75 AI-discovered molecules are entering clinical trials

A new generation of artificial intelligence tools is accelerating drug discovery and development, with multiple AI-discovered molecules now entering clinical trials.

The big picture: The pharmaceutical industry is witnessing a transformation as AI-driven drug discovery companies compete to develop new treatments for difficult diseases, potentially reducing the traditional 10-15 year, $2 billion development timeline.

• Insilico Medicine, a US-based startup, has developed an AI-discovered drug for idiopathic pulmonary fibrosis (IPF) that has shown promising results in early clinical trials
• At least 75 AI-discovered molecules have entered clinical trials, marking a significant milestone in pharmaceutical development
• Major tech companies like Alphabet have entered the space, launching specialized divisions such as Isomorphic Labs for AI drug discovery

Key applications of AI in drug discovery: AI is primarily being deployed in two crucial steps of the pharmaceutical development process, revolutionizing traditional research methods.

• AI systems mine large databases to identify therapeutic targets at the molecular level, making connections between molecular biology and diseases
• Generative AI, similar to the technology behind ChatGPT, designs drug molecules that can bind to these targets, replacing the costly process of manual chemical synthesis and testing
• The technology can also predict clinical trial success probability, helping to optimize the drug discovery process

Industry innovation and progress: Companies are developing unique approaches to overcome the challenges of limited data in AI drug discovery.

• Recursion Pharmaceuticals has installed what it claims is the fastest supercomputer in the pharmaceutical industry to generate and analyze massive quantities of molecular data
• Insilico Medicine has reduced typical drug discovery timelines from four years to 18 months, while testing fewer molecular variations
• The field maintains significant human involvement, with experts noting there is no clear definition yet of what constitutes an “AI-discovered” drug

Challenges and limitations: Data availability remains the primary obstacle in advancing AI drug discovery.

• Limited historical data can potentially introduce biases in both target identification and molecule design
• The true value of AI in drug discovery will only be proven when AI-discovered molecules successfully complete clinical trials
• The technology is expected to complement rather than replace pharmaceutical scientists, with the main benefit being reduced failure rates

Looking ahead: The validation of AI in drug discovery hinges on demonstrating improved success rates in clinical trials compared to traditional methods, which could reshape the future of pharmaceutical development and potentially deliver more effective treatments to patients faster.