Competing in gene therapy is unlikely to follow the same rules as competing in other therapeutic categories where — even in Orphan diseases — a better product can steal market share. The reason is that if the first gene therapies to market deliver on the promise of a functional cure, there may not be any patients left to treat. Investors have poured at least $3 billion into 20 companies developing gene therapies for Orphan indications that have one or more competitors working on the same gene — including $1 billion to bluebird bio Inc. alone. Most of these companies have touted potential benefits of their products over more advanced competitors targeting the same genes.

 

But in Orphan indications, if the first-to-market therapy addresses a large proportion of the prevalent patient population, it could be difficult for followers to even enroll clinical trials. And it is unknown whether the risk of immunogenicity to the vector or the transgene protein product would preclude re-treating patients who received a first-generation product. “If there is an approved gene therapy that is providing a functional cure, I think it will be hard for some of these companies to enroll trials,” Spark Therapeutics Inc. CFO Stephen Webster told BioCentury.

 

However, when the gene therapy delivers a therapeutic protein that doesn’t address the underlying pathology of the disease — and therefore isn’t a cure — improvements in efficacy or safety can potentially supplant first-in-class therapies.

 

And even in Orphan indications, there could be room for followers with new vectors if the first to market uses a vector to which a large enough proportion of the population has a pre- existing, natural immunity.

 

For companies pursuing larger indications, the rules of competition should be similar to those in other drug classes. Here, followers may be able to compete based on advances in vectors, gene expression levels, gene selection, administration or using more advanced technologies like CRISPR. The 30 gene therapy companies that are focused on large indications have received at least $3 billion since inception.

 

WINNER TAKES ALL?

 

Most gene therapy companies have shied away from competing with each other in Orphan diseases. According to BioCentury’s BCIQ database, out of 166 gene therapy products in development, 133 (80%) are being developed for indications where there is no competition, are using different genes than competitors for the same indication or are intended for a patient population large enough to support multiple players.

 

The other 33 products (20%), however, are in competitive Orphan indications (see “Crowding the Gene Pool”). According to Venrock’s Bong Koh, whether a second-in-class gene therapy will have a market will likely depend on how big a head starts the first in-class product has. “How quickly can you rapidly identify and penetrate the patients and cure them? My guess is if you have a one-year lead, it probably doesn’t mean all that much. But on the other end of the spectrum a 10-year lead probably means a lot,” said Koh. He added that for indications where there is a rapid decay due to disease, such as in spinal muscular atrophy (SMA), penetration is likely to be faster, potentially leaving a smaller opportunity for follow-on therapies. Koh has invested in four gene therapy companies, of which three have a chance at being first to market or are working in indications with room for competition: Audentes Therapeutics Inc., AveXis Inc. and RegenxBio Inc.

 

Audentes’ lead program is AT001, an adenoassociated viral serotype 8 (AAV8) vector encoding the myotubularin 1 (MTM1) gene that is in preclinical testing to treat X-linked myotubular myopathy. AveXis has chariSMA, a self-complementary AAV9 vector encoding the survival motor neuron (SMN) gene that is in Phase I testing for SMA. RegenxBio’s lead program is RGX- 501, an AAV8 vector encoding the low-density lipoprotein receptor (LDLR) that is expected to enter the clinic in 1H16 to treat homozygous familial hypercholesterolemia (HoFH). Koh’s fourth investment, Celladon Corp., was working in heart failure but has ceased R&D and is looking to sell itself or its assets, including Mydicar gene therapy, which failed a Phase IIb trial in April. Venrock’s one gene therapy investment that is in a competitive Orphan indication is Avalanche Biotechnologies Inc. The biotech’s AVA-311, an optimized AAV that encodes the retinoschisis X-linked juvenile 1 (RS1; XLRS1) gene, is in preclinical development to treat X-linked juvenile retinoschisis and is partnered with Regeneron Pharmaceuticals Inc. Applied Genetic Technologies Corp. (AGTC) is ahead with its XLRS, a recombinant AAV encoding RS1 that is in a Phase I/II trial for the indication. Koh wouldn’t discuss how Avalanche might compete as a follow-on gene therapy in X-linked juvenile retinoschisis. However, the biotech’s lead program is AVA-101, an AAV that encodes soluble VEGF receptor 1 (sFLT1; sVEGFR-1). The product has completed a Phase IIa trial for wet age-related macular degeneration (AMD) and is in preclinical testing for diabetic macular edema (DME) and retinal vein occlusion (RVO).

