The use of artificial intelligence and machine learning in drug development is revolutionizing the pharmaceutical industry.
Recognizing this, the US Food and Drug Administration plans to release a draft FDA’s Drug Guidance later this year on integrating AI/ML in to drug development processes
This forthcoming guidance is poised to provide a structured framework for incorporating AI/ML technologies, aligning with the objectives outlined in President Biden’s Executive Order 14110.
FDA’s Drug Guidance
Artificial intelligence and machine learning have become integral tools in modern drug development, offering unprecedented capabilities in areas ranging from drug discovery to post-marketing safety surveillance.
The FDA has observed a marked increase in submissions incorporating AI/ML components over the past several years, highlighting the technology’s growing role in the pharmaceutical industry.
Integration of AI/ML into drug development was minimal, with only one investigational new drug application featuring AI elements in 2016.
By 2021, this number had surged to 128 submissions, showcasing the rapid adoption and potential of AI/ML technologies. The applications of AI/ML span various therapeutic areas, with oncology leading the way, followed by psychiatry, gastroenterology, and neurology.
To better understand and regulate the use of AI/ML, the FDA has actively engaged with industry stakeholders through discussion papers and public forums.
In 2023, the agency released two discussion papers aimed at exploring the integration of AI/ML in drug development and pharmaceutical manufacturing.
These papers solicited feedback from various organizations, providing valuable insights that will inform the upcoming guidance.
The planned guidance is also a response to President Biden’s Executive Order 14110, issued in October 2023, which mandates that government agencies, including the Department of Health and Human Services, define objectives and goals for regulating AI/ML across different phases of drug development.
FDA’s Upcoming Guidance
The FDA’s upcoming draft guidance on the use of AI/ML in drug development is set to provide a comprehensive framework informed by the agency’s experience in reviewing over 300 submissions containing AI components and feedback from industry stakeholders.
This guidance will address various aspects of AI/ML integration in drug development, ensuring these technologies are used effectively while maintaining safety and efficacy standards.
Tala Fakhouri, associate director for policy analysis in the Office of Medical Policy at the FDA’s Center for Drug Evaluation and Research, emphasized the significance of this guidance during the Regulatory Education for Industry conference.
The guidance aims to fulfill a mandate from President Biden’s Executive Order 14110, which calls for defining regulatory objectives and goals for AI/ML in drug development.
AI/ML is increasingly used in various stages of drug development, including drug discovery, nonclinical studies, dose-finding studies, clinical research, site selection, participant recruitment, real-world data analysis, advanced pharmaceutical manufacturing, and post-marketing safety surveillance.
The FDA’s guidance will provide clarity on the scope of the agency’s oversight, how to operationalize a risk-based approach, ensure transparency, and the level of detail and documentation required in AI/ML applications.
Stakeholder feedback has been crucial in shaping this guidance. The FDA received 800 comments from 60 organizations on its discussion papers, with many urging the agency to harmonize its AI regulatory requirements with those of other global regulators and align with the Center for Devices and Radiological Health.
There is a call for partnerships to advance the creation and sharing of machine-readable datasets for drug development.
This guidance will also be developed with input from FDA’s medical product centers, the Oncology Center for Excellence, and the Office of Combination Products.
AI/ML in Drug Development
AI/ML is becoming increasingly pivotal across all stages of drug development. These technologies are being used to identify drug targets, enhance nonclinical studies, and optimize dose-finding studies.
In clinical research, AI/ML aids in site selection, participant recruiting, and retention, significantly improving the efficiency and effectiveness of these processes.
The FDA has observed a broad application of AI/ML, particularly in oncology, psychiatry, gastroenterology, and neurology. The use of AI extends to analyzing real-world data, which helps in understanding drug performance in broader patient populations outside of controlled clinical trials.
AI/ML also plays a crucial role in advanced pharmaceutical manufacturing, ensuring more precise and efficient production processes.
Post-marketing safety surveillance is another critical area where AI/ML is utilized. These technologies help monitor the safety and efficacy of drugs once they are on the market, providing real-time data and analysis that can lead to quicker identification of adverse effects and other safety concerns.
FDA’s Experience
The FDA has accumulated significant experience in reviewing drug development submissions that incorporate AI/ML components.
Since 2016, there has been a substantial increase in the number of submissions, with over 300 received to date.
These submissions have spanned various therapeutic areas, including oncology, psychiatry, gastroenterology, and neurology, reflecting the broad applicability and potential of AI/ML technologies.
Industry feedback has played a crucial role in shaping the FDA’s approach to AI/ML in drug development. The agency received 800 comments from 60 organizations on its discussion papers about incorporating AI into pharmaceutical manufacturing and drug development.
These comments highlighted the need for clarity on the scope of FDA’s oversight, how to implement a risk-based approach, ensuring transparency, and the level of detail and documentation required in AI/ML applications.
Stakeholders also emphasized the importance of harmonizing the FDA’s AI regulatory requirements with those of other global regulators.
They called for aligning these regulations with those of the Center for Devices and Radiological Health and establishing partnerships to advance the creation and sharing of machine-readable datasets for drug development.
This feedback will be instrumental in informing the FDA’s upcoming guidance, ensuring it addresses the industry’s concerns and facilitates the effective and safe use of AI/ML in drug development.
Regulatory Framework
The FDA is committed to developing a flexible, risk-based regulatory framework that promotes the innovative use of AI/ML in drug development while safeguarding public health.
This framework will address the unique challenges posed by AI/ML technologies, ensuring that they are used effectively and safely across various stages of drug development.
A key component of this framework is the operationalization of a risk-based approach. This means that the level of regulatory oversight will be proportional to the potential risks associated with the use of AI/ML.
The FDA aims to ensure transparency in how AI/ML is used, requiring detailed documentation and robust evidence to support AI/ML components in submissions.
The FDA’s upcoming guidance will also provide best practices for the use of AI/ML, ensuring consistent and reliable application of these technologies across the industry.
This includes developing consistent terminologies related to AI/ML to avoid confusion and provide clear communication among stakeholders.
Collaboration with other regulatory bodies is another important aspect of the framework. Harmonizing AI regulatory requirements globally will help create a cohesive regulatory environment, facilitating international cooperation and the sharing of best practices.
The FDA plans to work closely with its medical product centres, the Oncology Center for Excellence, and the Office of Combination Products to develop this guidance, ensuring it is comprehensive and addresses the needs of all stakeholders involved in drug development.
Future Directions
Looking ahead, the FDA’s forthcoming guidance on AI/ML in drug development is poised to be a pivotal document, providing a clear regulatory pathway for the integration of these advanced technologies.
Beyond the guidance, the FDA plans to develop a set of best practices to ensure the consistent and effective use of AI/ML across the industry.
This will include establishing clear and consistent terminologies related to AI/ML, which will aid in standardizing practices and fostering clearer communication among stakeholders.
The guidance will be informed by continuous stakeholder feedback and the FDA’s evolving experience in reviewing AI/ML submissions.
By incorporating input from various medical product centres and industry stakeholders, the FDA aims to create a robust and flexible regulatory framework that supports innovation while protecting public health.
The significance of this guidance extends beyond the regulatory sphere; it underscores the broader trend of increasing AI/ML integration into the pharmaceutical industry.
As AI/ML technologies continue to evolve, their potential to revolutionize drug development—from discovery through to post-market safety monitoring—becomes increasingly evident.
FDA’s upcoming guidance will play a crucial role in shaping the future of drug development, ensuring that AI/ML can be harnessed effectively and safely.
