As enterprise AI adoption surges, the life science industry stands at a pivotal juncture. “We are truly at a tipping point,” remarked Najat Khan, chief data science officer at Janssen, during the 2023 Stanford Drug Discovery Symposium. “It’s amazing how quickly we’ve been able to make progress across the whole value chain in terms of drug development,” she said. “The reality is that there’s a lot of hype, but there is already a lot that’s happening.”
While most Big Pharma companies are exploring AI applications, as a 2022 Clinical Trials Arena article noted. It’s easy to understand why. While developing a given drug might take more than a decade, AI could potentially enable the development of “drugs in one-tenth of the time, from being discovered to being able to treat patients,” wrote McKinsey partner Alex Deverson in late 2022.
Janssen is aiming to distinguish itself by building a team of “bilingual” employees — those proficient in both clinical research and data science. This approach has helped Janssen expand its involvement in AI and data science over the past two and a half years, Khan said. Janssen is “building from the ground up a team that has this expertise,” she added.
AI in drug development: From personalized medicine to manufacturing
Najat Khan
Janssen is applying AI across the entire drug development value chain, from drug discovery and clinical trial design to patient identification and manufacturing optimization. “AI can accelerate disease detection and enable precision medicine,” Khan explained, citing AI algorithms that diagnose pulmonary hypertension and AL amyloidosis up to three years earlier than current methods.
Khan emphasized the potential of AI-driven protein structure prediction and therapeutic design. “The Baker Lab, for instance, uses generative AI models to predict high-affinity protein binders, which serve as a starting point for developing more effective therapeutics,” she said. Janssen also is exploring the use of cell painting, a high-content profiling technique that visualizes cellular components and predicts toxicity before clinical trials.
Search engine traffic for “AI in drug discovery” and “AI in clinical trials.” [Data from Google Trends]
Also in manufacturing, Janssen leverages digital twin technology to improve bi-specific monoclonal antibody production processes. Khan noted that this approach has increased throughput threefold. The company is also applying machine learning to single-cell RNA sequencing data to improve cell therapy manufacturing and gain insights from patient responses.
Optimizing clinical trial design with AI
Khan highlighted the transformative role of AI in clinical trial design. AI has been instrumental in shaping the way Janssen designs and conducts clinical trials, Khan said. “How do you better understand, in a more precise way, if the patient is getting better?” Khan asked.
By analyzing patient data and trial outcomes, AI helps refine criteria, improve patient selection, and identify potential trial pitfalls. Janssen’s Trials360.ai platform uses machine learning to optimize site feasibility, site engagement, and patient recruitment, leading to better patient care.
“This targeted approach not only improves the chances of successful trial outcomes but also ensures that patients receive the most appropriate care for their condition,” Khan said.
With over 100 ongoing AI projects, Janssen employs a scalable, centralized approach to rapidly test and deploy AI technologies. The company’s Med.ai platform exemplifies this approach, facilitating AI integration and collaboration. “We’re fully committed to staying the course and exploring new ways to leverage AI for the betterment of patients,” Khan concluded.
Filed Under: machine learning and AI
