It’s been a turbulent year in the world of gene therapy, with another negative safety signal when Novartis confirmed that two children treated with its gene therapy to treat spinal muscular atrophy, Zolgensma, had died of acute liver failure. On the positive side, there were three new approvals (Roctavian for severe hemophilia A, Zynteglo for beta thalassemia, and Skysona for cerebral adrenoleukodystrophy).
While the Zolgensma safety issue was “not a new safety signal” per Novartis, this unfortunate incident shined a spotlight on one of several critical challenges that scientists, entrepreneurs and venture capitalists working to build on early successes like Zolgensma know all too well: gene therapy can offer hope to patients with previously untreatable diseases, but there’s more work to be done to improve both the safety and efficacy of this therapeutic modality. Three critical challenges include improving delivery, manufacturing, and conducting translationally-relevant pre-clinical studies.
Improving delivery of gene therapy
One challenge that scientists working in gene therapy are focused on is improving the delivery of therapeutic genes to specific organs and cells within the body. Zolgensma employs the most widely used delivery method for gene therapy today: adeno-associated virus (AAV). As a therapeutic modality, AAV use a retrofitted virus protein shell called a capsid to deliver a genetically encoded therapeutic payload called a transgene. AAVs are good at shuttling therapeutic genes into the body, but are often imprecise in its targeting and as a result require relatively high doses to achieve efficacy in the target organ or cell-type. At high doses, AAV, like other therapeutics, can raise safety risks due to off target effects. Compounding this problem, patient populations treated with gene therapy have some of the most severe underlying pathologies and can least tolerate these safety risks.
Scientists in academia and industry have been employing well-worn protein engineering methods such as mutagenesis, shuffling and directed evolution for decades in an effort to improve the targeting and delivery efficiency of AAV capsids. Newer methods utilize DNA-barcoded capsid libraries, next generation sequencing, and machine learning to generate robust in vivo data to optimize capsid delivery. If we are able to sufficiently enhance targeting and delivery efficiency, it will hopefully enable the field to lower the overall doses used in human clinical trials, and should improve the therapeutic window of gene therapy in patients.
Manufacturing and Quality
Another challenge with AAV gene therapies is that during manufacturing, some capsids end up “empty,” meaning they don’t actually include the therapeutic gene. Those empty and partial capsids have no benefit to patients and the field should endeavor to minimize any impurities in precision medicines. I am confident that chemistry manufacturing and controls (CMC) experts will continue to optimize large-scale manufacturing of AAV gene therapies and we will see overall yields increase and full-to-empty ratios improve to over 90%.
The costs associated with building, validating and operating a gene therapy process development lab and GMP facility are significant. In this time of challenging capital markets, companies should consider the benefits to partnering early with external contract development manufacturing organizations (CDMOs) to leverage their existing infrastructure and expertise.
Innovations such as producer and packaging cell lines for AAV production will also be important as gene therapy moves from rare to more prevalent conditions.
Improving early translation efforts
Many gene therapy programs start in academic labs, using animal models of genetic diseases. Unfortunately, what might make an excellent peer-reviewed scientific publication does not always provide the right framework for translation to drug development. As an example,
Many scientific labs at high-caliber research institutes dose mouse models with experimental gene therapies at the day of birth, or “p-zero.” This roughly corresponds to the third trimester in human development. In some cases, the severity of some disease models requires this approach, but should be evaluated empirically by dosing disease models at different timepoints to more closely match the patient journey. It is uncommon to immediately diagnose a patient with a rare disease at birth, and improvements in prenatal and newborn screening are important tools in our journey to shorten this timeline. Most patient journeys from symptom onset to diagnosis may be outside of the window studied in translational research. For example, the eye disease retinitis pigmentosa often doesn’t start to impair vision until the patient is in young adulthood.
If researchers administered gene therapy to animals at times that more closely approximate the ages at which human patients start to show symptoms, researchers, clinicians, patients, investors and payers would benefit from a more accurate picture of just how effective the gene therapy is likely to be in people. This would have a positive impact on the efficient deployment of human and capital resources.
Partnering to accelerate the field forward during challenging markets
To further advance the field of gene therapy and accelerate these potentially life-saving treatments, we should foster closer collaborations between academia, patient advocacy groups, venture capital investors and biopharmaceutical companies.
Investors and biopharma companies should provide academia the guidance, tools and resources to optimize early translational research. Patient advocacy groups should develop natural history studies, registries, biobanks, and key opinion leader networks to help educate and guide academics and investors. Academics in turn should be receptive to conducting the “hard experiment,” meaning the one that has higher translatability to the clinic. Effective partnership in these areas will be critical.
Manufacturing is another area prime for partnerships – especially in today’s capital markets environment. There has been a trend for early-stage companies to build manufacturing infrastructure for gene therapy – so companies can ‘own their own destiny’ and not be beholden to third-party contract development manufacturing organizations (CDMOs) – but not every program or company needs its own 100,000 square-foot plant. Emerging gene therapy programs and companies need to be judicious in how they deploy their capital. Beyond having the capacity to manufacture these complex medicines, the technological know-how is both rare and essential. Patients are waiting – why not partner with a company that has already built gene therapy manufacturing capacity and expertise instead of trying to build it from scratch?
Gene therapy has come a long way and we have only just begun to scratch the surface of its remarkable potential. Through meaningful collaboration and a focus on addressing some key challenges, we have an opportunity to reach more patients in search of cures.
