The Amazon of Gene Therapy

How scientists used genetic engineering and ML to bring a virus back from the dead as a better deliver method for gene therapies

Image for post
Image for post
Molecular structure of the Anc80 virus. [Figure 2C, Zinn et al. 2015]

Gene therapy is a powerful tool to treat diseases from cancer to deafness, but it requires safe and effective delivery to the correct cells in the body. This post is about the unique genesis of Anc80, a new viral delivery system.

Gene therapy has two parts: 1) the genetic package, to fix a disease-causing error, and 2) the delivery system, to deliver the fix to its intended location.

Existing delivery systems cannot reach all cells, limiting gene therapy’s utility.

Anc80 has proven 1000X more effective in targeting difficult-to-reach cells in mice and the first human trials start this year. If successful, Anc80 will enable drugs for many currently-untreatable diseases.

Delivering genetic packages is hard, but nature has given us a bespoke cellular infiltration system: the adenovirus.

Adenoviruses normally cause respiratory infections, but scientists in the 1980s began hijacking the infectious machinery to deliver genetic packages into cells without harmful virality.

There’s one big problem: humans have evolved a sophisticated immune defense against adenoviruses.

In 1999, scientists tried to treat 19-year old Jesse Gelsinger for a liver disorder using an adenovirus-delivered gene therapy. Jesse wasn’t the first to receive gene therapy via adenovirus, but he was the first to suffer an overwhelming immune response.

He died within four days.

Scientists began searching for alternative viruses that could deliver gene therapies without lethal immune reactions.

Adeno-associated viruses (AAVs) looked promising: AAVs don’t cause disease, so we haven’t developed immunity against them.

AAVs became the postman for many gene therapies, including the first FDA approval in 2017: Luxturna—a $850K drug to treat infant blindness. Roche is now buying Luxturna’s developer Spark Therapeutics for $4.8B.

Despite advantages over adenoviruses, AAVs still struggle to reach certain cell types and trigger immunity at high doses or in certain contexts, restricting gene therapy to a minority of potential uses.

Scientists have tried to reengineer AAVs, but enhancing one aspect (better immunity) often introduces new limitations (worse delivery).

It was time for a new approach.

In 2015, the scientist Eric Zinn, along with colleagues from Harvard Medical School and the Massachusetts Eye & Ear Infirmary, published a breakthrough.

Rather than modifying current AAVs, Zinn et al. explored the evolutionary history of the viral family, looking for common ancestors that no longer exist.

The big idea: earlier AAVs may have once had properties that were since lost during evolution because they didn’t help the virus at the time, but which could now be beneficial in the context of viral gene therapy delivery.

Zinn et al. identified Anc80, an ancestral AAV that they hypothesized could have enhanced properties. As the DNA sequence of Anc80 was lost to time, the team looked at comparable sequences from its descendants.

Then, using genetic engineering and in silico reconstruction, Zinn et al. recreated 776 possible variants of the ancient Anc80 virus.

The team then tested the ability of each Anc80 variant to infect cells.

On the 65th variant, Zinn et al. hit the jackpot: Anc80L65.

The 65th iteration of Anc80 outperformed existing AAVs in the delivery of a genetic package to liver, muscle and retinal cells in mice. It also displayed higher heat tolerance and improved molecular stability, important for delivering gene therapies in the tumultuous innards of a living organism.

Image for post
Image for post
Green is Good: green shows the amount of a gene that reached its target when delivered to liver (top row), muscle (middle row) and retina (bottom row) using Anc80 (right column) versus AAV2 (left column) and AAV8 (middle column). Anc80 is better because more of the gene reached its target. [Figure 4A, Zinn et al. 2015]

Anc80 proved safe and effective in monkeys. With significant differences from all of its descendant AAVs, Anc80 also didn’t stimulate immune reactions, because modern organisms haven’t encountered this virus for generations.

Zinn et al. went full Lazarus on Anc80 — and it worked.

Zinn’s colleagues began testing Anc80 against genetic diseases, starting with challenging disorders of the inner ear.

Five studies (Landegger et al. 2017, Tao et al. 2018, Yoshimura et al. 2018, Duarte et al. 2018 and Gu et al. 2019) proved Anc80 could deliver genes to ear cells that existing AAVs cannot effectively reach, including cells that lead to debilitating disorders when mutated.

One such disease — Usher syndrome — causes deafness, blindness and loss of balance. As AAVs don’t hit the responsible cells, over 15K people in the US alone have no good treatment.

Pan et al. 2017 showed that because Anc80 can reach those requisite ear cells, Anc80 could deliver a gene to correct Usher syndrome. Tested in mice, Anc80 proved 1,000X more effective than comparable delivery with current AAVs.

Suzuki et al. 2017 demonstrated that Anc80 could even fix genes in ears that were structurally damaged — a common real-world scenario that has stymied past gene therapy attempts — further validating Anc80.

Additional studies showed Anc80 can target other remote cell types: in the anterior eye (Wang et al. 2017), retina (Carvalho et al. 2018), kidney (Ikeda et al. 2018) and CNS (Hudry et al. 2018), establishing uses beyond the ear.

The catch: So far, all of the data is in mice or monkeys—not in humans.

Biotech companies have begun to develop Anc80 for human patients.

Akouos is a Boston-based gene therapy startup focused on ear diseases and co-founded by an Anc80 inventor. The company has licensed exclusive rights from the Mass Eye & Ear Infirmary to use Anc80 to treat hearing loss.

Akouos launched in 2017, then raised $50M from top investors including RA Capital and the venture arm of Novartis, which acquired AveXis and its spinal muscular atrophy gene therapy for $8.7B.

Akouos manufacturing partner Lonza has separately sponsored several Anc80 studies. In 2016, Lonza licensed rights from the Mass Eye & Ear Infirmary to manufacture Anc80 for each therapeutics company like Akouos.

While Akouos takes Anc80 to the clinic for ear diseases, its scientific founder is guiding GenSight, which has added Anc80 to its gene therapies for inherited eyesight loss. In parallel, the non-profit Odylia Therapeutics is using Anc80 for ultra-rare eye diseases.

Another startup, Vivet Therapeutics, raised $41M in 2017 (also partly from Novartis) and licensed rights to Anc80 for metabolic diseases. The later-stage company Selecta has licensed Anc80 for a combo treatment with its improved version of rapamycin for immunomodulation.

From its unique genesis to strong preclinical data, Anc80 is a useful case study in the discovery and development of novel science.

Like all new biomedical tech, Anc80 is risky. A quirk of human biology could lead to harmful effects not seen in earlier studies—a common problem in translational biomedicine. Or it may simply not work in people at all. Other emerging delivery tools like nanoparticles could also outcompete Anc80.

Despite these challenges, I’m bullish on Anc80 becoming a leader in the field of gene therapy delivery.

I hope you’ve enjoyed learning Anc80’s provenance and potential. Feel free to email me if you have questions, comments or other exciting biotech to discuss.


Written by

Biotech VC. Director or Observer on the boards of six RA Capital companies. Former CEO of Nivien Therapeutics. My writing does not represent RA Capital.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store