 

UNKNOWN TERRITORY

 

While second-to-market drugs in other therapeutic modalities might hope to treat patients who had failed the first-generation drug, it isn’t clear whether that will be possible with gene therapies. According to Webster, Spark has decided not to test re-dosing patients with its gene therapies due to the risk of an immune response against the protein product expressed by the transgene. “The fear scientifically is that if there is an immune response to the second dose, you could lose the benefit that the first dose conferred,” he said.

 

However, AGTC President and CEO Sue Washer said the risk of immunogenicity in re-treating human patients is actually unknown, and concerns are primarily derived from non-human primate studies, not clinical trials. Still, until more data are gathered on the risks of immunogenicity, follow-on therapies may be limited to gene therapy-naïve patients. Webster added that patients may be reluctant to enroll in a clinical trial of a new gene therapy if the risk of immunogenicity might preclude them from being re-treated with an approved gene therapy. “The informed consent for that will be brutal,” he said. “Who is going to volunteer for a clinical trial? Why would you chance an experimental therapy that may or may not be marginally better instead of receiving one that is approved?” Most gene therapy company executives who spoke to BioCentury said they think an improved duration of response compared with a more advanced product could allow a follower to compete. However, if the first product to market has a long enough head start and a duration of response lasting several years — even if not a lifetime — demonstrating an improvement could be a practical impossibility. uniQure N.V. CEO Jörn Aldag noted his company’s lipoprotein lipase deficiency product Glybera alipogene tiparvovec has demonstrated a duration of response of five years — so far. By the time another company could generate five years of response data, “we may have shown a benefit that lasts 10 years. So, you haven’t proven anything if you wait five years in your clinical trial. Proving superior durability of response will be very difficult,” Aldag said.

 

THE LCA EXAMPLE

 

Of course, in some Orphan indications, the competition may shake itself out before any products get to market due to natural attrition, or as companies reprioritize their investments. The latter appears to be the case in Leber congenital amaurosis (LCA), a form of retinitis pigmentosa that has about 3,500 patients in the U.S. and Europe. Three gene therapies that express the retinal pigment epithelium- specific protein 65kDa (RPE65) gene have been in development for the indication. Spark’s SPK-RPE65, an AAV2 encoding RPE65, is in Phase III testing to treat LCA, with data expected this year.

 

The other two completed Phase I/II trials, but neither is in active development: rAAV2-CBrhRPE65 from AGTC and HORA-RPE65 from Horama S.A.S. According to Washer, AGTC decided to discontinue its product based on a combination of factors, including the ultra-Orphan nature of the indication, difficulties in identifying patients and the degenerative nature of the disease pathology.

 

“We decided there were other ophthalmic discases better suited for us to pursue even before Spark existed as a company,” she said.

 

Horama, however, decided to deprioritize its RPE65 product because Spark was so much further ahead. Horama is seeking a development partner for HORA-RPE65 and is instead focusing its internal efforts on two ophthalmic gene therapies that have no competitors. Spark’s co-founder, President and CSO Katherine High noted, “The question I’m wondering about is if there are over 200 genes involved in vision and many of them fit into an AAV vector, why would someone go after something where somebody else already had a product in Phase III?” Koh agreed. “You have to ask yourself how big is LCA really? My understanding is other gene therapy companies have looked at that indication and passed on it because of its size,” he said. He added that with 7,000 rare diseases, small gene therapy plays may be better off focusing on untapped indications.

 

ROOM TO IMPROVE

 

In some indications a gene therapy might provide a significant benefit to patients but not a functional cure, potentially leaving room for follow-on therapies to provide improved efficacy. Hemophilia B, the most competitive of the Orphan diseases for gene therapy, may be an example. The less common of the two forms of hemophilia has a prevalence of about 4,000 patients in the U.S. and 80,000 worldwide. There are at least six companies and one academic group developing gene therapies for hemophilia B. Three compounds are either in or about to start Phase I/II testing: Baxalta Inc.’s BAX 335, Spark and Pfizer Inc.’s SPK-FIX and uniQure’s AMT-060. St. Jude Children’s Research Hospital and University College London have scAAV2/8- LP1-hFIXco in a Phase I trial. Two others are in preclinical development, including DTX101 from Dimension Therapeutics Inc. and Sangamo BioSciences Inc.’s zinc finger DNA-binding protein nuclease (ZFN) therapeutic. Despite the small patient population, given differences in the degrees of efficacy that could be attained in hemophilia, executives felt it could be possible for a follower to replace a first-in class product. Whether there would be a large enough patient population left to provide a return on investment for a follow-on gene therapy would depend on the uptake of the first-in-class therapy. Increased efficacy is part of the rationale behind Spark’s decision to use the Padua variant of Factor IX in its SPK-FIX therapy. The variant has 5-10 times the activity of wild-type Factor IX, meaning a similar expression level of the gene variant could significantly improve efficacy. According to High, 5% expression levels would convert severe hemophilia to mild hemophilia, which would provide a benefit to patients. Expression levels close to 50% would bring a patient within the normal limits for Factor IX expression. But High noted it’s not clear how much of a gain within the 5-50% range would be meaningful. “Is 10% expression clinically better than 5%? That would be debatable,” she said. There are other indications where there could be an opportunity to meaningfully improve efficacy. The question is whether it would be as straightforward to demonstrate efficacy in those indications as it is in hemophilia, where clotting factor is easy to measure.

