Gene Therapy Company Dyno Therapeutics Pairs Biology With AI For Therapeutic Gains

By Allison Proffitt

May 11, 2020 | Dyno Therapeutics emerged from stealth mode today and announced partnerships with Novartis and Sarepta Therapeutics to develop gene therapies for eye disease and neuromuscular and cardiovascular disease.

Dyno’s platform—CapsidMap—aims to create disease-specific vectors for gene therapy, explains Eric Kelsic, CEO and one of the six company co-founders. Kelsic and other co-founders  worked together in George Church’s lab at Harvard; Dyno has an exclusive option to enter into a license agreement with Harvard University for this technology. Church is also a co-founder of Dyno and Chairman of the company’s Scientific Advisory Board.

“Gene therapy is such a huge opportunity to treat disease; there’s huge unmet need there on the disease front. In addition to that, AAV vectors—it feels like we’re at the beginning of the field. There’s a lot of great work that’s been done on natural vectors, but they have limitations. They only go to certain cells and tissues,” Kelsic told Bio-IT World. “We decided to focus on engineering of the AAV capsid, which are the protein shell of the vectors.”

The company’s approach combines AI and wet lab biology to iteratively design novel adeno-associated virus vectors (AAV) that improve on current gene therapy. Kelsic calls it “high-throughput biology”, measuring many of the properties that are critical for gene therapy in high throughput, specifically efficiency of delivery, specificity to a target, the immune system response, packaging size, and manufacturing features.

“Those five things really make up all the characteristics that are critical for in vivo delivery,” Kelsic said. For gene therapy there’s a capsid profile for each disease. “Think about every disease that you want to treat, every potential therapy, and there’s a certain profile of what’s going to be the optimal vector for that treatment,” he said. “We built that profile into our platform to inform how we do our screening. Essentially we can measure all those properties independently using this high throughput approach.”

Dyno builds a large library of capsids with DNA synthesis, and labels them with DNA barcodes them. “We then run experiments both in vitro and in vivo in animal models,” Kelsic said. “We’re doing that at a very large scale: hundreds of thousands to millions of measurements can be done even in a single experiment.” The results are tracked using DNA sequencing, down to which capsids with which properties ended up in which tissues and organs.

“There’s a huge amount of information in that data, and up until recently no one has been directly learning from that and applying those learnings toward the design of the screening vectors,” Kelsic explained. The CapsidMap platform uses machine learning to pull out the “most meaningful insights” from the in vitro and in vivo models. “In doing that, we fill in all the gaps of our map of the AAV universe. Now rather than just being a handful of points, we’re learning a huge amount about the landscape of AAV.”

Finally, CapsidMap uses AI to design the next round of experiments—both wet lab and in silico—that will further develop the AAV landscape. “We’re programming in these targets: the capsid profiles that are the properties of the best vector. Then the AI takes that together with the data that we have and the models that we have [to] design the experiments that will enable us to identify those better factors and also accelerate the process,” Kelsic said. “It’s a pretty dramatic change from what has been done before.”

Funding and Partnerships

He’s not the only one who thinks so. Alan Crane was “absolutely blown away” when Kelsic explained the platform to him in June 2018.

Crane is a Polaris Entrepreneur Partner and has been exploring the role of AI in life sciences applications for years. “This was by far the most direct, most potential-for-creating-value-for-patients application of AI to biology that I had ever seen,” he told Bio-IT World. Not only did Polaris invest in the $9 million 2018 seed funding, but Crane joined the company as a co-founder and executive chairman.

Today Dyno announces two partnerships—with Novartis and Sarepta Therapeutics—and Crane says partnerships like these are the company’s funding model moving forward. He expects one to two more early partnership deals.

“We recognize that these deals could really fund the company,” Crane said. “We specifically chose to do a modest sized [seed] round, because at that point we already knew what the corporate interest was in the company, and we figured we’d be able to fund the company based on these partnerships.”

Both partnerships announced today involve upfront money, license fees, plus eventual milestones and royalties. Financial details with Novartis were not disclosed. If successful, Dyno could receive over $40 million from Sarepta in upfront, option and license payments during the research phase of the collaboration.

“With these early partnerships we have years of funding for the company,” Crane said. “We’ll probably never do another equity fundraise.”

The partnership with Novartis will focus on improving gene therapies for serious eye diseases. With Sarepta, Dyno will develop AAV capsids for neuromuscular and cardiovascular disease. In both cases, the partner will handle all eventual preclinical, clinical, and commercialization activities.

In addition to those indications, Kelsic said the company is also gene therapies for central nervous system and liver diseases. “In those four organs there’s huge amounts of activity today in gene therapy, and also lots of potential for partnering,” he said. “We’re also looking at other locations in the body, other organs where maybe delivery is challenging today and there’s not a lot of activity because of that.”

The company has about 20 employees today and is based in Cambridge, Mass. Two-thirds of the team is on the wet lab side; 1/3 are computational. The team will grow proportionally, Kelsic said, doubling over the next year.

“There’s a lot of potential to learn from every experiment we do. We certainly have generated a huge amount of data now, both at Harvard and from where we focused on one AAV capsid serotype called AAV2,” Kelsic said. But the company has no intention of abandoning its biology roots and letting AI do the rest of the heavy lifting.

“That’s the synergy that was so exciting for us before—bringing these two things together. Ultimately what we care about most is having the biggest impact we can on patients,” he said. “That’s always going to require coming back to the biology. That’s always going to require validation in vivo. The more in vivo data we have to inform that search process, the more likely it is that we’ll find really transformational vectors.”