AI fuels advances in biotech

AI in medicine discovery is not new, see the illustration above (click on the picture to enlarge it). However, we are currently in an acceleration period, thanks to the combination of these 3 factors:

1. Fast increase in accessible data.
2. Advancements in computing power.
3. Ongoing developments in AI algorithms.

AI applications with near-term impact

Both TD Cowen and McKinsey identified opportunities for AI in medicine development. TD Cowen’s are shown in the image at the right.

The 5 key AI applications with the potential for near-term impact identified by McKinsey are:

1. Enhancing scientific knowledge extraction can improve research quality whilst speeding up the time scientists spend on research.
2. Improving compound (molecule) screening processes, leads to faster identification of potential treatments.
3. Optimizing the design of large molecules and medicine vectors (delivery systems designed to transport medicines to specific tissues or cells within the body), enables more targeted delivery.

4. Facilitating indication selection for asset strategy. Increasing the likelihood to successfully uncover        novel indications with a high probability of success and reducing the number of blind alleys.
5. Streamlining clinical trial processes for faster and more efficient testing of new treatments.

AI in biotech, beyond efficiency: ‘de novo’ medicine design

One particularly exciting advancement is AI’s ability to design entirely new or personalized medicines. Designing a new medicine is like finding a specific key that fits a complex lock (the disease target). Traditional methods involve testing millions of existing molecules, a time-consuming and expensive process with a low success rate. By analyzing vast chemical libraries, AI algorithms can predict how molecules will interact with biological targets, increasing the discovery rate of more effective treatments.

Clinical trials: faster process and better success rate

Identifying the appropriate patients to study a medicine candidate is not easy. Therefore, clinical trials often mistakenly include patients for whom the investigated treatment is not suitable. This can lead to clinical trial failures and/or slow down treatment development. AI helps in selecting the right patients.

In addition to long timelines, trials are also subject to strict regulatory requirements. The use of AI can increase efficiency throughout the entire process. McKinsey estimates that this can result in:

• Up to 50% cost reduction by streamlining clinical trial processes and automatically generating trial documents. About 80% of the costs involved in bringing a medicine to market are related to clinical trials.
• Over 12-month acceleration in the time it takes to conduct a trial.
• Minimum 20% increase in the net present value, thanks to enhanced interactions with health authorities, quality control, and signal management.

By streamlining long and tedious processes and improving participant selection, AI has the potential to reduce costs, accelerate timelines, and improve clinical trial success rates.

All positive? Or also concerns?

Challenges remain, including the need for clear regulatory frameworks, concerns about data bias, and the transparency of AI algorithms. Regulatory bodies, such as the FDA, are increasingly engaging with stakeholders to address these concerns and ensure the responsible use of AI in healthcare.

AI algorithms are only as good as the data they’re trained with therefore biased data must be avoided as it can lead to inaccurate predictions.

The ‘black box’ nature of some AI algorithms can make it difficult to understand how they reach their conclusions.

Conclusion

In conclusion, AI represents a transformative opportunity for the biotech industry. As companies continue to further harness the power of AI, we can expect rapid advancements that will revolutionize medicine discovery and patient care, offering compelling prospects for both investors and patients alike.

AI is rapidly changing the landscape of medicine discovery, offering solutions to the long-standing challenges of cost, speed, and success rates. While challenges remain, AI continues to evolve, and we can expect many breakthroughs that will revolutionize healthcare.

Some examples in the Aescap Portfolio

• Almirall and AI specialist Absci announced a partnership to develop 2 AI-designed dermatology therapeutics.
• Moderna has developed its own version of ChatGPT. The tool is now used company-wide as a ‘personal assistant’ to employees. One example is Dose ID GPT. This uses ChatGPT Enterprise’s Advanced Data Analytics feature to evaluate the optimal vaccine dose determined by the clinical trial team. Dose ID provides justification, references sources, and generates informative graphs illustrating key findings. In doing so, Moderna enhances safety and optimises dosing for testing in clinical trials.
• Novartis started partnerships with various AI companies. In this video Novartis CEO Vasant Narasimhan, shares his views on how AI will impact medicine development.
• ProQR uses AI to find the most interesting and promising applications to fully exploit its Axiomer RNA-editing platform.
• Sanofi has been using AI for years. For example in using neural networks to identify the biological target for treating a disease, and for designing medicines, and graphical models that integrate clinical and molecular data. Thereby improving clinical trials. Some of the tools the company is currently deploying have shortened the traditional innovation cycle with days or months even. In June 2023, Sanofi CEO Paul Hudson stated that Sanofi’s ambition is to “become the first pharma company powered by artificial intelligence.” By using AI, Sanofi improved potential target identification in areas such as immunology, oncology, and neurology by 20-30%. Hudson stated that Sanofi’s ambition is to use this technology to “transform the practice of medicine.” Sanofi had disclosed collaborations with multiple AI-partners.

Sources

• Boston Consulting Group: Unlocking the potential of AI in Drug Discovery
• McKinsey & Company : The potential of AI in drug discovery
• Nature: AI’s potential to accelerate drug discovery needs a reality check
• Nature Biotechnology: AI in drug discovery: Impact and challenges
• PWC: AI in healthcare: How artificial intelligence is poised to transform the industry
• Reuters
• Science: Deep learning for de novo drug discover
• TD Cowen: AI In Drug Development From Hype To The Clinic & Beyond