Manufacturing Gene Therapy

The Food and Drug Administration cleared the first two gene therapies for inherited diseases in short order, with just a year and a half separating historic approvals for the blindness treatment Luxturna and the spinal muscular atrophy drug Zolgensma.

Since the May 2019 OK for Zolgensma, however, two years have now passed without an addition to the ranks of approved gene therapies, despite a broad and growing pipeline. A spate of regulatory setbacks last year, including a surprise rejection for BioMarin Pharmaceutical's hemophilia treatment Roctavian, have spurred questions about whether the agency is enforcing a tougher line on new candidates.

One sign of increasing scrutiny is delays involving cell and gene therapy manufacturing. Last fall, Sarepta Therapeutics, Voyager Therapeutics, Iovance Biotherapeutics and Bluebird bio were forced to reset development timelines as the FDA requested more information about production processes.

"It's in part due to the general evolution of the field," said Timothy Cripe, hematology and oncology division chief at Nationwide Children's Hospital and an early stage gene therapy researcher. "The technology is improving and the methodology is improving for measuring purity and potency and genome copy numbers," he added, referring to assessments of manufacturing quality.

"This is increasing [the FDA's] bar a bit, as would naturally occur," Cripe said.

Manufacturing-related delays, however, came alongside clinical trial pauses for Solid Biosciences and Astellas Pharma's Audentes division due to safety concerns, most notably the death of three participants in a trial of an Audentes gene therapy. Those pauses have since been lifted. 

"That also gives the agency pause. That gives us all pause. We're asking, 'Are we dosing the kids appropriately? Is dosing by kilogram appropriate?'" Cripe said.

FDA officials, for their part, said the delays and holds aren't related to any change in policy, but rather are a reflection of the agency's increasing workload as more gene therapies are tested in humans.

"I think what you're seeing is the increase in activity in the field," Wilson Bryan, director of FDA's Office of Tissues and Advanced Therapies, said in a session of the virtual Biopharma Congress held by the analyst group Prevision Policy.

"We have over 1,000 active [Investigational New Drug applications]," Bryan said. "As we get more and more active INDs more and more bad things are going to happen. What gets attention are the bad things."

That view is shared by Patricia Zettler, an associate law professor at Ohio State University and a former associate chief counsel at the FDA. "It's more likely that there just have been some bumps along the road," she said. "It would be more surprising if no gene therapies ever encountered those kinds of bumps in the road."

Some of the "bumps" may also be the product of gene therapies moving out of academic labs, where many of the ones furthest along were birthed, and into large-scale clinical or commercial production at big pharmaceutical companies. A data manipulation scandal over animal testing for Zolgensma, for example, was blamed on this type of handover.

Successful reviews of manufacturing data or sites, meanwhile, may not be the sort of event that companies disclose to their investors.

"If you're a company like BioMarin and you put out a press release that said, 'We had an FDA inspection and everything is fine,' investors would be calling up and asking what is wrong," said Jacob Sherkow, a University of Illinois professor who is an affiliate of the university's Carl R. Woese Institute for Genomic Biology.

Current setbacks aside, the two approvals and a review, albeit eventually rejected, of a third show an agency much more open to considering gene therapies than it was after safety scares two decades ago, most notably the death of teenager Jesse Gelsinger in a clinical trial. So too do new guidelines for developers and FDA officials' predictions of a near future with many more cell and gene therapies approved.

"[Gene therapy] was the third rail of FDA policies for a long time," Sherkow said. "They're a lot more open now. We have actual guidance about how to do this, how to do a clinical trial and what endpoints you should use."

Cripe, at Nationwide Children's, agreed the FDA's comfort with the risk-benefit balance of gene therapy has shifted. "With Zolgensma and with Luxturna and reports of successful trials in muscular dystrophy, the benefit part of that equation is becoming more clear," he said.

One variable potentially at play as more gene therapies go under FDA review is the relative urgency for approval when other therapies exist. Luxturna, for example, treats an otherwise incurable form of blindness, while, at the time of Zolgensma's approval for spinal muscular atrophy, only one alternative treatment existed for the disabling and often fatal neuromuscular disorder.

