CRISPR-Based Therapeutics Blaze an In Vivo Path to the Clinic

Pioneering companies are showing that CRISPR systems can be tucked inside delivery vehicles and injected locally or systemically to bring about cures

Therapeutic applications of genome editing were envisioned at least as early as the mid-1990s, when the first sequence-specific genome editing technologies emerged. Initially, such applications were considered distant prospects, but by 2012, they suddenly seemed near to hand. It was at that time that CRISPR technologies emerged.

CRISPR, which stands for clustered regularly interspaced short palindromic repeats, came to notice as a curious pattern in bacterial DNA. Attempts to understand CRISPR led to the discovery of bacterial immune systems. Even better, these systems were found to include RNA-programmable nucleases, that is, CRISPR-associated proteins (Cas proteins) that complex with guide RNAs to target specific DNA sequences.

Cas proteins and guide RNAs have become mainstays of CRISPR genome editing technology, and this technology has been used to realize ex vivo therapies. For example, CRISPR-modified T cells are being deployed in immunotherapies, and CRISPR-modified hematopoietic stem cells are showing promise against sickle-cell disease and β-thalassemia.

More ambitious CRISPR-based in vivo therapies are now in development. These therapies, which allow genes to be edited inside patients, would be impossible but for the development of improved CRISPR-based genome editing tools, such as the alternative or modified Cas proteins discussed in this article. When these tools are encapsulated by viral capsids or lipid nanoparticles (LNPs), they constitute therapeutics that may be administered locally, within specific tissues, or systemically throughout the entire body.

In vivo CRISPR therapy enters the clinic

In November 2020, the first systemically delivered CRISPR-Cas9 therapy entered clinical trials. This therapy, which is called NTLA-2001, is being developed by Intellia Therapeutics and Regeneron. NTLA-2001 consists of a lipid nanoparticle that is designed to target the liver, and that encapsulates a Cas9-enzyme-encoding messenger RNA and a guide RNA. Once expressed within liver cells, the Cas9 enzyme is directed by the guide RNA to the gene for transthyretin (TTR).

NTLA-2001 diagram
Intellia Therapeutics and Regeneron are developing NTLA-2001, a therapeutic that is designed to target liver cells and disable a gene that is responsible for the production of misfolded transthyretin (TTR) protein. In November 2020, NTLA-2001 became the first systemically delivered CRISPR-Cas9 therapy to enter clinical trials. It is being evaluated for its effectiveness against TTR amyloidosis. [Intellia Therapeutics]

Through targeted knockdown of the gene for TTR, NTLA-2001 reduces the serum concentration of the protein. In this way, NTLA-2001 treats TTR amyloidosis, a rare genetic disease that is caused by the accumulation of misfolded TTR in tissues such as the nerves, heart, and kidneys. The therapy is administered just one time, through an intravenous injection.

In a Phase I clinical trial that included six patients, an 87% average reduction in serum TTR was observed for the patients who received the highest dose (Gillmore et al. N. Engl. J. Med. 2021; DOI: 10.1056/NEJMoa2107454). The reductions in serum TTR could be a durable effect, suggest the developers of NTLA-2001. They point to preclinical studies in which deep and long-lasting TTR reductions were observed in rodents and nonhuman primates. In fact, TTR levels remained low even when the drug was evaluated in a rodent model of accelerated liver regeneration.

“Because we’re changing the DNA, the genetic makeup of that cell, the expectation is that once you do that, that cell will never revert back,” says Laura Sepp-Lorenzino, PhD, executive vice president and chief scientific officer at Intellia Therapeutics. “These technologies in medicine will have a long-lasting, potentially curative effect in humans.”

In the clinical study, researchers assessed NTLA-2001’s potential for deleterious off-target effects by performing genome-wide assays and targeted sequencing. The researchers identified seven candidate off-target sites, all of which were located in noncoding regions. For these sites, no detectable levels of off-target editing were found, even when primary cell cultures of human hepatocytes were treated with concentrations of NTLA-2001 that were up to three times higher than the 90% maximal effective concentration.

