Transforming Clinical Trials through the Power of AI

Prakriteswar Santikary, PhD | |

The stakes are higher than ever in clinical research. The clinical development marketplace has become more competitive, with stricter regulatory standards and greater emphasis on trial oversight and patient safety. 

At the same time, the costs and time needed to commercialize a new drug are escalating – exceeding 8 years and over $2 billion. 1 A big factor in these staggering rates is the increasing complexity of clinical trials, driven in large part by study designs needing more endpoints to demonstrate product value, including data from mHealth devices.

It takes 8 years and costs $2billion to commercialize a drugAgainst this backdrop, the clinical trials industry needs disruption more than ever before.

Enter Artificial Intelligence

This is where Artificial Intelligence (AI) and emerging technologies such as the cloud and Data Lakes come in. By cutting costs, improving data quality, and reducing trial durations, AI, machine learning and deep learning techniques are driving clinical trial efficiencies, optimizing end-to-end clinical trial processes, and helping pharma companies bring new drugs and therapies to the market faster.

Artificial intelligence can disrupt every stage of the clinical trials processAI offers a number of significant uses that can disrupt every other stage of the clinical development process as well — from finding biomarkers and gene signatures that cause disease to bringing new diagnostic tools and treatments for terminal illnesses.

By augmenting and assisting human intelligence, leveraging data and making clinical trial predictions to detect trends, risks and outcomes, AI combined with big data holds the potential to solve even more of today’s key clinical trial challenges.

How AI can transform Clinical Trials:  3 Examples

AI-driven uses cases can be grouped into three broad categories. Let’s review them here:

AI-driven protocols make clinical trials more intelligent

  • Make Clinical Trials Intelligent. AI-driven protocol designs powered by AI algorithms and deep learning techniques can make clinical trials more intelligent in numerous ways (See figure). And, AI can provide pharmaceutical researchers with additional predictive data that can help to determine whether taking a drug will result in a positive or negative outcome, whether patients will drop out of studies, and whether trials will be successful or not. Using predictive algorithms and AI techniques, trial risks can be mitigated well before they become big problems, thereby keeping trials on track and on budget.
  • Optimize Clinical Trial Processes. With nearly half of all investigative sites under-enrolling and approximately 30% of patients dropping out of clinical trials, 2 patient recruitment and retention are among the top challenges facing the clinical trial industry today. AI can help pharma to address both of these challenges. By extracting pertinent EMR information, sifting through physicians’ notes, reading binary data from images and medical scans and comparing them to a study’s inclusion and exclusion criteria, AI can more efficiently and effectively identify the appropriate patients for clinical trial enrollment. And, during trials, AI can help by predicting which patients are at risk of dropping out by utilizing Real world Evidence (RWE) from medical claims, labs and prescription data.
  • Gain Greater Insight for Smarter Decision Making.The promise of detecting patterns using AI, machine learning and deep learning techniques when managing vast volumes of clinical research data and using it to accelerate drug discovery is tantalizing. AI can ingest millions of words of text, molecules, genomic sequences, and images in minutes, aggregate the data and devise hypotheses beyond human ability. So, with these tools, biotechnology or pharma companies could combine all their early stage data to determine in which indication a drug is more likely to succeed.

Conclusion

It’s not uncommon for researchers to think ‘Why change it when it’s working?”  But in clinical development, it’s not.  Less than 10% of trials end on time2 and the costs to develop new drugs are sky-rocketing.

But, AI can help to reverse these trends and enable sponsors to optimize clinical trials and accelerate new product development. From maximizing patient recruitment and retention to improving data collection and risk monitoring, Artificial intelligence has the potential to disrupt every stage of the clinical trials process. Researchers who embrace these new technologies can dramatically shorten time to market for life-saving drugs and deliver a huge win for the patients who need them most.

 

Click here to learn how a modern, cloud-based data platform that delivers real-time analytics can keep your clinical trials on track and on budget.

 

Prakriteswar Santikary, PhD is the Chief Data Officer at ERT

References:

  1. http://csdd.tufts.edu/news/complete_story/pr_tufts_csdd_2014_cost_study
  2.   http://csdd.tufts.edu/news/complete_story/pr_tufts_csdd_2014_cost_study
Share