Gene Therapy – Between Hope And Hype

Introduction

[Figure omitted: Photograph 51, the X-ray diffraction image of DNA. Caption: “The famous ‘photo 51’, elucidating the double-stranded, helical structure of DNA for the first time.”]

Over the past few years, you couldn’t tune into the biotech / pharma blogosphere without hearing about gene editing, CRISPR or RNAi on an almost daily basis. The hype surrounding CRISPR got picked up to a significant degree by the business and mainstream press and casual observers might think that we are about to enter a new age of medicine where just about any disease will not simply be treated but cured by these novel therapeutic approaches.

The hype machine has recently been kicked into overdrive by a 60 minutes episode which interviews some of the main protagonists behind the discovery of CRISPR.

[Figure omitted: a still from the 60 Minutes segment on CRISPR. Caption: “Source: CBS”]

As is often the case with trendy investment themes, there is genuine risk that vast exaggerations and hype will supersede what is realistically achievable in the near- and medium-term; at the same time, there is genuine promise in new approaches that interact and interfere with DNA and RNA one way or another. The purpose of this article is to equip readers with a solid foundation, a ‘primer’, on the underlying science and technology before exploring commercial and M&A dynamics in the sector.

Since the discovery of DNA mid-20th century (there had been prescient speculation even earlier), our understanding of genetics and epigenetics has advanced significantly. The Human Genome Project achieved (near-)complete sequencing of the human genome by the beginning of the 21st century; discovery of transcriptional inhibition by antisense RNA, the RNAi process and CRISPR have all contributed significantly to opening up of possible applications. Genotyping of bacterial strains is an important tool for researchers monitoring epidemiological trends. The identification of specific genetic mutations helps identify specific patient sub-populations, for instance children with CDKL5-related epilepsy or many different genetic factors involve with cancer.

Several therapeutic pathways emerge from the knowledge that a genetic mutation (‘polymorphism’) is in fact causing, or contributing to, a disease state.

• Direct modification of the underlying gene à la CRISPR may appear to the casual observer like the most straightforward treatment option but this ‘optimal’ solution is in reality hampered by all kinds of practical limitations, which we will address later.
• At present more feasible is post-transcriptional interference, that is, interference with messenger RNA (‘mRNA’)

It is intriguing to note that nearly our entire therapeutic spectrum comes ultimately down to interference with proteins: where classic small molecule and biologics therapeutics interfere with proteins directly (such as by binding to a specific epitope with a view on inhibiting protein function), both RNA interference and gene editing aim to quell the production of a specific protein altogether.

Furthermore, it is worth keeping in mind that many ‘old school’ therapeutics like certain antibiotics (such as aminoglycosides) and radiation therapy ultimately aim to disrupt DNA integrity or DNA transcription in diseased cells or in bacteria.

Having (hopefully) injected some realism into a complex debate, let’s move into a discussion of novel gene-targeted therapeutics, their limitations and challenges.

Post-transcriptional approaches

If you spend any amount of time on SeekingAlpha, bio-twitter or other online platforms prone to hype, chances are you’ve heard of mysterious entities dubbed ‘oligonucleotides’, ‘antisense’ and ‘RNAi’. It doesn’t help that some commentators appear to be themselves blissfully unaware of the meaning of these words they so readily throw around.

• Oligonucleotides are simply very short DNA or RNA fragments.
• ‘Antisense’ oligonucleotides are very short, single-stranded DNA or RNA fragments that are complimentary to a given DNA or RNA sequence.
• RNAi stands for RNA interference, a complex cascade of biological processes that ultimately inhibits gene expression. The distinction between ‘pure’ antisense-based approaches and RNAi, and the ensuing debate as to the merits of either approach, is clouded by a supremely annoying rivalry between two companies, Ionis Pharmaceuticals (IONS) and Alnylam (ALNY), and their respective fan-base. In my eyes, RNAi can be described as a more complicated implementation of ‘direct’ antisense inhibition and the pros and cons of both approaches need to be weighed on a case-by-case basis.