Alternatives to Cas9

Although the CRISPR-associated protein derived from the Streptococcus pyogenes nuclease, known as SpCas9 or Cas9, is the most widely used Cas protein, it is not the only option. Alternatives are being explored by several companies. For example, AsCas12a is being developed by Editas Medicine.

AsCas12a is derived from a gut bacterium called Acidaminococcus sp. Studies have demonstrated that this protein has fewer off-target effects than Cas9. However, it also has a lower editing efficiency, meaning fewer cells receive the desired edit.

AsCas12a Ultra diagram
AsCas12a Ultra, a fine-tuned nuclease specifically tailored for ex vivo therapeutic applications, is being developed by Editas Medicine and Integrated DNA Technologies (IDT). Left panel: AsCas12a Ultra shown binding to DNA. Right panel: Results from an experiment in which all possible guide RNAs for the B2M gene were tested with the wild-type AsCas12a and AsCas12a Ultra. All guide RNAs tested achieved highly efficient editing with AsCas12a Ultra. [Editas Medicine]

Editas has been working to address this limitation. In collaboration with Integrated DNA Technologies (IDT), Editas has developed a more active version of AsCas12a that is called AsCas12a Ultra. Scientists representing the companies published a study (Zhang et al. Nat. Commun. 2021; 12: 3908) indicating that AsCas12a Ultra, like AsCas12a, has much better on-target specificity than Cas9. AsCas12a Ultra was also shown to be nearly 100% efficient at gene editing across all cell types and target sites tested, a capability that has never been seen with any other gene editing nuclease. Finally, AsCas12a Ultra was shown to be highly potent.

“A more potent nuclease means you need to add less of your CRISPR-Cas agent to achieve the same effect,” says John Zuris, PhD, associate director of editing technologies at Editas. “This is especially important for multiplexed editing, where you need to add many different CRISPR-Cas agents at the same time. We show that AsCas12a Ultra is, on average, over 40-fold more potent than wild-type AsCas12a.”

Editas has engineered an AsCas12a protein based on AsCas12a Ultra for cell therapy. Currently, the engineered protein is being evaluated for its ability to treat sickle-cell disease. The company is now enrolling patients for a Phase I/II study.

Another alternative to SpCas9 that the company is exploring is a nuclease derived from Staphylococcus aureus. This nuclease is called SaCas9, and it engages in less off-target activity than SpCas9. Another advantage offered by SaCas9 is its compactness, a quality that Editas is exploiting to develop a treatment for an inherited retinal disorder, Leber congenital amaurosis (LCA), that is caused by mutations in the CEP290 gene.

The treatment consists of an adeno-associated virus (AAV) vector that incorporates a plasmid that expresses SaCas9 and human CEP290-specific gRNAs. When EDIT-101 is administered via a subretinal injection, it delivers the gene editing machinery directly to photoreceptor cells.

“SaCas9 is smaller than SpCas9 and enables us to package all the CRISPR-Cas elements in a single AAV vector for delivery to photoreceptors,” Zuris asserts.

In vivo base editing

The familiar SpCas9 nuclease and the relatively new alternatives cited thus far in this article are not the only tools in the CRISPR gene editing toolbox. Other options are engineered enzymes that eschew nuclease activity and instead serve as base editors.

The first two base editors, a cytosine-to-thymine base editor (CBE) and an adenosine-to-guanosine base editor (ABE), appeared in 2016 and 2017, respectively. Both were developed in a laboratory led by David Liu, PhD, at Harvard University. Both consist of a catalytically impaired Cas9 fused to a base-swapping enzyme.

Because the base editors incorporate catalytically impaired Cas9, they don’t create a double-strand break in the DNA, and they don’t rely on the cell’s repair system to make the desired changes. Instead, the base editors cause single-strand breaks. At each break, a base editor will swap one base for another.

According to a recent study (Anzalone et al. Nat. Biotechnol. 2020; 38: 824–844) from Liu’s team, base editors could theoretically be used to correct approximately 30% of the known pathogenic point mutations listed in ClinVar, a public database developed by the U.S. National Institutes of Health.

An ABE base editor is the lead product candidate at Verve Therapeutics, a company that focuses on cardiovascular diseases. This base editor is called VERVE-101, and it is being used to inactivate the PCSK9 gene and reduce levels of low-density lipoprotein (LDL) cholesterol in blood.

