abstract gray background
Blog
Insights Hub

Harnessing AI and Real-World Data: The Future of Clinical Development

Explore how the convergence of real-world data, technology and artificial intelligence is playing a vital role in accelerating drug development.

The pharmaceutical industry is currently experiencing a significant transformation. The convergence of real-world data (RWD), technology and artificial intelligence (AI) is playing a vital role in accelerating drug development. In a recent panel discussion at DIA Global, our experts explored how these elements are reshaping clinical research and drug discovery.

Real-world data paired with machine learning is a game changer in drug development

RWD is becoming increasingly important in influencing the drug development landscape, particularly when used to develop the natural history and patient journey through disease. This can be converted to a clinical trial simulator, which can model a clinical trial before the trial design is finalized and initiated. Running the simulator multiple times (each time modifying various parameters including those related to anticipated efficacy and safety issues — each with its own probability of discontinuation or mortality — and various design parameters of the trial such as number and location of visits) results in an optimal combination of factors. RWD is also used in conjunction with natural language processing (NLP) and machine learning (ML) to automate data collection and cleaning processes. Machine learning is a subset of AI that identifies patterns in large datasets, leveraging identifiable conditions or attributes that are needed to inform sample selection for trials and other prospective data. It identifies cases in instances where diagnosis codes and other information available in each database are too non-specific to do so. With good performance, these algorithms can then be used to maximize use of existing secondary sources. ML is also used to identify predictors of response to a therapy or risk for adverse events (AEs) that can be used to inform value-based contracting or treatment strategies intended to minimize “wasted” use of a drug.

AI offers the biggest expected return on investment in drug discovery

The use of AI in drug discovery is offering significant expected returns on investment. AI techniques such as protein structure modeling, disease representation models and special biological approaches are enabling researchers to identify novel targets and create new molecules with new mechanisms of action.

AI is also proving useful in drug repurposing or indication selection. By associating molecule design with biological targets, AI is helping to expand the label of existing drugs, providing a significant return on investment.

AI approaches are optimizing trial designs

With a significant increase in the number of regulatory submissions to the U.S. Food and Drug Administration (FDA) that is expected to continue to rise, it’s critical to find methods that materially help improve disease outcomes. Data, technology and AI are being used in innovative ways to optimize study design and trial protocol development and predictive analytics. For example, AI can help reduce overall patient and site burden, the number of patients required for trials and trial duration. One panelist shared that a recent meta-analysis of FDA submissions also showed that AI can help reduce cancer mortality on trial by 15-25%.

Decision modeling is also leveraged to empirically measure patient preferences of protocol designs and predict how these will affect a patient’s willingness to participate in an objective way. This further helps to optimize study designs, making them more patient-centric and effective.

The future of AI in drug development

Clinical trials are changing fundamentally and have reached a critical inflection point in the confluence of data fueling AI and the promises and challenges it represents. AI has the potential to disrupt the entire clinical trial process. AI technologies are fueling drug discovery and asset and trial ROI optimization. It can predict the effects of medical compounds to the point where we only need a fraction of the experimental evidence that is normally required.

It is critical for drug developers to partner with a knowledgeable CRO partner like the PPD™ clinical research business of Thermo Fisher Scientific. We know how to navigate the intricacies of new technologies and provide the necessary expertise and integrated approach to maximize the benefits of these transformative tools. Our comprehensive approach, integrating various aspects of drug development, such as clinical development, post-approval activities and market access, ensures optimal utilization of tools and strategies. This leads to improved research efficiencies, faster approvals and access to treatments, and ultimately, reduced time to deliver the right treatment to patients. Further, the ease of working with a single partner eliminates the need to engage multiple service providers, simplifying the drug development process. This ensures consistency and efficiency in utilizing real-world data, technology and artificial intelligence in a rapidly evolving industry.

Ready to learn more about convergence of real-world data and technology for clinical trials?