As a quick parenthesis, the CEO of Alnylam who happens to be the chairman of biotech industry association BIO, ranks near the top of my list for ‘overtly attention-seeking’ biotech executives. Head over to his twitter feed for an endless stream of vaguely inspirational superficiality. The otherwise desperately cynical Adam Feuerstein also happens to be favorably inclined towards Maraganore – perhaps Adam’s sheen of impartial criticism suddenly pales when dealing with the chairman of BIO.

Regardless of my reservations around Maraganore’s persona, ALNY is deserving of a dedicated write-up. At $10bn Mcap with little to show for commercially, my sense is that ALNY might make a decent short at present, but I would do readers a disservice if I skipped the required legwork before making such a recommendation. Let’s resume a broad discussion of this therapeutic space first.

Let’s dig a little deeper into what differentiates ‘pure antisense’ from RNAi.

“Antisense-RNA control functions in all three kingdoms of life—although the majority of examples are known from bacteria. In contrast, RNAi, gene silencing triggered by double-stranded RNA, the oldest and most ubiquitous antiviral system, is exclusively found in eukaryotes. Our current knowledge about occurrence, biological roles and mechanisms of action of antisense RNAs as well as the recent findings about involved genes/enzymes and the putative mechanism of RNAi are summarized.”

Source: Antisense-RNA regulation and RNA interference, Sabine Brantl

“RNAi is the name of the process (RNA interference). The RNA interference is a molecular process where an antisense RNA binds to a complementary target mRNA and drive the mRNA cleavage and degradation. This cause a gene expression silencing at a non-transcriptional but translational level. The action is mediated in the end by a nucleoproteic complex made by a single strand short antisense RNA and the RISC complex (RNA interference silencing complex). So RNAi is a generic name for this kind of non-genomic gene silencing mediated by different RNA species. It’s the name of the effect, not the name of the effector.

There are different kind of RNAs that can activate DICER/RISC complexes and trigger gene expression at translational level through RNAi mechanism and this is the reason why you can find long-dsRNA, miRNA, siRNA. All these RNAs finally work as a 19-20 nucleotides single strand antisense RNA attached to the RISC complex for sensing and triggering the right mRNA but they’re different at the begin.

The main difference is between miRNA and siRNA because the siRNA is a double strand short RNA, they’re mainly exogenous, for example a viral interference RNA that can aim the host gene regulation, but it can also can be designed experimentally and produced by in vitro transcription for knockdown experiments. Instead the miRNA is a single stranded RNA which fold himself after the synthesis into the nucleus by RNA pol. This stem-loop structure made by the same single strand RNA is cleaved at the level of the loop generating a more or less long double strand RNA derived by an initial single stranded RNA. It is then exported into the cytosol, cleaved and loaded into the same RISC complex. So miRNA is an endogenous alternative gene silencing process, there are a lot of known miRNA expressed in certain situations by cells for a fine gene-regulation.

The long dsRNA is the name of a long (200-500 nt) double strand RNA which is synthetized as a double strand RNA through in vitro transcription and then experimentally delivered into hosts cells for knockdown purposes. Once it is into the cells this long RNA is cleaved by DICER complex exactly like siRNA in many 19-20 nt fragments and these fragments are loaded into the RISC complex for sensing mRNA population searching for the target mRNA.”

Source

As you can tell, there is a significant degree of complexity involved with RNAi: it entails a cascade of events which involve different proteins and small strands of RNA working in concert to inhibit gene expression.

I’ve gone over a number of 3D animations which explain the RNAi process in preparation for this article. If you are keen on granularity, I recommend this animation by Nature:

[Figure omitted: embedded Nature video on the RNAi process.]

An easier-to-understand animation is provided by Alnylam:

[Figure omitted: embedded Alnylam animation of RNAi.]

Additionally, I can also recommend this Oxford animation on siRNA and miRNA, as long as you can get over the relatively poor narration:

[Figure omitted: embedded Oxford animation. Caption: “Generation and action of siRNAs and miRNAs”]

To sum it up, the centerpiece of the RNAi process is ‘RISC’, a multi-protein complex which incorporates a single strand from a micro RNA (‘miRNA’) or a small interfering RNA (‘siRNA’). In theory, all one has to do in order to use the RNAi process therapeutically is to introduce suitable miRNA or siRNA into the desired cells and RISC (as well as other processes) will take care of the rest.

