Gene edting… editn… editing, time to fix our genome (2/3)

From primitive to advanced tools: the rise of gene editors

The field of gene editing has rapidly advanced in recent years and finally reached the clinical-trial-ready state, opening up exciting new possibilities for treating genetic disorders, improving human health, and attracting investors.

In this three-part study, we aim to provide a comprehensive overview of the current state of gene editing and its potential for the future.

• Part I:, we examined the field's history, the need for gene editing, its likely impact, and the potential $3tn market size.
• Part II: in this opus, we focus on the tools and technologies that enable gene editing players to gain market share, including the latest advances in gene editing techniques and delivery methods.
• Part III: we will explore the crucial aspects of market access, intellectual property, and investment opportunities in gene editing.

Together, these three parts provide a comprehensive picture of the current state and future potential of gene editing, offering valuable insights for investors.

DNA is the primordial information storing system and still the best

Humans take pride in their data storage systems, which have evolved from cave paintings and books to CDs, hard drives, and SSDs. However, the complexity of these systems often means they have a limited lifespan, with cave paintings lasting up to 70,000 years, books potentially surviving for millennia if preserved well, and CDs and hard drives lasting a few decades at best. Fortunately, life has a far more durable system for storing data: DNA.

Genomic information, encoded in DNA, holds billions of years of data that living organisms still use today. The amount of data that can be stored in DNA has increased exponentially, from the viral genomes of a few thousand base pairs (bp) to bacterial genomes containing a few million bp, and even some plants with close to 100 Giga bp. This represents an almost one billion-fold increase in storage capacity with only a marginal increase in volume.

The largest capacity for a modern SSD sits currently at 100 terabytes. The same human hand that could hold that SSD,  contains enough cells to cumulate a staggering 150mn terabytes of genetic information. Given these capabilities, no wonder companies like Twist Biosciences are now exploring the use of DNA as a data storage medium. 

To store information, it necessary to be able to write it, read it and make sure there are no mistakes in the process. But solving these issues, opens up the way to more significant usages for gene editing technology. 

The DNA is not easily messed up with

A genome with too many sequence errors cannot sustain a living cell. However, errors can still exist as long as they are not lethal. The main reason is that the genomic material within each cell cannot check its library of genes against another cell; therefore, no cell knows if its information actually carries an error or not.

To prevent errors from becoming harmful (and ideally from happening at all), evolution has selected strong packaging and repair mechanisms to limit access and potentially mitigate damage. Strong packaging systems and repair mechanisms mean that DNA data is often in “Read-only mode”.

To make matters even more complicated, cells have internal detectors for the presence of genetic material outside of the nucleus. As no human cell exchanges genetic material, the presence of genetic material outside the nucleus is under strict control, as it could be foreign DNA from bacteria, parasites, or viruses. No cell wants its machinery hijacked by pathogens. So, to engineer genomes with biological tools, one needs to craft them and deliver them very carefully.

The gene editing tools are a marvel of biological engineering

Simply put, there are two approaches to engineering a biological system: combining parts that do not occur naturally or repurposing an existing biological mechanism.

Four primary gene editing tools are available: TALEN, MegaTAL, ZFNs, and CRISPR. The first three fall into the former category, while CRISPR belongs to the latter.

TALE and ZF are systems that recognize DNA patterns but only bind to them under normal circumstances. By combining these systems with an enzyme from the family of “N”ucleases, which can cut genetic material, scientists built genetic engineering tools that bind and cut DNA, hence the name TALEN and ZFN. MegaTALs are based on another DNA-binding protein called the Meganuclease. Meganucleases are naturally occurring enzymes with a longer DNA-binding domain than TALENs, allowing them to recognize longer DNA sequences with greater specificity. However, MegaTals are relatively new and have not been used as extensively in research compared to TALENs, and are virtually inexistent in clinical pipelines. 

On the other hand, CRISPR is a bacterial immune system that identifies and cuts DNA sequences in viruses. The DNA sequence recognition occurs via a specific RNA piece, called gRNA for guide RNA, which can be designed and customized to target any DNA sequence. Therefore, CRISPR is known as the "magic" DNA scissor for its high level of specificity.

These systems, including CRISPR, have been turned into human-ready genome editing tools since 2005, and within the past decade, they have entered clinical trials.

All four systems are incredibly precise DNA scissors.

All four systems bind to a sequence of ~20bp on the DNA and cut it. Knowing there are four letters in the DNA, a sequence of 20bp would lead to over 10¹²  DNA sequence combinations. Given today's knowledge, even with millions of organisms having different DNAs, 20bp is enough to recognize any unique sequence in any genome in any species. 

To our great despair, nothing is 100% reliable in biology. The enzymatic activity of the four systems is sometimes cutting at unwanted places, called off-target effects. Scientists are working to limit these off-target effects. A wrong cut in the wrong gene could lead to a terrible genetic disease or cancer. Clinical trials need to be particularly mindful of these effects, and adapt their procedures accordingly.

CRISPR technology is leading the pack and is going beyond the cut

CRISPR 1.0 has taken the lead in terms of engineered versions, especially due to its incredibly simple targeting system: the single guide RNA. By synthesizing the desired guide RNA one could cut any sequence in any genome. TALEN and ZFN, despite their versatility and efficiency, are clunkier and more difficult to use, therefore losing ground to CRISPR systems. Champions of this tech are Intellia and CRISPR therapeutics.

CRISPR 2.0, called base editing, on the other hand, does not aim at cutting the DNA to delete an error but at replacing a defective base pair with the right one. This is because gene products can be activators or inhibitors; depending on the disease, one might want to mess up or restore the gene sequence. To that effect, two base editors have been developed: cytidine base editors (CBE) allowing C to T conversions and adenine base editors (ABE) allowing A to G conversions. Not cutting but replacing leads to less stress for the cell, which should limit off-target effects and deleterious consequences. Companies such as Beam or Verve lead the pack in the base-editing market. In particular, Verve got some spotlights last year for launching the first in vivo base editing clinical trial.

Engineering prowess never stops, and a seminal paper of 2019 showed another way of using the CRISPR system: prime editing or CRISPR 3.0. Within the skinny 2022 IPO list of biotechs, one could not have missed Prime Medicine, which raised $175 million and is now valued at ~$1.2bn. Prime editing can correct and insert "genetic text" at multiple points in the genome of a cell at once, hence its nickname of “word processor”. Despite an in-silico claim to be able to correct 89% of all genetic diseases, the issue stays the same; the delivery has not been worked out yet, and they are even further from clinics than all the others. But among the few names at the forefront of gene editing, Prime Medicine deserves to be kept under a watchful eye.

First target: Ex-vivo gene editing

The idea of engineering cell ex-vivo is not new per say

The birth of cell therapy can be traced back to the mid-20th century when scientists first began exploring the potential of using cells to treat diseases. The first successful bone marrow transplant was performed in 1956, marking the beginning of a new era in medicine.

Over the next few decades, researchers made breakthroughs in cell-based therapies, particularly with modified T cells. As a reminder, T cells are a type of white blood cell that play a crucial role in the immune system's response to infections and diseases. They are produced in the bone marrow and mature in the thymus gland, from where they get their name.

It was during the 80s that researchers first began exploring the potential of genetically modifying T cells ex-vivo to improve their anti-tumor activity and re-inject them. These efforts culminated in 2002 when a group of researchers at the University of Pennsylvania developed the first Chimeric Antigen Receptor (CAR)-T cells targeting CD19, a protein found on the surface of B cells, another immune cell type present in the blood. The researchers showed that these cells could eradicate leukemic cells in mice, and subsequent clinical trials in humans also showed promising results. This breakthrough led to the development of other CAR-T cell therapies targeting different types of cancer, including lymphoma and leukemia.

In 2017, the FDA approved the first CAR-T cell therapy for treating pediatric acute lymphoblastic leukemia, marking a significant milestone in cell-based therapies. There are now 6 CAR-T cell therapies on the market. With sales revenue up >70-100% yearly, the field is attracting numerous players.

But none of the approved CAR-T therapies are based on gene editing. They involve gene transfer techniques, which are forms of gene therapy, but do not do gene editing. This has led to many roadblocks along the way.

Moving from gene-modified to gene-edited cell therapies

Challenges, including manufacturing, treatment delays, toxicity, heterogenous durability, and efficacy on solid tumors have limited cell therapy's current use to later lines of treatments. Today the best manufacturing processes have a ~20% defects rate. Given the manufacturing cost is in the $100k per batch, such a high defect rate alone is a significant hurdle.

These challenges come from two sides: manufacturing and gene-modifying techniques.

Regarding manufacturing, there are two primary approaches for cell therapies: autologous, which uses cells derived from the individual patient, and allogeneic, which uses engineered cells derived from healthy donor cells.

Currently, only autologous cell therapies have been approved. Autologous cell therapy involves 1) isolating immune cells, 2) genetically modifying the cells ex vivo 3) allowing the manipulated cells to proliferate, and 4) infusing the cells back into the patient. This is a time-consuming and expensive process, and there may also be limitations on the number and quality of cells that can be collected. In addition, autologous cell therapy may not be effective in cases where the patient's cells are damaged or diseased. Most patients still have to wait 4-6 weeks from the sampling of their cells to receiving the treatment, which is often too late for cancer patients on their last line of treatments.

Allogeneic cell therapy involves 1) isolating immune cells from donors 2) genetically modifying  the cell ex vivo 3) proliferating the cell in industrial batch size 4) injecting the cells in any patient in need. The main difference is in the source and scale. Some go even further with truly “off-the-shelf” approach by taking generic stem cells from cell banks and differentiating them into immune cells. The idea behind the allogeneic approach is that the odds of reaching 100% correct batch manufacturing and faster time to treatment are much higher. The field is spearheaded by Allogene and CRISPR Therapeutics. Both have reduced by a factor of up to 10 the time to injection for patients in clinical trials. 

If the allogeneic approach might not beat the efficacy of autologous ("if you want something done right, do it yourself" i.e. with your own cells), the sheer production capacity is more than making up for it. Today no autologous-approved therapy is serving its market correctly due to delays and defects in production. An issue the allogeneic player will not face.

Regarding gene-modifying technics: all companies currently on the market use viral gene transfer. However, gene editing tools in an allogeneic setting could become the gold standard, with cells precisely edited and standardized from a donor source. In other words, it could overcome both hurdles of cell therapy: fast delivery and reproducible manufacturing. We will describe next the market these players want to address.

Blood cancers are the primary targets, followed by hereditary hematologic diseases

All blood cells have a common lineage, and thus blood cancer cells as well, making it a well defined cell population and an easier target. Also, immune cells are located in the blood, and genetically modified ones are similarly re-injected in the blood. It is therefore obvious the target of choice for cell therapy are blood cancers.

The blood cancer drugs market is expected to grow to ~$80bn in 2027 at a 10% CAGR. Notably driven by better diagnostics and demographic expansion for five major blood cancers: Acute Lymphocytic Leukemia (ALL), B-cell non-Hodgkin Lymphoma (B-NHL), Acute Myeloid Leukemia (AML), Multiple Myeloma (MM), and Chronic Lymphocytic Leukemia (CLL). The Multiple Myeloma market alone is expected to reach $17bn by 2027 (CAGR 11.2%).

Unsurprisingly cell therapy players have crowded the market with more than 800 cell therapy prospects since 2017. Given incredible efficacy, they are likely to take the biggest share of the market. The market is currently led by Novartis and Gilead, and a player like Legend Biotech, with results in MM expected this year, is in the spotlights. All previously mentioned companies manufacture autologous cell therapies that are virally modified. Despite that, intense competition from allogeneic players based on gene-editing technics is mounting up:

• CRISPR 1.0: CRISRP Therapeutics, Intellia, Editas.
• CRISPR 2.0: Beam
• TALEN: Allogene and Cellectis. 

A big focus is also on two main hereditary hematologic diseases: Hemophilia and Sickle Cell Disease. Both indications are unmet medical needs with high costs for the healthcare system; a single patient with Hemophilia in the U.S. will cost ~$6.4mn over his lifetime in hospital care and blood transfusion. Hemophilia is expected to reach $18bn in 2027 (7% CAGR) and Sickle Cell Disease $6bn (6% CAGR). Both indications result from mutations in a single gene and are perfect targets for gene editing players. If CSL addressed hemophilia first with viral gene therapy, all eyes are on CRISPR Therapeutics in partnership with Vertex. They should deliver the first Phase III results with CRISPR-edited cell therapy for Sickle Cell Disease. It is one of 2023 most anticipated events in the space.

T cells are not the only genetically edited cell therapy around

Next to T cells, we find Natural Killer (NK) cells and Macrophages, any Lymphocytes. All four cell types are subtypes of white blood cells. CAR-T cells have revolutionized the cellular therapy landscape by showing impressive efficacy in relapsed / refractory blood cancers. They are not the only type of cells being genetically engineered.  Three of the five major cell types used in cell therapy are genetically modified. The two unmodified ones are Tumor Infiltrating Lymphocyte and stem cells, and the three modified ones are T cells, Macrophages, and NK.

