AI In Drug Development: From Hype To The Clinic & Beyond

The TD Cowen Insight

With milestones such as the first Generative AI-based drug entering Phase 2, AI is delivering on the promise of more efficient and cost-effective R&D via many use cases. Much has transpired since our Primer on AI in Pharma R&D. AI is now mainstream thanks to ChatGPT, and making its mark globally with a Presidential Executive Order on AI.

We also introduce our proprietary Clinical Trial Scorecard. It lists companies using AI-enabled drug discovery, breaks down disease focus, preclinical and clinical status, and drug-type modality to allow investors to track.

More Efficient Drug Development With AI

Despite a better understanding of biology, automation, and lower costs of data acquisition plus the increase in R&D dollars spent, the time to develop drugs is increasing. This conundrum, coined “Eroom’s Law” (Moore’s Law spelled backwards), was first observed in the 1980s.

AI is delivering on the promise of dramatically improving the speed and efficiency of drug development. We are seeing an increase in AI-based clinical trials and a significant increase in the number of partnerships with AI tools players. We believe that this is only the beginning. We see AI adoption and implementation becoming an increasingly larger part of biopharma industry strategy (our estimates show a doubling of AI focus the next 3-5 years).

A New Era in Drug Discovery

Near term, we believe AI will usher in a new era of drug discovery. This includes enabling the discovery of candidates in “hard-to-drug” indications such as neurology and autoimmune/inflammation and “hard-to-drug” targets, such as ion channels and G-Protein Coupled Receptors (GPCRs), which will manifest in many more drug candidates entering the clinic in the next few years. We believe there is a compelling investment opportunity given recent valuation pullbacks.

AI Applications in Drug Discovery And Development

This multi-analyst, multi-sector report provides a comprehensive overview of the rapidly evolving field of AI in drug discovery and development. We outline the basics of AI and drug discovery, the history of advances in AI that have led up to this point, the current state of the market, and the expansive range of opportunities for AI to speed drug discovery and development.

We detail the growing applications of AI & case studies including:

• Identifying potential failures or successes
• Accelerating patient selection for clinical studies
• Unraveling discoveries in disease pathology
• Identifying novel targets for drug candidates

TD Cowen Clinical Trial Scorecard

This work is supported by our proprietary “Clinical Trial Scorecard,” which analyzes the publicly disclosed preclinical and clinical programs in the industry for both private and public companies. We break down programs by disease focus, preclinical and clinical status, and drug-type modality to allow investors to track the “game.” We’ve also done channel checks with our covered companies offering AI tools and/or developing drugs supported by proprietary AI platforms, including those from Nvidia’s Head of Health Care.

Valuating The Impact of AI on Pharmaceuticals and Biotech

We believe many investors are misunderstanding the opportunity as “too early”. However, it’s clear to us that the adoption of AI by large pharma and biotechs via collaborations in the last 3 years ($41B+ in potential value) will continue, driven by efficiency gains and increased probabilities of success.

Given the annual R&D spend in pharma (+$250B worldwide by 2026 per Evaluate Pharma), the economic value of accelerating the time to market by 3-5+ years and extending the number of years before patent expirations is immense. In addition, our estimates show potential R&D cost-savings of ~$50BN annually by 2026. As a result, we expect financing to remain available for differentiated companies and for several private companies to go public in the next few years. Investors, pharmaceutical, and biotech companies now have real world examples of drugs progressing through the clinic which highlights the tremendous potential upside of AI and in silico approaches to dramatically improve R&D productivity.