A recent analysis titled “How successful are AI-discovered drugs in clinical trials? A first analysis and emerging lessons,” published in the June issue of Drug Discovery Today, gave insights into a topic that is becoming increasingly important for the pharmaceutical R&D sectors: the role of Artificial Intelligence (AI) in drug and new molecule discovery. In the decade since the advancement of symbolic and generative AI solutions, the number of AI-discovered drug and vaccine molecules has increased substantially.
Accelerating drug discovery and shortening clinical trial timelines
The authors of the analysis, affiliated to the Boston Consulting Group, conducted a first analysis of the clinical pipelines of AI-native Biotech companies; their findings indicate that AI promises to revolutionize drug discovery. They write: “If we take, at face value, the success rates we observed for AI-discovered molecules in Phase I and II and assume that these hold in the future, and if we combine these with historic Phase III success rates, a striking picture emerges: the probability of a molecule succeeding across all clinical phases end-to-end would increase from 5–10% to ∼9–18%. This would represent almost a doubling of pharmaceutical R&D productivity overall, which would bring enormous benefits.”
In a recent piece VP, Open Research at Digital Science, Mark Hahnel provides insights which reflect the findings of the analysis. Writing about the role of AI in drug discovery and clinical trials, he highlights the work done by companies such as Exscientia which developed a clinical pipeline for AI-designed drug candidates. He points out that the platform generates highly optimized molecules that meet the multiple pharmacology criteria required to enter a compound into a clinical trial, in addition to cutting the industry average timeline for molecule development from 4.5 years to just 12 to 15 months. “These companies have the technical know-how to build the models, and most likely some internal data with which to train them on. But they need more,” he writes.
Drawing from existing data
This is where tools like Dimensions Knowledge Graph, powered by metaphactory, which demonstrate the potential of structured data, and OntoChem solutions, which transform unstructured data into actionable insights through semantic normalization to turn data into knowledge, come in. For example, Dimensions Knowledge Graph, powered by metaphactory, can tap into data and deliver insights derived from global research and public datasets, represented in 32 billion statements in the knowledge graph. “Connecting internal knowledge with such vast external data provides a trustworthy, explainable layer for AI algorithms, enhancing their application across the pharmaceutical value chain,” writes Hahnel.
The authors from the Boston Consulting Group in their conclusion highlight the potential of AI in drug discovery to deliver more innovative medicines to patients more quickly, efficiently, and affordably. They write that while the impact of these (AI drug discovery) techniques is already evident, particularly in speeding up and reducing costs in preclinical workflows, there are also indications that similar benefits are beginning to emerge in clinical trials as well. Hahnel, in his piece, emphasizes that pharmaceutical companies should deepen their collaboration with open academic data aggregators to improve metadata quality and utilize highly curated linked datasets. “The limiting factor is not the AI capabilities, it is the amount of high-quality, well described data that they can incorporate into their models,” he writes.
Do you want to learn more on how to tap into the potential of data in your organizations? Please reach out to the teams to learn more about Dimensions Knowledge Graph and OntoChem.
Related reading: