Pharma companies have been early and active adopters of artificial intelligence (AI) approaches for discovery of new pharmaceuticals and are starting to focus this technology on clinical trials.
Trials, a critical yet complex part of the R&D process, are a major cause of the high failure rates and costs associated with development of new pharmaceuticals. The traditional process spanning 3-6 years with billions spent, no longer meets the requirements of the post pandemic world. The entire process needs to be structured to drive R&D productivity, lower cost and shorten timelines.
This acceleration can only be achieved with the quick adoption of the right technologies. AI as a technology is taking centre stage in spearheading pharma’s transformation to meet this new requirement.
Here we will introduce you to applications of AI that have gained traction and solve challenges in trial design, site selection, recruitment, retention, and trial data handling that have traditionally caused high failure rates and exorbitant costs in R&D.
Applications in Clinical Development
Decision AI applications provide ways to make quicker, better-informed decisions along the path to market and beyond.
- Making trial recruitment faster and more accurate.
Matching patient records to trial criteria can identify the right candidates and reduce screen failure rates. Multiple data sources such as EHRs, medical imaging, and ‘omics’ are mined, analyzed, and interpreted using AI to achieve optimal patient selection.
- Optimizing site selection
Site qualities like administrative setup and clinician experience influence trial timelines as well as data quality and integrity. ML algorithms reduce trial failure risk by analyzing multiple attributes like historical performance, patient burden score, trial competition, and resource availability for optimal selection.
- Faster and more accurate trial data handling
Inaccuracies and inefficiencies abound in manual entry and transfer of data during trial execution. AI, ML systems can enable faster and more accurate trial data capture with intelligent document processing, auto-population of forms, and avoid miscoding or miskeying errors.
- Risk monitoring and outlier detection
AI algorithms can identify sites likely to witness a particular risk behavior and patients at risk for non-compliance. Subject outlier detection improves patient safety monitoring and enables proactive planning for risk mitigation, patient safety, and site performance.
- Safety reporting and Adverse event monitoring
The number of safety incidents is rising, while safety reporting staff is not. AI can deliver easy wins in this case at a much lower time and cost, screening 100% of all calls with a 99% accuracy in detecting adverse events. Implementation is simple, as adverse events reports are in a structured data format, and there is a rich historical record of adjudicated events to enable ML training.
AI will impact the life-sciences industry from molecule to market. By leveraging data science and AI, pharma companies can unlock precise insights that enable decisions for the path to market and beyond.
With such AI-led decision intelligence in place, companies can discover new opportunities in conventional approaches, explore new avenues for competitive differentiation, and quickly, confidently, and innovatively address emerging business challenges – to bring better care to the ultimate stakeholder: patients.
Take the case of how AI seems to have turned R&D on its head; conventional methods for conducting research and navigating through development have changed. The linear and sequential process of clinical development has now become hyper-iterative and integrated. We need not look further than the development of vaccines and the rollout now underway to see how these changes are transforming our ability to respond to current and emerging health challenges.