The transformative power of AI in clinical trials

Hannah Burke our consultant managing the role
Posting date: 24/12/2024
The Transformative Power of AI in Clinical Trials | Proclinical

The transformative power of AI in clinical trials

Artificial Intelligence (AI) has been transforming industries across the globe, from finance to entertainment, it is reshaping the fabric of innovation. AI is revolutionising the life sciences industry, driving exciting breakthroughs in research, diagnostics, personalised medicine, and drug development, and reshaping the future of healthcare. Although still relatively novel, the use of AI in clinical trials has already demonstrated its potential to transform research, reduce errors and expedite approvals by regulators.

Clinical trials are a crucial step in the process of testing and licensing new drugs or treatments for commercial use. However, it has become increasingly difficult to bring new drugs to market. Rising costs and the time it takes to complete a trial are often cited as key challenges for researchers. In this article, we discuss how AI is transforming the design and management of trials and what it could mean for the future of clinical research.

How is AI being used in clinical trials?

AI was first used in the 1970s to identify effective blood infection treatments. However, its use has become more prevalent in clinical trials in the last ten years, with its benefits and potential to transform the field driving this surge.

Common ways in which AI is used in clinical trials are:

  • Trial design
  • Patient recruitment
  • Patient management
  • Trial data collection and analysis
  • Trial design

    Designing a clinical trial is one of the first steps in seeking approval for medical devices and new drugs or treatment. It lays the groundwork for activities at each stage of the trial and other important aspects, including patient recruitment, sample size determination, randomisation, study protocol and funding. Even for a group of experienced researchers, this is often a lengthy process that can take up valuable time.

    Researchers are able to use AI to analyse historical data to inform their planning and design more robust clinical trials.

    Recent examples of AI use in trial design include:

  • HINT (hierarchical interaction network): Developed by a computer scientist at the University of Illinois Urbana-Champaign. It can predict whether a trial will succeed based on the drug molecule, target disease and patient eligibility criteria.
  • SEETrials: Created by Medical Objects in Rosemont, Illinois, which developed a method for prompting OpenAI’s large language model GPT-4 to extract safety and efficacy information from the abstracts of clinical trials. It looks at historical trials and can inform trial designers on what has already been done.
  • As seen in the above examples, machine learning models can simulate various trial designs and predict different outcomes based on information inputted in by researchers. Armed with this information, drug and medical device developers can tailor their designs to generate varying outcomes.

    Plus, because of AI’s versatility, clinical trial coordinators can input interim data and modify protocols in real-time if they need to, leading to more accurate and faster results.

    Patient recruitment

    The most challenging aspect of clinical trials can often be patient recruitment. Due to the hurdles at this stage of clinical trials, patient recruitment takes up around a third of the study’s length, causing many trials to exceed their recruitment deadlines.

    AI can help by analysing large amounts of data to identify potential patients, e.g. looking at social media and other publicly available sources, to identify who meets the trial’s criteria. By using AI, researchers can speed up their recruitment process and guarantee more accurate matches for their patient pool.

    Patient management

    Not only do trial recruiters struggle to find the right candidates, but the biggest issue comes from retaining these patients once they have been enrolled. Some trials can last for years and getting commitment from candidates is crucial. More than 90% of trials are delayed to failure at the enrolment stage, and on average, around a quarter of participants drop out of trials. AI is crucial to helping improve participant retention and can be deployed in several ways to help retain any patients that have been enlisted.

    AI can track real-time data and behaviour from patients through wearable devices and AI-powered apps. Reducing time needed on the site and the need for personnel on clinical trial sites. Some of the benefits for researchers using AI is that they can:

  • Tailor their approach to retain patients by looking at historical behaviours
  • Adapt their strategies to handling patients by using real-time data
  • Analyse videos to ensure that patients take doses correctly and allowing them to intervene when they don’t
  • Trial data collection and analysis

    Another useful application of AI in clinical trials is its ability to monitor data in real-time. Researchers can use AI to track anomalies, trends and anticipated problems before they impact any of the trial’s results.

    A major benefit of AI is its capacity to automate and analyse large amounts of data from multiple data sets, minimising the risk of human error. AI can also be employed is to perform complex data analyses, which has the potential to uncover insights that may have been previously missed by clinical researchers. Researchers can use AI to inform their decisions, making the trial more robust and increasing the accuracy of results.

    Advantages of using AI in clinical trials

    Applying AI to clinical trials offers many advantages, making it an appealing tool for clinicians. Some of the pros of using AI in clinical trials include:

  • Recognising patterns, outliers and anomalies
  • Enhancing recruitment processes and patient retention
  • Increasing accurate trial results
  • Automating routine tasks and reporting
  • Accelerating clinical trial stages
  • Bringing novel drugs and devices to market sooner
  • Challenges of using AI in clinical trials

    Despite AI’s many benefits in clinical trials, including speeding up processes and accelerating drug market entry, researchers still face challenges. Some of the main challenges that clinicians encounter when using AI in clinical trials are:

    • Results using large amounts of training data may be difficult to reproduce
    • Researchers could become too dependent on AI
    • AI models that are complex can be hard to understand
    • Guidelines surrounding AI are still developing and can’t account for all issues that may arise
    • Biases in training data are more likely to arise

    AI can also pose an ethical risk to participants and researchers, such as:

    • Large patient datasets could violate patient privacy
    • Security risks are increased
    • Responsibility for issues that emerge during the trial can become unclear

    To manage and reduce these risks, it is important that researchers outline clear protocol and policies to ensure patient safety and compliance.

    Looking ahead: The potential of AI in clinical trials

    The use of AI offers significant potential to transform clinical trials in unprecedented ways. It can revolutionise trial design, patient recruitment, patient management, and trial data collection and analysis. While AI holds many benefits, it is still in its infancy and poses challenges and ethical risks that researchers are working to overcome. Despite these hurdles, AI will play a key role in helping companies bring drugs and treatments to market faster and more cost-effectively.

    Expand your clinical research team with Proclinical

    Our team of hiring experts know how to locate the best talent. We work with clinical research companies across the globe to place highly-skilled researchers to companies driving innovation to fill role gaps. Submit a vacancy and we will get started on looking for the perfect candidate, or if you would like to find out more about what we can do for you, get in touch.

    Find your next research opportunity with Proclinical

    Here at Proclinical, we pair you with the best opportunities to develop in your clinical research or life sciences career. You can browse our clinical research positions or submit your CV, and one of our experts will contact you about our latest job openings. Take the next step in your career with Proclinical and unlock your potential today.

    close