BioMarin's Roctavian, on the other hand, treats hemophilia A, a bleeding condition that while potentially severe has multiple treatments that can limit or prevent spontaneous bleeds in most patients.

Treatment availability could given FDA reviewers more leeway to carefully consider the risks and benefits of experimental gene therapies, including controversial questions like the durability of benefit.

While Bryan did not directly address the rejection of Roctavian at the Biopharma Congress, he said the agency wants gene therapies to represent substantial improvements over existing treatments, if they exist.

"Sometimes people look at gene therapy and [talk about] the advantages: it could be a one-time treatment versus the drugs that are out there that may require repeated administrations, or it might replace a more invasive administration," he said. "We don't take that into account so much."

More and more gene therapies for inherited diseases with few or no treatments are advancing, however, while second-generation therapies for more common diseases are under study as well.

"Cell and gene therapy," Bryan said, "is a growth industry."

Three years ago, the Food and Drug Administration granted a landmark approval to the first gene therapy for an inherited disease, clearing a blindness treatment called Luxturna.

Since then, the regulator has approved one more gene therapy, the spinal muscular atrophy treatment Zolgensma, and given a green light for dozens of biotech and pharmaceutical companies to start clinical testing on others. Genetic medicines for a range of diseases, including hemophilia, sickle cell and several muscular dystrophies, appear in reach, and new science is galvanizing research. 

But, entering 2021, the gene therapy field faced major questions after a series of regulatory and clinical setbacks have shaded optimism. "The ups and downs of adolescence are on full display" analysts at Piper Sandler wrote in September, summing up the state of gene therapy research.

Here are four questions facing scientists, drugmakers and investors this year. How they're answered will matter greatly to the patients and families holding out hope for one-time disease treatments. 

Are recent high-profile setbacks a warning sign? 

The FDA was widely expected last year to approve a closely watched gene therapy for hemophilia A, the more common type of the blood disease. Instead, the agency in August 2020 surprisingly rejected the treatment, called Roctavian, and asked its developer, BioMarin Pharmaceutical, to gather more data. 

The next day, Audentes Therapeutics reported a third clinical trial participant had died after receiving the biotech's experimental gene therapy for a rare neuromuscular disease. The tragedy brought flashbacks to past safety scares in gene therapy, although the current wave of treatments being tested have generally appeared safe. 

Less than five months later, the gene therapy field grappled with two more setbacks. UniQure reported a case of liver cancer in a study of its gene therapy for hemophilia B. And Sarepta, one of the sector's top developers, faced new doubts about its top treatment for Duchenne muscular dystrophy after disclosing a key study missed one of its main goals. 

In each case, the drugmakers involved offered explanations and reasons for optimism. BioMarin still expects to obtain an approval; Audentes' trial is now cleared by the FDA to resume; UniQure later gave evidence showing its gene therapy to be unlikely the cause of the cancer case; and Sarepta argued its negative data were the product of unlucky study design. 

But taken together, the developments are powerful reminders of both the stakes and uncertainty still facing gene therapy.

All four events also highlighted lingering worries about one-time genetic treatment. In rejecting Roctavian, for example, the FDA seemed to be concerned the impressive benefit hemophilia patients initially experienced may wane over time. The deaths in Audentes' study, meanwhile, renewed warnings about extremely high doses of gene therapy. Researchers have long watched for evidence that replacing or altering genes may cause cancer to develop in rare instances, particularly after four infants developed leukemia in a gene therapy study in the early 2000s. And Sarepta's negative findings were surprising because early signs of dramatic biological benefit that didn't seem to translate into clear-cut functional gains for all patients.

Experts are still confident gene therapy can deliver on its promise. But recent events suggest getting there may take a bit longer than some expected. 

Has the FDA raised its bar?

"The process is the product," is an often-used cliche about gene therapy, which are complex treatments with exacting manufacturing standards. 

