Dyno inks potential $1b neurological gene therapy deal with Roche


Research collaboration aims to overcome key challenges facing AAV gene therapies for neurodegenerative diseases.

Gene therapy platform company Dyno Therapeutics has signed a new research collaboration with pharma giant Roche to drive advances in gene therapy for neurological diseases. Using Dyno’s AI-enabled platform for designing adeno-associated virus (AAV) vectors, the collaboration aims to develop next-generation AAV gene therapies that address critical issues in gene delivery, immune response and therapeutic precision.

In addition to a $50 million upfront payment from Roche, Dyno’s potential earnings from the deal could exceed $1 billion from future milestone payments, along with royalties on commercialized products. The new collaboration is the second research agreement between the companies, following an initial partnership in 2020, which focused on neurological and liver-targeted therapies.

Dyno’s platform addresses limitations of conventional AAV-based gene therapies, which often rely on a narrow range of naturally occurring vectors with constraints on delivery efficiency, immune response and manufacturability. These challenges restrict the potential of gene therapies for certain diseases, particularly neurodegenerative diseases such as Alzheimer’s and Parkinson’s.

Dyno’s approach, which combines AI and high-throughput in vivo testing, enables the accelerated engineering of AAV capsids. These capsids are the protein shells of viral vectors and are critical for targeting specific tissues and achieving effective gene transfer. Combining its platform with in vivo sequence-function analysis, Dyno says it can design AAV capsids that achieve better tissue targeting and immune evasion, ultimately advancing the efficacy and reach of gene therapies.

“Our approach combines extensive in vivo data with the world’s most advanced AI models for sequence-function prediction, empowering Dyno capsid engineers with industry-leading capabilities for solving the longstanding challenge of therapeutic gene delivery,” said Dyno founder and CEO Dr Eric Kelsic. “Dyno’s platform brings the entire field closer to realizing a future where all gene therapies are safe, effective, and widely accessible to all patients who need them.”

Dyno’s design process begins with AI-driven capsid sequence predictions, followed by DNA synthesis on specialized printers that generate vast capsid libraries. Each variant is then tagged with unique DNA barcodes, allowing high-throughput sequencing to monitor and analyze how specific capsids perform in terms of key therapeutic properties. Machine learning models trained on this data further refine predictions of capsid function and continually expand the fitness map of AAV sequences.

By balancing exploration of untested sequences with optimization of known effective ones, Dyno claims iterative approach not only improves existing gene therapies but also opens doors to treat diseases previously inaccessible to gene therapy by targeting new organs and cell types.

Under the terms of the new agreement, Dyno will focus on the design and discovery of optimized AAV capsids for neurological indications, while Roche will oversee the validation, preclinical and clinical stages, as well as commercialization for therapies developed with these novel capsids.

 “Our previous collaboration with Dyno Therapeutics gives us great confidence to increase our investment in therapeutic gene delivery, to support our neurological disease portfolio,” said Boris L Zaïtra, head of corporate business development at Roche. “Our combined knowledge and resources will allow us to explore new treatments for historically difficult-to-treat neurological diseases.”

In 2021, Dyno landed a monster $100 million Series A funding round led by Andreessen Horowitz, and the company has also secured partnerships with other gene therapy developers, including Astellas and Sarepta.

Photograph: Marlon Trottmann/Shutterstock



Source link

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top