Much the same holds true for the ‘antisense’ approach where, in theory, you ‘just’ need to get the antisense single-strand into the desired cell environment.

Of course, this is precisely where the limitations of these technologies – including CRISPR which we will discuss later on – for therapeutic use enter the picture. Some discussion of these considerations, including the ever-present risk of off-site binding, is provided towards the end of this highly recommended interview with Prof. Anna Krichesky of Harvard:

[Figure omitted: embedded interview with Prof. Anna Krichesky.]

Vehicles for delivery

This brings us to the all-important topic of how these RNA- and DNA-targeted therapeutics can best be delivered to their intended location. Invariably, the challenge is to get:

1. RNA-targeted therapeutics into the cytoplasm
2. DNA-targeted therapeutics into the cell nucleus. To do so in a lab setting is trivial; to do so in a highly localized fashion in humans, for instance by subretinal injection in the case of Spark Therapeutics’ (ONCE) Luxturna, is tricky but achievable; to do so systemically in humans is a serious challenge.

“Enthusiasm for nucleic acid technologies has waxed and waned; while they promise new avenues for addressing a host of targets deemed undruggable by traditional modalities, they have yet to enter the mainstream, in part due to the difficulty of delivering them efficiently to tissues beyond the liver, kidney and circulation.”

Source: Biocentury

[Figure omitted: chart of deal volume in post-transcriptional approaches over time. Caption: “Dealmaking relating to post-transcriptional approaches has picked up steam since 2013. Source: Biocentury”]

Viral vectors remain the go-to vehicle for this purpose, since researchers can leverage the innate processes which enable viruses to infect their host and replicate. Setting aside differences between different classes of viruses, the common denominator is release of DNA or RNA into the host cell alongside key enzymes such as transcriptase and integrase which catalyze insertion of the genetic material in the host cell. These processes, used by viruses for the purpose of self-replication, can be harvested for integration of DNA or RNA into human cells for therapeutic purposes.

[Figure omitted: ribbon rendering of an AAV2 capsid. Caption: “Ribbon illustration of a AAV2 viral capsid. Source”]

However, this is where basic PK/PD and immunological considerations come into play. The viral load required for systemic administration of gene therapy would be enormous and it would be non-trivial to:

1. quell an immune response from the host to the presence of this viral load
2. make sure that viral replication does not take place. There may be additional safety considerations which I am unaware of.

Last week, Merck and Moderna added to the deal Moderna’s mRNA-5671. The cancer vaccine encodes the four most common KRAS mutations. Merck will invest $125 million in Moderna’s series H round. The biotech closed a $500 million series G round in February.

Marett said modifications in nucleic acid chemistry that tune the stability and immunogenicity of RNA have opened the doors to partnerships. “You have chemical modifications that extend the messaging life of the mRNA so it can be transcribed in sufficient quantities to deliver its effect. That’s certainly one thing we’ve seen that perked interested in this space.”

According to Marett, partners have moved on to asking practical questions on issues like manufacturing scale-up for mRNA vaccines, but for chronic delivery to treat genetic disorders much of the conversation is still centered on PK/PD issues.

Source: Biocentury

Despite these concerns, viral vectors remain a vehicle of choice, as illustrated by the recent acquisition of AveXis (AVXS) and the therapeutic success of Spark Therapeutics’ (ONCE) Luxturna. We will discuss these case studies further down.

[Figure omitted: schematic of viral delivery feeding into the RNAi cascade. Caption: “An illustration that combines viral delivery of RNA with the downstream steps involved in the RNAi process. Source: UofT”]

Now that we’ve discussed antisense and RNAi, let’s have a look at the gene editing tools at modern scientists’ disposal & on the mind of every biotech investor.

Gene editing

[Figure omitted: chart of fund ownership in gene-editing names. Caption: “The popularity of gene editing players like CRSP is leading non-specialist funds to invest. Source: Barron’s”]

I will again refer to nature video for a 3D explanation of the CRISPR / cas9 mechanism: https://www.youtube.com/watch?v=4YKFw2KZA5o

If nothing else, you need to understand that CRISPR-Cas9 is akin to protein-based, biologic machinery able to seek out and cut genomic sequences with accuracy. That accuracy is not perfect however, which has serious implications with regards to the safety of any future therapeutic applications based on this particular gene editing mechanism.

