Photo: AI robot hands holding biopharma products/for Taylor Tyden Biospace
As the biopharma industry explores the potential of artificial intelligence, use cases are rapidly emerging. Companies are already trying to use generative AI — think ChatGPT — to improve pharma R&D, from target discovery to drug development to commercialization and regulatory approval for postmarket pharmacovigilance.
Last July, Hong Kong-based Insilico Medicine claimed to be the first company to enter a phase II clinical trial to develop a drug entirely with generative AI. But as with any new and rapidly evolving technology, there are differing perspectives on how and when to best use generative AI, and there is still plenty of skepticism.
According to experts helping to establish AI-based R&D protocols, there are several ways in which the biopharma industry can apply generative AI.
Improving data collection and management
Biopharma researchers have historically relied on data scientists to find information, and those data scientists “then have to figure out whether it's the right data before they can do anything with it.” can begin,” said Alastair Campbell, vice president and global head. of science and technology at R&D-focused software company Dotmatics.
Generative AI changes this equation by automating data collection. Although humans still have to validate and verify the machine's outputs, Campbell said. Biospace that generative technologies can help accelerate leads in discovery and design;
At the JPMorgan Healthcare conference in January, Jean-Philippe Wert, chief research and development officer at AI biotech Okon, said that using generative AI helps his company look at all the data in its search for treatments.
Reuse drugs
Mike King, senior director of product and strategy, technology solutions at IQVIA, noted that Pfizer's Viagra was originally tested for heart problems, but the most popular indication today, erectile dysfunction, was accidentally Discovered. A properly designed algorithm can provide additional clues to evaluate so companies don't have to rely on sheer luck, he said. Biospace.
Administering medications to patients
Rachael Brake, chief scientific officer at life sciences informatics startup Zephyr AI, said AI can help fill patient shortages with emerging compounds, reducing R&D time and costs in the process. “The value proposition of these new treatments is that they're addressing an existing unmet need in the field,” Brake said. “Creating a drug that doesn't actually solve a problem isn't really worth much to anyone.”
Faster discovery
Generative AI can automate the understanding of basic biological processes. For example, King said, “you can train AI on protein structures based on broad-based biology, science and known amino acids, and suggest how certain proteins look based on specific structures. will come.” This information can then be used to develop new drug candidates.
Kimberly Powell, VP of healthcare at NVIDIA, suggested that a biopharma company could use Google DeepMind's AlphaFold 2 protein-folding AI technology to predict protein structures, then develop small-molecule drugs. NVIDIA's MolMIM autoencoder for discovery.
“The concept of algorithmic methods and computational methods to design new drugs is not new,” Campbell said. “I think what's changed a lot is that the methods have improved.”
Clinical supervision
Breck also said AI may be able to scan patient profiles as well as new literature to uncover emerging drug resistance to approved treatments. For example, AI can analyze tumor data “to understand why they might be considered sensitive or resistant to that particular therapy,” Brake said.
King added that drug safety is another area where AI can help parse data. “The ability to combine structured and unstructured data to find potential adverse events and product quality issues, that technology is live today,” he said. “This provided a significant benefit in understanding the aftermarket of product performance, but also in identifying potential critical failure modes, where the volume of data is not necessarily large enough to trigger something”.
Caution remains.
Despite the excitement, there is considerable skepticism and concern about generative AI in the biopharma industry.
Campbell said there is a “stigma” in its use, in part because the work can be vague, and peer reviewers and regulators need to see explainable methods. This can only be overcome if creative AI delivers results that scientists can trust, he said.
The concern extends to job security. “Scientists fear being replaced by robots,” Campbell said. But at this point, he added, “we're so early in the process that it's not realistic.” The regular layoffs we're seeing today in the biopharma space are a reflection of the current economic climate, not a shift in staffing strategy as a result of AI implementation.
Neil Worsell is a former business editor. Biospace. Follow him on LinkedIn or x.