Most of the roughly 60,000 pages in Spark Therapeutics' application for approval of Luxturna, for instance, involved what's known in the industry as "chemistry, manufacturing and controls." 

The therapeutic basis for gene therapy, by contrast, is much clearer for many of the rare, monogenic diseases that developers are targeting. If mutations in a single gene lead to disease, replacing or otherwise fixing that gene should have a large benefit. 

"Genetic medicine is not industrialized serendipity," said Gbola Amusa, an analyst at Chardan, contrasting gene therapy with chemical-based drugs. "It often is an engineering question."

In 2020, the FDA gave ample notice that it's watching gene (and cell) therapy manufacturing closely. Sarepta, Voyager Therapeutics, Iovance Biotherapeutics and Bluebird bio were all forced to revise their development timelines after the agency asked for new details about production processes. 

"The FDA is saying to companies that you've got to up your standards," Amusa added. 

For their part, FDA officials have indicated the spate of data requests are a product of the sharply higher numbers of companies advancing through clinical testing.

Can a wave of new startups develop better tools?

Tasked with replacing faulty genes with functional ones, scientists for the most part have turned to two types of viruses to safely shuttle genetic instructions into cells. Adeno-associated viruses, or AAVs, are typically used for infused treatments, while researchers working on cells extracted from patients generally opt for lentiviruses. 

Each virus class has advantages, but also notable drawbacks. AAVs, for instance, can trigger pre-existing immune defenses in some people, making those individuals ineligible or poor candidates for gene therapy. Lentiviruses, by contrast, are known to integrate their DNA directly into the genomes of cells they infect — a useful attribute in some regards but limiting in others. 

Over decades of gene therapy research, scientists have found ways to tweak and modify these viral vectors to better suit their needs, but the basic tools are the same. Jim Wilson, a gene therapy pioneer who ran the study that led to the death of teenager Jesse Gelsinger in 1999, told attendees at a STAT conference last fall that he's "somewhat disappointed" by slow progress in viral vector research. 

And as more and more gene therapies enter clinical testing, the limitations of current viral vectors have become more apparent. 

The pace of research might be picking up, however. Recently, a number of companies aiming to build better delivery tools have launched, including Harvard University spinout Dyno Therapeutics and 4D Molecular Therapeutics, which recently raised $222 million in an initial public offering. 

Larger companies are interested, too. Roche, Sarepta and Novartis have all partnered with Dyno, for example. 

In gene editing, meanwhile, researchers are developing new ways to cut DNA, while Beam and others are advancing different editing approaches altogether. 

Will large drugmaker bets bear fruit? 

Billions of dollars have flowed from pharmaceutical companies into gene therapy over the past few years, leaving few large multinational drugmakers without a research presence. 

2020 was no different, with sizable acquisitions inked by Bayer and Eli Lilly, as well as an array of smaller investments from Pfizer, Novartis, Johnson & Johnson, Biogen, and UCB. And CSL Behring, best known for its blood plasma products, spent nearly half a billion dollars to buy UniQure's most advanced gene therapy, a treatment for hemophilia B. 

Over the past three years, there's been at least $30 billion spent on biotechs involved in gene or cell therapy. (Four deals account for the majority of that value.)

All of that dealmaking, while following promising and compelling science, is ultimately a bet that one-time genetic treatments can be scaled up and commercialized into a lucrative business. 

Many of the acquired companies are working on therapies for very rare disorders affecting hundreds or thousands of people. A handful, however, are taking aim at more prevalent conditions, starting with still relatively uncommon diseases like hemophilia to ones affecting millions of people like Parkinson's. 

"For gene therapy to meet our lofty expectations — not just for investors, but for society — it has to make the leap from these ultra-rare diseases," said Loncar.

Commercially, the track record for the few therapies on the market in the U.S. is mixed. Luxturna, now owned by Roche, is a niche product. Zolgensma has broader use and earned Novartis about $1 billion in the year and a half it's been commercially available. 

Two cell therapies from Novartis and Gilead, meanwhile, have struggled to gain traction.