Another approach to gene editing is based around zinc finger nucleases (‘ZFNs’) and Sangamo Therapeutics (SGMO) is a leader in this field. Much like the duality between ‘pure’ antisense and RNAi, the distinction between CRISPR-Cas9 and ZFNs and the competition between their respective developers drives a lot of online debate.

I haven’t formed a strong opinion about which approach should be preferred and in which scenario. For now, suffice to say that neither CRISPR-Cas9 nor ZFNs are flawless technologies and as usual in the medical domain, trade-offs will have to be accepted.

For deeper insights and necessary context on the hype surrounding CRISPR, I have found this lecture by Prof. Church supremely helpful:

[Figure omitted: embedded lecture by Prof. George Church.]

Regulatory considerations

Readers will be familiar with the ‘regular’ FDA pathways aimed at speeding up approval of important new treatments:

• Fast track
• Breakthrough Therapy
• Accelerated Approval
• Priority Review

Of course, all of these can potentially be applied to gene therapy. Of greater interest is the specially established RMAT– regenerative medicine advanced therapy designation – established by the 21st century cures act, not unlike LPAD which Achaogen (AKAO) are leveraging for the approval of Plazomicin. RMAT applies to therapies which satisfy the following conditions:

• The drug is a regenerative medicine therapy, which is defined as a cell therapy, therapeutic tissue engineering product, human cell and tissue product, or any combination product using such therapies or products, except for those regulated solely under Section 361 of the Public Health Service Act and part 1271 of Title 21, Code of Federal Regulations;
• The drug is intended to treat, modify, reverse, or cure a serious or life-threatening disease or condition; and
• Preliminary clinical evidence indicates that the drug has the potential to address unmet medical needs for such disease or condition

Sponsors of eligible therapies benefit from earlier interactions with the FDA (e.g. for trial design discussions) and the therapy obtains priority review.

To date, 16 RMAT designations have been awarded and 40 applications have been received, demonstrating the attractiveness of this relatively recent designation.

In Japan, legal ‘acts’ have provided similar provisions with regards to regenerative medicines. For instance, AveXis has obtained “Sakigake designation from Japan’s MHLW” for its lead product.

Companies, both public and private, awarded RMAT designations in 2017:

1. Asteria Biotherapeutics
2. Athersys
3. Bluebird Bio
4. Cellvation
5. Humacyte
6. Enzyvant
7. jCyte
8. Juno therapeutics
9. Kiadis

…and for 2018:

1. Abeona
2. Capricor
3. MiMedx

Gene therapies currently pending approval:

1. LentiGlobin (bluebird bio)
2. BMN 270 (Biomarin)
3. Lenti-D (bluebird bio, Orphan drug)
4. JCAR017 (Juno/Celgene)
5. Bb2121 (bluebird bio/Celgene, Breakthrough Therapy)

A list of gene and cell-based therapies approved around the world to date:

1. Gendicine - approved in China
2. Glybera - most expensive therapy in the world at >$1m / treatment, discontinued)
3. Imlygic – not exactly ‘gene therapy’ but rather an oncolytic virus used in unresectable melanoma. Cf. Forbes article
4. Invossa
5. Kymriah – Novartis (NVS) first-to-market Car-T therapy, cost of therapy $475k
6. Luxturna – the previously mentioned therapy for a type of genetically caused blindness, developed by Spark Therapeutics (ONCE) and partnered with Novartis (NVS)
7. Neovasculgen – approved in Russia for peripheral arterial disease
8. Oncorine – oncolytic virus approved in China
9. Rexin G – approved in the Philippines

Chinese, Russian and Philippine approvals are not as confidence-inspiring as FDA approvals, but they go to show that gene- and cell-based therapies are making strides around the globe. More flexible regulatory standards outside the U.S. help accelerate such developments. Readers should also note that the term gene therapy is used broadly as an umbrella term which includes Car-T therapy, in which immune cells from a patient are removed, genetically modified and then re-injected.

