FDA issues Draft Guidance on the use of AI in Drug and Biological Product Development

The US Food and Drug Administration (FDA) recently provided a draft guidance on the application of artificial intelligence (AI) in regulatory decision making. By championing a balance between innovation and safety, this initiative highlights the significance of AI while upholding its rigorous scientific and regulatory standards.

Entitled ‘Considerations for the Use of Artificial Intelligence To Support Regulatory Decision-Making for Drug and Biological Products’, the guidance lays the foundation for integrating AI into the evaluation of drug and biological product safety, efficacy, and quality. It offers a structured framework for ensuring AI model credibility, emphasizing risk based approaches and robust public engagement.

Highlighting the transformative potential of AI, FDA Commissioner Robert M. Califf, MD, remarked that these advancements could redefine clinical research and medical product development, ultimately enhancing patient outcomes. As AI-driven innovations gain momentum—spanning applications from disease modeling to real-world data analytics—the FDA’s proactive approach seeks to inspire confidence and catalyze progress in the future of healthcare.

Key Highlights from FDA Draft Guidance:

• Risk Based Credibility Assessment:
  The guidance introduces a risk based framework for evaluating the credibility of AI models. This approach ensures the validation and evidence provided align with the intended use and potential impact of the AI model on regulatory decisions.
• Transparency and Documentation:
  Sponsors must provide clear and comprehensive documentation of the AI model, including its design, intended use, training, testing, and performance characteristics. This ensures regulatory transparency and facilitates an understanding of how the AI contributes to decision-making.
• Context of Use:
  AI model credibility is evaluated within its specific context of use (e.g., drug development, manufacturing, or post-marketing surveillance). This context determines the validation standards and acceptable risk levels for the AI system.
• Algorithm Development and Testing:
  The FDA emphasizes the importance of robust training, testing, and validation datasets that are representative of the target population and use cases. Clear metrics and methodologies for assessing performance should be provided.
• Bias and Fairness Considerations:
  Sponsors should proactively assess and mitigate potential biases in AI models to ensure equitable outcomes for diverse populations, particularly in clinical trial data and patient-centric applications.
• Lifecycle Approach to AI Model Management:
  The FDA recommends a lifecycle management approach, ensuring AI systems are continuously monitored, updated, and revalidated post-deployment to maintain accuracy and effectiveness.
• Stakeholder Collaboration:
  Encourages dialogue between sponsors, regulatory agencies, and other stakeholders to clarify expectations and resolve uncertainties about AI integration in regulatory contexts.

Source: Considerations for the Use of Artificial Intelligence To Support Regulatory Decision-Making for Drug and Biological Products | FDA

Navitas Life Sciences: Driving Innovation in Life Sciences

Navitas Life Sciences is a global, full-service Clinical Research Organization (CRO) dedicated to accelerating drug development and improving healthcare outcomes. With over three decades of expertise, we provide comprehensive support for clinical trials, regulatory consulting, real world evidence generation, post marketing studies, and pharmacovigilance services.

Our services empower pharmaceutical, biotech, and medical device companies to integrate advanced technologies, including AI and machine learning, ensuring compliance, efficiency, and innovation across the drug development lifecycle. By delivering tailored solutions, we enable our clients to achieve faster regulatory approvals, optimize clinical trial outcomes, and drive patient-centric advancements in healthcare.

To learn more about our services and solutions, reach out to us at This email address is being protected from spambots. You need JavaScript enabled to view it.