Although modified Macrophages are still early in development, modified NK cell therapy players have had ups and downs. In January, Fate Therapeutics, the posterchild of NK cell therapy, lost its $3bn deal with Johnson & Johnson. As opposed to Gilead paying $225 million upfront and making an additional $100mn equity investment in Arcellx last December. Only Editas has a gene editing program for NK.

These fields could benefit from gene editing tools for simpler and faster development. We are expecting more players to enter the space with gene editors.

From the patient to the shelf

Hospitals have been pushing for Point-of-care production for autologous therapy, while the industry and Contract Development and Manufacturing Organizations (CDMOs) are pushing for allogeneic cell therapy.

However, decentralized manufacturing in hospitals is quite challenging since all centers should use the same manufacturing protocol, the same or comparable in-process and lot release assays, and the quality programs from each center must work closely together. Consequently, manufacturing cellular therapies using a decentralized model is more complicated than manufacturing cells in a single centralized facility, despite the advantage of cutting production and delivery time by proximal manufacturing.

2022 has seen a flurry of news from CDMOs pushing single manufacturing sites. For example, Catalent acquired the manufacturing facilities of Erytech. The CDMO operations of Novartis signed a cell therapy production deal with Carisma. AGC Biologics is likewise expanding its production capacities.

The holy grail: In-vivo gene editing

The biggest opportunity is here

In our previous installment, we highlighted a significant challenge in the healthcare industry: there are over 6,000 single gene defects that can lead to debilitating diseases with no known cure, leaving only symptom management as a viable option. As mentioned above, advancements in gene editing offer promising possibilities in tackling these ailments and even extend beyond hereditary genetic disorders.

Most applications are in the Central Nervous System, Ophthalmology, Metabolic diseases, Cardiovascular, and Hematology. Another area where gene editing may prove especially effective is in treating certain types of solid tumors. By targeting specific genes or mutations associated with cancer, gene editing could potentially eliminate cancerous cells, sparing patients the toxic side effects of chemotherapy and radiation. Furthermore, gene editing could help boost the body's immune response to viruses and other pathogens, providing a powerful tool in the fight against infectious diseases.

Despite the immense potential of gene editing, the market for this technology is currently just 0.004% of the global drug market. This is partly due to the challenges surrounding the delivery of gene-editing therapies. Unlike traditional drugs, gene therapies require precise delivery of delicate cargo to the appropriate cells and tissues within the body - precise and delicate being the operative words. The treatment is useless in the wrong cell type or if the genetic sequence delivered is altered in the slightest.

The delivery mechanism is the central issue.

As stated in the previous installment, in laboratories, the gold standard is to use chemical and/or physical procedures to deliver genetic material to cell cultures to circumvent their innate immunity. Although highly efficient, they are not cell-specific, meaning one cannot target only muscle cells or neurons and therefore cannot be used on simpler living beings, let alone complex ones like us.

For living organisms, viruses were the first delivery vector used. Viral vector techniques use the innate aptitude of viruses to deliver genes directly into target cells. Unfortunately, viruses present a risk for marked immunogenic responses, insertional mutagenesis, and cytotoxicity - in layman’s terms, your immune system will overreact, or the virus will be inserted at the wrong place in your DNA increasing cancer risk, or it will trigger cell damage or death. All three risks have unfortunately been observed in various clinical trials with all virus types used.

As a response, the industry split two ways long ago: keeping improving viral vectors or developing non-viral vectors. One of the most famous non-viral vectors to deliver genetic medicine is lipid nanoparticles (LNPs). LNPs are small vesicles (30 to 200 nm in diameter) composed of fatty acids (lipids) capable of carrying and delivering a therapeutic load within our body. LNPs have been in the clinics, in approved drugs, or in diagnostics agents since the mid-90's but were limited to transporting small molecules. Quite a substantial tweaking was needed to accommodate genetic material. Worldwide attention was bestowed upon them in 2020 with the mRNA Covid19 vaccines from BioNtech and Moderna encapsulated in LNPs. Of course, those privy to biotech information know that the first LNPs encapsulated genetic drug was approved for Alnylam's ONPATTRO drug in 2018.  One can guess how hard it is to make LNPs encapsulated genetic drugs by looking at all the patent infringements trials ongoing in the space: Arbutus & Genevant vs Moderna, Moderna vs BioNtech, Alnylam vs Moderna and Pfizer, Acuitas v. Arbutus and Genevant, CureVac vs BioNtech and Pfizer. 