Deal-making

To get a sense for what sort of gene therapy assets and platforms are en vogue with Big Pharma decisionmakers, let’s consider some of the most meaningful deals of late.

Arguably, Novartis (NVS) are establishing themselves as a key player in a more classical field of gene therapy: viral-vector gene transfection. Their recent acquisition AVXS has pushed adeno-associated virus 9 as an ideal vector for gene therapy delivery in the CNS. NVS have also partnered with Spark Therapeutics (ONCE) on Luxturna.

However, a major step toward the goal of systemic gene delivery to the CNS using rAAV may have been taken by Foust and colleagues in their recent report. 5 In this exciting study, peripheral intravascular (i.v.) administration of rAAV9 in neonatal or adult mice resulted in a widespread transduction of either the spinal cord or the brain. In addition, transduction of non-neuronal cells in the CNS by rAAV has historically proven to be extremely limited. However, in this study, depending on the age of the injected animal, a significant portion of non-neuronal cells in the brain and the spinal cord were infected. Although previous studies have shown that intraventricular injections of rAAV1 or rAAV2 in newborn mice resulted in widespread transduction of the brain, 18, 19 no indication of non-neuronal infection was seen in these earlier studies; thus, the findings by Foust et al. 5 could be of great clinical importance, in that this would expand the repertoire of rAAV delivery to other cell types in the CNS that have been inaccessible thus far. For the purposes of clinical strategies, the ability to transduce CNS cells across the BBB in the adult is the most promising part of the Foust et al. study, in that closure of the BBB to large proteins occurs much earlier in development in the human as opposed to the rodent. 3

Source: AAV9: a potential blood-brain barrier buster

NVS is rationalizing the deal in a presentation to shareholders by highlighting the potential for broader applications of AveXis’ AAV9-based approach. Versatility of a platform and potential synergies with the acquirer tend to be defining attributes of attractive acquisition targets.

Then there are of course the GILD & CELG bids to acquire Car-T behemoths KITE and JUNO, which can also be considered gene therapy players.

Some years back, Shire (SHPG) merged with Baxalta in a deal valued at $32bn (cash and shares) but later scrapped a gene therapy it inherited through the deal.

While the recent NVS / AVXS deal may suggest that M&A is picking up in the sector, in reality the vast majority of deals are in the form of partnerships.

[Figure omitted: chart of deal counts by company in the oligonucleotide sub-sector. Caption: “Leaders in the oligonucleotide sub-sector like Ionis and Moderna are serial dealmakers. Source: Biocentury”]

Preferring partnerships over acquisitions is congruent with a ‘wait and see’ approach towards novel technologies and treatment modalities. The relatively modest expenses incurrent by Big Biotech and Big Pharma when partnering with gene therapy players allow for diversification. Acquisitions in a field brimming with investor confidence tend to be expensive; case in point AveXis.

Among notable partnerships, let’s just consider the recent extension of Biogen’s (BIIB) collaboration with Ionis (IONS), which builds upon the successful launch of spinal muscular atrophy treatment Spinraza. While the deal may appear rich, and it is certainly motivated by BIIB’s drive to rejuvenate its own pipeline, the prior commercial success with Spinraza makes it much easier for BIIB to justify this investment to its shareholders.

Closing remarks - stay tuned

There is no denying that gene therapy is achieving some remarkable therapeutic milestones - helping blind patients regain sight is no small feat. The endless possibilities afforded by CRISPR shouldn’t be overlooked either. From the investors’ point of view however, it can be difficult to find value in a decidedly ‘hot’ field of biotech. This introduction is just that, a first step towards gaining a more comprehensive understanding of an exciting but slightly overhyped sub-sector in biotech. I intend to follow-up with company-specific considerations and hopefully one or the other plausible investment opportunity. For the time being, I would suggest readers interested in surfing the gene therapy wave ought to diversify rather than betting on a single company or platform. Among the smaller players, I like DiceRNA (DRNA), which enjoys backing by Bain Capital and which recently settled an important legal dispute with Alnylam on what appear to be favorable terms. Given the stock’s rapid appreciation, in keeping with other players in the field, I would advise cautionary and paced out buying at this point (if any).

Disclosure: I am/we are long AKAO. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.