The future is here: Artificial intelligence and clinical trials
Prognosticators expect robots, drones and medical artificial intelligence (AI) to play a more prevalent role in clinical trials this year, thanks to the ability of computers and machines to perform tasks traditionally requiring human thought or intervention.
In the long term, all that should improve the quality, safety and speed to market of emerging therapies. In fact, one report published last year in the International Journal of Clinical Trials asserts such use could halve the cost of some trials.
“If Tesla cars can drive themselves on the roads and park themselves in the garage, why can’t we have a fully connected clinical trial platform that effectively manages all aspects of the clinical trial?” asks Peter Benton of Worldwide Clinical Trials on AppliedClinicalTrialsOnline.com.
Three predictions on how AI will impact trials this year:
- Automated experimentation systems will play a bigger role in trial structure. AI will increasingly be utilized to predict how drugs will interact with targets, allowing clinical trial experimentation to be reduced by up to 70 percent. AI can reportedly predict drug activity and toxicity with much higher precision than other computational methods, partly because its prediction power improves as it inputs experiment results.
The authors of a recent paper published via Carnegie Mellon University discuss how AI can reduce the random nature of certain trials. “We simply cannot perform an experiment for every possible combination of biological conditions, such as genetic mutation and cell type,” they explain. “Researchers have therefore had to choose a few conditions or targets to test exhaustively, or pick experiments themselves.”
Machine learning reduces such uncertainty by applying algorithms that repeatedly choose meaningful experiments based on emerging patterns. The authors point to a recent case in which an algorithm developed a model that was accurate 92 percent of the time, though it only physically conducted 29 percent of the experiments needed for the study. “Our work has shown that doing a series of experiments under the control of a machine learner is feasible even when the set of outcomes is unknown,” they report. “The immediate challenge will be to use these methods to reduce the cost of achieving the goals of major, multi-site projects.” A forerunner in such AI technology is London start-up BenevolentAI, which plans to test the technology in mid-2017 using clinical stage drug candidates licensed from Janssen. Other pharma players edging into the AI world include Berg, Insilico Medicine, Atomwise and Cloud Pharmaceuticals.
- More data will come directly from patient devices. Thanks to the increasing connectedness of personal and/or wearable computing devices, trial data can be captured more easily, automatically, quickly and regularly without the need of human intervention. That’s partly why the wearable technology market is expected to reach $75 billion by 2025. “Collecting clinical trial data exclusively from investigators and at clinical trial sites alone is no longer cost effective or sufficient,” asserts pharma exec Jack Lawler on the Applied Clinical Trials site. “Data collected directly from the patients is a better reflection of what happens in the real-world setting.” Others caution that related privacy and data quality issues must be addressed to ensure the data comes from the patient and is inaccessible to others.
- Clinical trial systems will increasingly move into the cloud. This allows for better communication between people and AI-enabled systems and devices. Worldwide adoption of electronic health records would speed up such integrations, but would also necessitate more consistent privacy and data retention policies.
In short, analysts say the trials of tomorrow may be structured entirely differently to take advantage of the ability of machines to counter their most burdensome and expensive elements. And that can only be good for the patients waiting for lifesaving therapies.
“For progressive organizations and entrepreneurs, this presents a tremendous opportunity to improve upon existing solutions and services or create completely new offerings,” predicts Zikria Syed on the Applied Clinical Trials site. “For others, it will be necessary to adapt to these changes just to survive.”