Main menu

Pages

Prominent Pharmaceutical Firms Embrace AI to Accelerate Clinical Trial Processes

 


Pharmaceutical giants are increasingly relying on artificial intelligence to expedite the process of clinical trials. Human trials have historically constituted one of the most expensive and time-consuming phases of drug development, often stretching over several years and incurring costs exceeding a billion dollars from initial drug discovery to final approval.

For a while now, pharmaceutical companies have been experimenting with AI with the hope of harnessing machine-driven insights to uncover groundbreaking drugs. While AI has identified a few potential compounds that are currently in development, the fruition of these endeavors is still years away.

Nevertheless, interviews with numerous CEOs of pharmaceutical companies, regulatory authorities, public health experts, and AI firms, as reported by Reuters, suggest that AI is assuming an increasingly vital role in human drug trials.

Noteworthy players in the industry, such as Amgen, Bayer, and Novartis, are employing AI systems trained on vast datasets encompassing public health records, prescription records, medical insurance claims, and their internal data. This utilization of AI is aimed at expediting the patient recruitment process, potentially halving the time required.

The US Food and Drug Administration (FDA) has received nearly 300 requests to integrate AI or machine learning into drug development between 2016 and 2022. The overwhelming majority of these requests, over 90%, have been submitted within the past two years, with most targeting AI deployment in various phases of clinical development.

Prior to embracing AI, Amgen spent extensive periods sending surveys to doctors across the globe to ascertain the availability of patients with specific clinical and demographic characteristics for trial participation. Often, established relationships with medical facilities and practitioners influenced the selection of trial sites.

However, it's been estimated that nearly 80% of trials fail to meet their recruitment targets due to factors such as clinics and hospitals overestimating patient availability, high attrition rates, or patients not adhering to trial protocols. Amgen's AI tool, known as ATOMIC, now scans extensive internal and public datasets to identify and rank clinics and doctors based on their past success in recruiting trial participants.

Where Amgen once estimated an enrollment period of up to 18 months for mid-stage trials, depending on the disease, ATOMIC has the potential to halve this time under optimal circumstances. Amgen intends to integrate ATOMIC into most of its studies by 2024, with the expectation that AI will trim two or more years from the typical decade-long drug development timeline by 2030.

Despite the potential benefits, it's important to note that less than 25% of health data is currently accessible for research, according to Samir Bhatt, an AI expert at the World Health Organization.

Bayer, a German pharmaceutical company, successfully used AI to reduce the required number of participants by several thousand in a late-stage trial of asondansetron, an experimental drug designed to mitigate the long-term risk of stroke in adults. Typically, pharmaceutical firms seek prior approval from regulatory authorities to utilize external control groups in trials. Bayer is currently in discussions with regulatory bodies, including the FDA, regarding the use of AI in generating external controls for pediatric trials. However, the European Medicines Agency has not received any such requests from companies to use AI in this manner.

Some scientists, including the head of oncology at the FDA, express concerns that drug companies may be overly reliant on AI for establishing external controls across a wide spectrum of diseases. Patients participating in clinical trials often believe they are receiving effective treatment and additional medical attention, potentially leading to an inflated perception of a drug's efficacy.

In light of these concerns, regulatory authorities must maintain vigilance and caution as the use of AI in clinical trials continues to evolve and expand.

Comments