Our investment in Weave Bio: Using AI to alleviate regulatory friction in drug development
By Joel Dudley and Nick Olsen
Development of new medicines is the ultimate knowledge work. A new medicine is often the culmination of decades of work, hundreds of scientists, dozens of disciplines, and ultimately one hypothesis to improve the lives of patients. It is hard to tell if it is in spite of, or because of, this complexity that data and knowledge management is a significant challenge for the industry.
We’ve previously written about our excitement for the future of this industry, a data-driven revolution, and much of our excitement stems from new technologies to interrogate biological systems and an emergence of powerful new therapeutic modalities. We are not alone in our belief that advances in high-throughput chemistry and biology are driving a reimagining of the discovery process through the lens of computation and modern technology stacks. Early research and discovery represents an important yet limited subset of the complex activities that are required to bring a drug to market. The further a drug program moves through the drug development process–from early discovery to clinical trials and beyond–the more the program is de-risked and value is created. However, a majority of activities and investments into technology innovation in the pharmaceutical industry have focused on the earlier, discovery-oriented phases of drug discovery programs.
All of this work, all of these experiments, all of the hypotheses tested result in a single drug embodiment that hopefully enters clinical trials. The first time a new drug is taken in a first-in-human study is a massive achievement. It’s the culmination of too many hours and dollars to count. And it’s passed the first major regulatory hurdle — the FDA has deemed the drug safe enough to progress into a human study. This first regulatory filing is called the Investigational New Drug (IND) application that the FDA reviews prior to a sponsor having the opportunity to progress the drug into clinical trials, and it is the first of many submissions to the FDA through the lifecycle of a drug. The IND is a thousands-of-pages long document that meticulously summarizes all the information known about that drug candidate that includes detailed scientific data, manufacturing information, and the clinical study plan for the first trials.
It’s an interesting dynamic for companies at this stage. By this point in a drug’s journey, the sponsor has likely filed key IP, and the patent clock is ticking loudly. Yet, given how manual and labor-intensive the effort is, necessitating the collaborative efforts of large, multidisciplinary teams, most companies wait until the very last experimental data comes in before sprinting for months to summarize years of work in a single document. This is inefficient and can create costly errors that result in clinical holds that slows the drug from impacting patient care. Additionally, this critical data and knowledge is often ossified in static regulatory documents, limiting a company’s ability to learn from and leverage this critical information for the rest of the drug’s life as it progresses into clinical trials and eventually (hopefully) as a commercial product.
Enter Weave Bio
Weave Bio wants to enable a different future.
One where scientists weave together a living record of their therapeutic candidates enabled by modern digital technologies and AI. Their first product, AutoIND, is an IND authoring and preparation platform that augments the teams responsible for putting together the IND data package. The product includes a number of helpful features that enable teams to move quickly, collaborate seamlessly, and review and refine robustly. We are not yet in a world where AI can run without intervention and review, but with AI as a robust navigation tool or GPS that facilitates an intuitive workflow, scientists can significantly increase their throughput and spend more of their time on activities that drive meaningful value.
To get this right, Weave built a team that bridges life science, regulatory, AI, and software expertise. They forge new technology paradigms to solve hallucination challenges, to develop methods that link together biomedical concepts, and that faithfully reference the source documents. The leadership team at Weave Bio brings together a wealth of expertise across the life sciences. Ari Caroline (CEO), Umut Eser (CTO), Brandon Rice (CPO), and Lindsay Mateo (CCO) lead the team.
This is just the beginning of the journey for Weave. In essence, we view Weave as building a drug intelligence platform. Instead of these regulatory interactions being point-in-time submissions, each drug should have a living, digital embodiment. By seamlessly integrating all relevant information, that digital record becomes a mechanism for all future regulatory filings, strategic discussions about lifecycle management, and a more seamless mechanism for engaging with regulatory authorities.
This is why we are thrilled to announce our lead investment in Weave Bio, a groundbreaking company that is poised to transform the regulatory reporting landscape in drug development. We are proud to partner with Magnetic Ventures and the entire Weave Bio team on this exciting journey. The road ahead is long, but the potential impact on patient care and the drug development process is immense. If you are involved in drug development and are interested in learning more about Weave Bio’s platform, please reach out to us or the team at Weave.