Today there are ~280 active clinical trials for viral vector‐based in vivo gene therapies and around ~30 for LNP. The race is on to find the best drug delivery vector. The CDMO market is very interested in both technologies, as the corresponding subsegments were valued respectively at $1.2bn in 2022 (CAGR 22% until 2028) and $5.4bn (CAGR 15% until 2028). 

Going beyond the liver is the frontier for LNPs and gene editing

As good as LNPs can be, and the recent partnership of Moderna with Generation Bio shows that everyone has an interest in them, specific organs in the human body are hard to target to. Currently, most clinical trials are for liver diseases with metabolic or cardiovascular consequences, but this is not coincidental. It is because the liver, given its biological functions, tends to concentrate any drugs circulating through the blood. So even systemic injections done by Intellia or Verve will mechanistically end up in the liver.

To overcome this challenge, one strategy is to target stationary cells that do not undergo division. This is why Editas focused on ophthalmology. However, their chosen treatment option only proved effective for a smaller-than-anticipated subset of patients. Despite this setback, there is still hope. Editas is currently seeking partnerships to develop treatments for severe eye diseases such as Leber Congenital Amaurosis 10 and Rhodopsin-Associated Autosomal Dominant Retinitis Pigmentosa. In addition, their gene-editing tools have enabled them to branch out into ex-vivo gene editing with NK cell therapy for the rest of their pipeline, demonstrating the versatility of their technology.

For viral delivery, the limiting factor is the payload size. Packaging the classic Streptococcus pyogenes (CRISRP/Cas9) and a gRNA together (~4.2k base pairs) into the most studied viral vector, called AAV, is challenging due to its size (shown on the graph above) limiting its max packaging capacity (~4.7k bp). Miniature versions of the CRISPR system are in development, leading to a more acceptable total size like the CasMINI system (~2.5k bp). The other gene editing tools are too big to fit in viral vectors. Nevertheless, even if technically possible, the issue remains the marked immunogenicity linked to the viral nature of the delivery system.

The risk/reward ratio for patients is hardly one-sided

The risk/reward ratio associated with gene editing for patients, even independently of the delivery mechanism, is hardly one sided. On the one hand, the potential rewards are significant, with the possibility of curing genetic diseases that have no known cure, reducing the need for lifelong treatments, improving the quality of life for patients, and saving billions in hospital bills. On the other hand, the risks of gene editing cannot be ignored, with potentially deadly unintended gene mutations, immune reactions, and toxicity. Proof of the sane level of skepticism is the number of clinical holds put on gene therapy as a whole and gene editing clinical trials, which as grow as fast as the industry.

The FDA and the Institute for Clinical and Economic Review know balancing risk, efficacy, and price will be tough to achieve. First-mover advantage is probably still the best hedge for companies because if the cure is successful, the need for competition heavily decreases, thus securing non-eroding revenues.

A word to conclude

To summarize this piece, gene editors’ technologies are still in their infancy both in terms of tools engineering and in terms of applications. Cell therapy remains the first market investors want to be exposed to, especially in allogeneic players, despite the risk of overcrowding which is part of the current competitive landscape. At the same time, the biggest opportunity lies in vivo gene editing. Still, its coming to maturity is further down the line and positions should be built slowly and carefully.

The next piece will cover pricing, regulations, market entrance, biosimilars, and several ways to invest in the field.

Catalysts

• More Cell and Gene Therapies drugs approved. With over 300 assets in development, every success will help better size the TAM.

• Advances in drug delivery. Better viral delivery and nanoparticles are the dream for targeted genetic therapies. Passing this hurdle will expand by a factor of 10 the market for numerous actors.

• Lower DNA/RNA synthesis price. With decreasing synthesis prices come more testing and cheaper production, benefiting both biotech and CDMOs.

Risks

• Clinical setbacks. No rewards come without risks. Today hype has calmed down with valuations cratering, but there is still room below if clinical failures happen.

• A precarious funding outlook. The debacle of SVB may have long-term negative repercussions on biotech funding.

• Ethical roadblocks. Today confined to some extreme cases of cancer or hereditary disease, gene editing is a tool able to reshape our approach to life. The ethical concerns around when modifications are too extensive or out of the medical field, e.g. plastic surgery, will surely slow advances.