Gene therapy's biggest commercial test yet was supposed to come this year, with the expected approval of BioMarin's Roctavian in hemophilia A. The FDA's surprise rejection could mean a yearslong delay in the U.S., but the challenges of pricing, reimbursement and patient access in gene therapy remain dauntingly large.

Nearly $20 billion in funding flowed into biotech companies developing cell-, gene- and tissue-based therapies last year, widely eclipsing the total invested in 2019 and ending up 50% higher than the previous record of $13.5 billion set in 2018.

The numbers, compiled by the industry association Alliance for Regenerative Medicine in a report published in March, reflect a fast-expanding sector that's drawing increased attention from venture capital backers, public company investors and large pharmaceutical companies. 

More than $5.5 billion of last year's investments came via venture capital financing for startups in the space, most notably the $700 million raised by cell therapy developer Sana Biotechnology ahead of its initial public offering last month. Nearly $10 billion came via IPOs and secondary stock offerings, and another $3 billion through upfront payments in partnerships and collaboration deals. 

Biotech as a whole had a strong year in 2020. The Nasdaq Biotechnology index, which tracks the industry's stock market performance, rose by nearly 25%, recovering from a spring slump as COVID-19 became a pandemic to regain ground strongly. 

Cell and gene therapy companies did even better, according to ARM, which calculated in its report stock performance that surpassed the broader NBI index. 

The regenerative medicine sector, which includes tissue-based treatments as well as cell- and gene-based medicines, got larger, too. ARM counted roughly 1,100 developers worldwide, up about 100 from 2019. 

"The future is now," said Janet Lambert, ARM's CEO, in an interview. "It's not like we're waiting for there to be a big and meaningful cell and gene therapy sector. There is a big and meaningful cell and gene therapy sector."

Recently, however, some of the most advanced companies have run into regulatory roadblocks or revealed disappointing study results. Cancer cases reported in trials of two closely followed gene therapies renewed safety concerns, even if it appears the experimental treatments did not play a causative role. 

Setbacks are to be expected amid the sector's fast growth, said Lambert, who noted the roughly 150 late-stage studies now ongoing. Many of those programs likely won't succeed, given the usual rates of clinical trial failure in biotech. 

Unlike in the past, however, the pipeline of cell and gene therapies is so broad, and the number of companies involved so high, that setbacks for any one program are less likely to slow the entire sector than in past decades. And while the Food and Drug Administration has not cleared any new gene therapies since landmark approvals for Roche's inherited blindness treatment Luxturna and Novartis's spinal muscular atrophy therapy Zolgensma, the agency recently OK'd new CAR-T cell therapies for types of lymphoma. 

Across Europe, the U.S. and China, regulators are expected to decide on approvals for eight regenerative medicine therapies this year, according to ARM. In the U.S., cancer cell therapies from Bristol Myers Squibb and Johnson & Johnson could reach market, as well as a tissue-based treatment from Mallinckrodt for severe burns. 

Developers and regulators are also learning quickly, particularly in areas like manufacturing and quality control.  

"One of the important things we need to work on is how best to regulate the [chemistry, manufacturing and control] aspects of cell and gene therapy," said Lambert.

"It's clearly a place we've struggled," she added, noting recent disagreements between the FDA and developers over CMC issues like testing assays. 

ARM members are hoping to have more conversations with the agency earlier, Lambert said, although the FDA division in charge of cell and gene therapies has been stretched thin. In comments to the agency, ARM has advocated for the division to receive more resources and staff in renegotiations for the next FDA user fee agreement that will start in 2023. 

Instruments maker Thermo Fisher plans to spend $180 million to build a 290,000-square-foot site that will double capacity to make viral vectors, the inactivated viruses used to convey gene therapies into human cells.

The contract development and manufacturing organization plans to complete work on the facility in Plainville, Massachusetts, in 2022 and add more than 200 jobs.

Thermo's investment is the latest in a series of expansions designed to meet growing demand for gene therapies and vaccine. The company paid $1.7 billion in 2019 to buy viral vector contract manufacturer Brammer Bio. The company later opened a $90 million manufacturing facility in Lexington, Massachusetts, and it's expanded capacity at sites in Cambridge, Massachusetts, and Alachua, Florida, as well.

