda K, Mingozzi F. AAV Vector Immunogenicity in Humans: A Long Journey to Successful Gene Transfer. Mol. Ther. 2020; 28(3): 723–746. DOI: 10.1016/j.ymthe.2019.12.010.