Verve Therapeutics base editing system illustration
Verve Therapeutics is developing base editing systems for treating cardiovascular diseases. For example, the company believes that its approach could allow three independent atherosclerotic cardiovascular disease pathways—low density lipoprotein (LDL), triglyceride-rich lipoprotein (TRL), and lipoprotein(a) (Lp(a))—to be targeted at once, realizing combinative effects.

Like NTLA-2001, Verve’s therapy uses LNPs to deliver messenger RNA that, in this case, encodes for ABE as well as guide RNA to the liver. It is also designed as a one-time, intravenous injection and, so far, has had very promising results.

In a recent study on nonhuman primates (Musunuru et al. Nature 2021; 593: 429–434), researchers at Verve Therapeutics found their therapy maintained an 89% reduction in PCSK9 expression and a 59% reduction of LDL for the eight months of the study. As with the NTLA-2001 results, this result suggests that even as the liver regenerates, the edits to the DNA remain. In addition, the researchers found no evidence of edits at the top 141 sites that had been identified as possible off-targeting sites.

“I think what is terrifically exciting about this is that it’s a very validated biology, but a very new approach,” says Andrew Bellinger, MD, PhD, chief scientific officer at Verve Therapeutics. “The excitement is the potential of this approach, [which can make] single base pair changes in one gene at one spot in your genome, [and which] can be so effective at turning off the gene so completely and so durably.”

Verve Therapeutics is focusing on patients who have genetic diseases like familial hypercholesteremia, and who are at high risk of atherosclerotic cardiovascular disease (ASCVD). The company also plans to eventually expand the treatment to anyone who has or is at risk for ASCVD.

“We can do this in very high-risk patients, and that’s the plan for the next few years,” Bellinger relates. “Ultimately, we think this approach of lipid nanoparticles delivering adenine base editors is one that’s very generalizable for larger patient populations.”

Exploring the new frontier with prime editing

Although base editors allow for a lot more specificity, they come with their own limitations. The main limitation is that base editors can do nothing but introduce transition point mutations by swapping one base pair for another. Base editors do not make insertions or deletions. However, an even newer technology called prime editing, again developed by Liu, can do much more.

Like base editors, prime editors can “nick” DNA (that is, cut just one of the two DNA strands) and bring about base-to-base conversions. In addition, prime editors can make small insertions and deletions. With these capabilities, prime editors could allow for the precise treatment of even more diseases caused by genetic mutations, about 90% of those listed in ClinVar.

In 2019, Liu co-founded Prime Medicine to develop prime editing as a search-and-replace technology. In a study from Liu’s group published the same year (Anzalone et al. Nature 2019; 576: 149–157), search-and-replace prime editing was described as the use of “a catalytically impaired Cas9 endonuclease fused to an engineered reverse transcriptase” that is programmed with “a prime editing guide RNA (pegRNA) that both specifies the target site and encodes the desired edit.”

In the study, over 175 edits were made across four different human cell lines, including insertions, deletions, and point mutations, with varying degrees of success. Overall, the researchers found that prime editing causes much less off-target editing than Cas9 and has strengths and weaknesses that complement those seen with base editing.

In one demonstration of prime editing’s abilities, a mutation that causes sickle-cell disease could be corrected with a gene editing efficiency of 44%, and an occurrence of indels, or unintended insertions and deletions, of about 5%.

“Prime editing makes the correct edit of a gene mutation at the exact target site and exhibits minimal off-target activity,” says Jennifer Gori, PhD, vice president of research at Prime Medicine. “Prime editing also does not cause double-strand breaks in chromosomal DNA or affect cell viability, which we believe may contribute to better patient outcomes, fewer side effects, and overall improved safety.”

The company, which was launched in July 2021 with $315 million in financing, will be partnering with another one of Liu’s companies, Beam Therapeutics, to develop multiple drug discovery programs.

“We are working on a variety of delivery methods across multiple modalities, including methods such as nonviral nanoparticles and viral vectors,” Gori notes. “We look forward to further developing prime editing technology and progressing our preclinical programs toward the clinic. We are focused on making a difference to patients, and our hope is that our programs may cure or halt the progression of genetic diseases.”