Thermo's new space in Massachusetts will rival a 300,000-square-foot plant operated by Swiss drug manufacturer Lonza in Pearland, Texas, outside of Houston. Lonza called the plant the world's largest dedicated to cell and gene therapy production when it opened in 2018.

Lonza has also been steadily making deals in an effort to support biotech and pharma customers through every step of the process in gene therapy. As of April 2019, the company had worked with more than 45 viral vector customers and expected that number to keep climbing.

Other companies are expanding, too. Fujifilm in November 2019 announced plans to spend about $120 million in the gene therapy field, including a new innovation center in Texas. Also in 2019, Catalent paid $1.2 billion for Paragon Bioservices to bolster its manufacturing capacity for gene therapies.

Thermo said its new plant will take advantage of the latest technology, with digital connectivity and advanced operator training. The company said it chose to construct the site in Plainville to take advantage of nearby Thermo facilities and draw on the Boston-area talent pool

Catalent will spend $130 million to build five more gene therapy manufacturing suites at its plant in Harmans, Maryland, to meet "a growing customer pipeline and market demand."

The new suites are targeted for completion in the first half of 2022. By then, the contract manufacturer expects to have a total of 15 gene therapy suites in two buildings, along with cold storage warehousing.

Located near the Baltimore-Washington airport, the Harmans campus is one of five Catalent gene therapy sites in Maryland. With the new additions, Harmans will have a footprint of about 350,000 feet and include fill/finish and supply chain services among others, Catalent said.

The company's most immediate focus in gene therapy is the production of materials for Zolgensma, a spinal muscular atrophy treatment from Novartis. Catalent last summer announced the Food and Drug Administration cleared the Harmans facility for commercial work on the therapy. It's the most expensive treatment ever brought to market, with a price tag of more than $2 million.

While some gene therapy companies have experienced setbacks in development, the field overall is still "at the beginning of a steep trajectory of innovation," analysts at Piper Sandler wrote in a September 2020 report. Data due from clinical trials of gene therapies may also offer a boost, they said.

One of the biggest hurdles in the gene therapy field is the incredible complexity involved in manufacturing. Some companies have opted to outsource some of the production to contract manufacturers such as Catalent, creating a surge in demand and shortages in supply.

Catalent moved to increase its production capabilities by buying Paragon Bioservices for $1.2 billion in 2019 and Masthercell Global for $315 million last year. It's also continued to expand facilities, including a $50 million plan announced last week to install a new high-speed vial filling line at its plant in Bloomington, Indiana.

Meanwhile, Catalent has become a go-to partner for coronavirus vaccine developers. The company is working with Moderna, Johnson & Johnson and AstraZeneca.

Biogen plans to spend $200 million to build a gene therapy manufacturing facility at its Research Triangle Park campus in North Carolina as the company invests further in the fast-growing field of genetic medicine.

The 175,000-square-foot plant should be operational by 2023 and employ about 90 people, Biogen said in March. The company said it chose the location because it's found the area attracts diverse and "highly qualified and passionate employees."

Biogen didn't detail the types of therapies that would be produced at the facility but said it will have "differentiated, sustainable and advanced manufacturing capabilities." The company has primarily focused on gene therapies for eye diseases, most notably through a roughly $800 million buyout of Nightstar Therapeutics.

Building a manufacturing plant is a notable sign of Biogen's commitment to developing and commercializing gene therapies. The company has made several acquisitions and research deals in the field, seeking to diversify its pipeline.

Most recently, Biogen announced plans to work with Germany's ViGeneron on gene therapies for inherited eye diseases. Last year, the company signed a deal with Sangamo Therapeutics' to develop gene editing medicines for neurological conditions including Alzheimer's disease.

In 2019, Biogen bought Nightstar Therapeutics for about $800 million to acquire a group of experimental gene therapies for inherited retinal diseases. One of the treatments is in Phase 3 testing for choroideremia and another is in a Phase 2 trial for X-linked retinitis pigmentosa. At the time of the acquisition, Biogen said both could be on the market early this decade.