 

For example, in retinopathies, companies can’t simply measure expression levels of the protein of interest. “You have to look at downstream effects of it on various visual and retinal functional measures,” said High. Koh said he expects there will be both big and small indications where better expression or better vectors will make a difference for follow-on gene therapies. But he added, “there are going to be quite a few indications where it just doesn’t matter. The expression the first product has is good enough and provides a functional cure, and whatever sort of side effects exist are just going to be well managed.”

 

DIFFERENTIATING FACTORS

 

Development of non-immunogenic vectors could allow follow-on therapies to gain market share in either large or Orphan indications. Many patients have a natural, pre-existing immunity to some of the most common AAV vectors. For instance, Aldag said as much as 40% of the population has a natural immunity against AAV8, and patients can be screened for the presence of neutralizing antibodies against a relevant AAV serotype. In that example, if the first product to market uses an AAV8 vector, a sizable portion of the patient population would be ineligible for treatment, and better suited for a lentiviral vector or an AAV with lower rates of pre- existing immunity. “If someone comes to market first with an AAV8 hemophilia product, there is still a significant market left for someone with an AAV5 for example,” Aldag said. Both Spark’s AAV8-hFIX19 and Baxalta’s BAX 335 use an AAV8 vector. uniQure’s AMT-060 uses an AAV5 vector. Oxford BioMedica plc CEO John Dawson agreed, noting that lentiviral vector-based gene therapies would provide a potential alternative for patients with pre-existing immunity to AAV vectors. However, such opportunities would likely be limited to diseases where there is systemic exposure to the gene therapy. High noted pre-existing immunity against AAV is less of a concern in ophthalmology, because the eye is immuno-privileged.

 

DIVIDE AND CONQUER

 

In larger indications like congestive heart failure (CHF) or Parkinson’s disease (PD), companies see enough patients to support multiple gene therapies and expect the competitive paradigm would be similar to other therapeutic categories. Thus, technologies that increase efficacy, such as self-complementary transgenes that have higher expression profiles, or approaches that improve safety, such as vectors that can better target the tissue of interest, could provide differentiation for a follow-on product. These large indications are also typically diseases in which gene therapies are used as delivery vehicles for a protein product that may provide a clinical benefit but does not address the underlying pathology of disease and therefore does not represent a cure. For instance, in Parkinson’s there are at least five companies working on gene therapies that aim to stimulate the production of dopamine or other neurotrophic factors. Wet AMD is another crowded indication, with seven companies developing gene therapies that encode a VEGF inhibitor or some other antiangiogenic protein. Both diseases have populations estimated at more than 1 million patients in the U.S. alone.

 

BEYOND ‘FUNCTIONAL’

 

In the long term, most gene therapy companies said they were looking beyond the current generation of gene therapy — in which a functional gene is simply added to the cell alongside the mutated gene — to gene editing technologies like CRISPR that could instead repair the endogenous gene. According to Fulvio Mavilio, scientific director at Genethon, the next wave of gene therapies will be the in vivo application of gene editing technologies. “Gene editing is the next generation, because we don’t add things, we fix things,” he said. Genethon’s lead gene therapy uses autologous CD34-positive cells transduced with a lentiviral vector encoding human Wiskott-Aldrich syndrome gene to treat Wiskott-Aldrich syndrome. The next wave may be coming sooner than most expected. While gene editing technologies are already in man in ex vivo settings such as in CAR T cell therapies, the problem for in vivo applications has typically been that the CRISPR construct has been too large to fit into a delivery vector. Abeona Therapeutics Inc. CEO Tim Miller said his company is working on a modified CRISPR construct that can be delivered using an AAV vector, and the company expects to complete preclinical proof-of-concept studies within the next year. Miller said the AAV-delivered CRISPR program will first target the rare blood disorder Fanconi anemia.

 

Stephen Hansen

Associate Editor, BioCentury