How AI and machine learning are shaping the future of R&D in the pharmaceutical industry

5 advantages of AI and ML in the R&D of the pharmaceutical industry

Advantage n°1: Faster and novel drug design thanks to Generative AI

Generative AI is driving significant breakthroughs in drug discovery by enabling faster and more precise identification of viable drug candidates. A prime example is AlphaFold 3, recently unveiled by Google DeepMind, which can predict the structures and interactions of proteins, DNA, RNA, and other biomolecules with unprecedented accuracy.¹https://firstwordpharma.com/story/5854987 AlphaFold 3 has the potential to transform drug discovery, particularly in modelling how various biomolecules interact. However, Chai Discovery’s new AI model, Chai-1, has outperformed AlphaFold 3 in some benchmarks, offering more accurate protein folding predictions with fewer data requirements and being available for commercial use, unlike AlphaFold.²https://firstwordpharma.com/story/5894918

Generative AI, especially deep learning models, has become a key focus in pharmaceutical collaborations. Companies like Absci and Isomorphic Labs are leveraging AI to design novel molecules and antibodies. For instance, in early 2024, Absci partnered with AstraZeneca to use its generative AI platform to develop an antibody drug for a specified oncology target​.³https://pmlive.com/pharma_news/astrazeneca_and_generative_ai_company_absci_enter_247m_oncology_partnership_1504590/#:~:text=AstraZeneca%20(AZ)%20and%20Absci%20have,for%20a%20specified%20oncology%20target. Moreover, collaborations such as the one between Isomorphic Labs and Eli Lilly aim to leverage AI technologies to identify small-molecule candidates for various therapeutic targets. This partnership highlights the ability of AI to transform how we approach previously hard-to-drug targets, enabling faster, data-driven discoveries​.⁴https://www.nature.com/articles/d43747-024-00084-w\

Advantage n°2: Improved Success Rates in Drug Discovery

One of the major hurdles in drug R&D has been the high failure rate, especially during the clinical phase. Companies like Insilico Medicine, Exscientia, and BenevolentAI are actively using AI to model complex biological systems and predict how different molecules interact with biological targets, helping to improve success rates. AI can now predict whether a drug candidate is likely to succeed in human trials, reducing the risk of costly failures later in the development process. These AI tools simulate biological environments, allowing researchers to test drug candidates more thoroughly in silico before moving to expensive laboratory and clinical testing​.⁵https://www.sciencedirect.com/science/article/pii/S266732582400205X

For instance, the first AI-designed drug, INS018_055, developed by Insilico Medicine, has entered Phase 2 clinical trials. This potentially first-in-class small molecule inhibitor targeting idiopathic pulmonary fibrosis was discovered and designed in just 30 months which is lower than the industry standard of 10-15 years. This milestone signals AI’s increasing role in advancing molecules to clinical phases faster, with improved targeting and a greater chance of success​.⁶https://www.clinicaltrialsarena.com/news/insilico-medicine-ins018055-ai/

Advantage n°3: Revolutionizing Drug Discovery for Complex Diseases

AI is proving particularly beneficial in drug discovery for complex diseases that have been traditionally difficult to target, such as neurological conditions and autoimmune diseases. AI can help identify novel targets and design drugs for “hard-to-drug” targets.

In oncology, for instance, AI-driven platforms are identifying innovative therapeutic antibodies and small molecules that may treat various cancer types, including hormone-sensitive cancers such as ER+/HER2- breast cancer.⁷https://www.menarini.com/en-us/news/news-detail/menarini-group-and-insilico-medicine-enter-global-exclusive-license-agreement-for-novel-kat6-inhibitor-for-potential-breast-cancer-treatment-and-other-oncology-indications This is a step forward in the precision treatment of diseases where one-size-fits-all therapies often fall short​.

Advantage n°4: Enhancing Clinical Trial Efficiency

AI is increasingly being used to streamline clinical trials by improving patient selection, predicting outcomes, and monitoring trial progress. This approach can significantly lower costs and reduce the time needed for trials. For example, AI can analyze large datasets to identify which patients are most likely to respond to a treatment, which optimizes the inclusion of patients in trials. This not only improves the likelihood of trial success but also accelerates drug approval.⁸https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9974218/

Companies such as IQVIA are pioneering these efforts by applying AI and ML in clinical trials to reduce manual intervention and increase trial efficiency. Their systems use AI to automate data capture, enhance patient recruitment, and optimize trial designs based on real-time data.⁹https://www.iqvia.com/blogs/2024/02/transforming-clinical-trial-workflows-with-ai In addition, Oracle Health Sciences applies AI to its cloud-based clinical trial management systems, automating data collection, patient monitoring, and protocol adjustments.¹⁰https://www.oracle.com/life-sciences/clinical-trials/ These companies are driving major innovations in clinical trials, allowing for more adaptive trial designs and reducing the time and cost associated with manual trial processes.

Advantage n°5: Economic and Strategic Implications

The adoption of AI in drug R&D is also being driven by the economic need to reduce costs and increase the return on investment (ROI) in pharmaceutical research. The global pharmaceutical industry spends over $250 billion annually on R&D, and AI-driven tools are expected to save up to $50 billion per year by 2026.¹¹https://www.cowen.com/insights/ai-in-drug-development-from-hype-to-the-clinic-beyond/ By accelerating drug discovery timelines and reducing trial failures, AI can extend the time drugs can stay on the market before their patents expire, which enhances profitability​.

Pharmaceutical giants like Bristol Myers Squibb and Novartis are making large investments in AI platforms, as seen with Bristol Myers’ $674 million deal with VantAI and Novartis’ collaboration with Isomorphic Labs.4 These deals highlight the growing trend of integrating AI into the core drug discovery and development pipeline, positioning AI not just as a tool but as a strategic pillar for the future of pharmaceuticals​.

Discover how Nautilus.ai, our dedicated activity in AI and Digital solutions for Life Sciences players can support you in your projects >

What are the challenges to the full integration of AI and ML in the R&D of the pharmaceutical industry?

AI and ML have great potential in pharmaceutical R&D, but several key barriers hinder its full integration:

1. Regulatory Compliance and Safety: Adhering to stringent regulations while integrating AI into drug development is challenging, as regulatory bodies are still adapting to AI-driven processes.
2. Ethical and Privacy Concerns: Maintaining patient privacy and preventing algorithmic bias are critical, particularly with sensitive health data.
3. Data Quality: AI requires vast, high-quality datasets, but existing data is often fragmented and inconsistent, impacting accuracy.
4. Technical Integration: Legacy systems in pharmaceutical companies can be incompatible with new AI technologies, making adoption difficult and costly.
5. Employee Literacy: Many staff lack the necessary AI skills, requiring extensive training to use AI tools effectively.
6. Complexity and IP Issues: The intricate pharmaceutical landscape and unclear intellectual property rights for AI-generated innovations further complicate AI adoption.

While the integration of AI and ML into drug R&D holds promise for transforming the pharmaceutical industry, questions remain about its true return on investment (ROI) and practical impact. Despite significant advances, recent perspectives argue that the initial hype surrounding AI in this field may have been overly optimistic, with some now scrutinizing its real-world use cases and measurable ROI.

However, ongoing collaborations with tech and biotech companies continue to push forward innovations, aiming to reduce costs, improve success rates, and bring effective treatments to market more efficiently. As these technologies mature, their success in achieving tangible, scalable benefits will determine the extent to which they truly revolutionize pharmaceutical R&D. Alcimed can help you navigate the evolving landscape of AI and ML in drug R&D. Don’t hesitate to contact our team to explore how you can integrate these cutting-edge technologies into your R&D workflows.

About the author,

Mikka, Consultant in Alcimed’s healthcare team in Germany