But Nightstar didn't have its own in-house manufacturing capabilities, a crucial asset for gene therapy developers. The new plant in North Carolina will give Biogen more control and oversight over gene therapy development and commercialization of any that might succeed.

The company joins a number of gene therapy developers — including Audentes Therapeutics, Pfizer, Bluebird bio and AveXis — who have chosen North Carolina to build manufacturing plants.

Even as companies build out internal manufacturing capabilities, contract manufacturers are also rushing to meet demand. Just last year, Catalent said it would spend $130 million to build five more gene therapy manufacturing suites at its plant in Harmans, Maryland, and Thermo Fisher announced a $180 million investment in a site for the viral vectors used to convey gene therapies into human cells.

A coalition of drugmakers in March agreed to work with Vineti, a San Francisco-based technology company, to usher in new standards that could ease some of the safety and logistic concerns surrounding advanced medicines like cell and gene therapies.

These medicines often pass through complex networks of manufacturers, transporters and healthcare providers before they're delivered to patients. As many of them are personalized and meant to treat very ill patients, it's critical to know exactly whom each therapy is for and where it is along the supply chain. And yet, modern cell and gene therapies still use the same patient identifiers as more "legacy" treatments like bone marrow transplants — standards that, according to Vineti, "do not always cover all the use cases presented by today's personalized therapeutics."

Vineti wants to establish a new, flexible patient identifier akin to a digital ID badge that could work for all types of advanced medicines and improve traceability, control costs and better ensure patient safety. The company now has more support, too, from that coalition of drugmakers, which includes several biotechs as well as large pharmaceutical firms like Novartis and Takeda.

Over the last few years, the cell and gene therapy field hit major milestones. In the summer of 2017, the Food and Drug Administration approved a first-of-its-kind cellular medicine from Novartis for a hard-to-treat form of leukemia. Less than two months later, another cell therapy developed by Kite Pharma was cleared for patients with a different blood cancer.

Before the end of that year, Spark Therapeutics received the first-ever FDA nod for a gene therapy targeting an inherited disorder. And since then, a small crop of advanced medicines developed by Kite, Juno Therapeutics and AveXis have come to market.

These successes have, in turn, enticed more companies to invest in cell and gene therapy research. Gilead, best known for its work fighting viruses, dropped $12 billion to buy Kite right before its first treatment got approved. Novartis then agreed to buy AveXis for almost $9 billion in 2018, and was soon followed by Roche, which announced in early 2019 it would acquire Spark for $5 billion.

Experts don't foresee the field calming down anytime soon, either. The FDA has said it expects to be reviewing and approving between 10 and 20 cell and gene therapies each year by 2025.

Yet, an influx of advanced medicines could exacerbate some of the existing challenges in making and delivering them. Cell therapy developers like Novartis, for example, have already run into problems commercializing their products, which has caused certain investors to take a more tepid view of the field.

Vineti is hoping its new identifier can iron out some of these problems. As it stands, Vineti's proposal is to have a seven-part ID that would work for all advanced medicine types, including personalized cancer vaccines, cancer immunotherapies, and autologous cell and gene therapies.

The ID would have a "core ID" piece, which shows the company making the product and the specific patient meant to receive it. Then there would be a "treatment ID" piece to identify the product and the order. Lastly, there would be a "specimen ID" piece to detail what procedure is taking place, the run of this procedure, and the number of items produced from it.

So, according to one example posed by Vineti, if a patient is receiving an advanced medicine called ABC from the hypothetical company Acme Pharma, and the first step in that process is to collect one bag of raw material through one session of apheresis, the resulting code would be ACM123456-ABC01-AP0101. In that code, ACM represents the company, 123456 is the patient number, ABC is the therapy, the first 01 is the order number, AP signals apheresis and the last pair of 01s indicate that this is session 1 and should create one bag of material.

Vineti is currently accepting industry-wide feedback on its proposal.