Artificial Intelligence has the Attention of Regulators

“The regulatory uses of AI are real: In 2021, more than 100 drug and biologic applications submitted to the FDA included AI/ML components. These submissions spanned a range of therapeutic areas, and sponsors incorporated the technologies in different developmental stages,”  wrote Patrizia Cavazzoni, M.D., Director of the CDER at FDA in a recent FDA Voice article.

The FDA has issued two discussion papers on the use of AI.  As a follow-up to the discussion papers , the Agency is planning a workshop to discuss how the community can work together to realize the potential of AI/ML for product development while being mindful of potential challenges.

“AI technologies are important in drug manufacturing because they can enhance process controls, identify early warning signals, and prevent product losses.”

-Patricia Cavazzoni, M.D., Director of CDER, US FDA

Discussion Paper 1: Using Artificial Intelligence and Machine Learning in the Development of Drug and Biological Products

This discussion paper specifically states many of the AI/ML scientific and regulatory science principles outlined in the document may be applicable across all medical products, including in the development of medical devices intended to be used with drugs (including, but not limited to, combination products, companion devices, and complementary devices).

The discussion paper also includes references to how AI/ML is being applied to real-world data (RWD) and data from digital health technologies (DHTs) in support of drug development. Some of the general challenges and considerations with utilizing AI/ML in different drug development use cases are also discussed, including overarching standards and practices for the use of AI and ML; and considerations and practices for AI/ML in drug development.

Current and potential use cases for AI and ML outlined in the discussion paper include:

  • Drug Discovery – uses of AI/ML for drug target identification, selection, and prioritization, as well as compound screening and drug design in drug discovery.
  • Clinical Research – uses of AI/ML to inform the design and efficiency of non-traditional trials such as decentralized clinical trials, and trials incorporating the use of real-world data (RWD) extracted from electronic health records (EHRs), medical claims, or other data sources.
  • Nonclinical Research – current efforts to explore the use of more novel AI/ML algorithms (e.g., artificial neural network models and tree-based models) for PK/PD modeling.
  • Post Market Safety Surveillance – use of AI/ML to detect and evaluate drug event associations from literature and to screen social media for adverse events.
  • Advanced Pharmaceutical Manufacturing – the use of AI/ML in the optimization of process design, advanced process control, smart monitoring and maintenance, and trend monitoring.

FDA’s CDER, CBER, and CDRH, including Digital Health Center of Excellence (DHCoE), aim to initiate a discussion with stakeholders and solicit feedback on three key areas in the context of AI/ML in drug development:

1) Human-led governance, accountability, and transparency

2) Quality, reliability, and representativeness of data

3) Model development, performance, monitoring, and validation

    Discussion Paper 2: Artificial Intelligence in Drug Manufacturing

    In this discussion paper, the agency specifically states the areas of consideration presented in the discussion paper focus on the manufacture of drug products that would be marketed under a New Drug Application (NDA), Abbreviated New Drug Application (ANDA), or Biologics License Application (BLA).

    Potential use cases for AI in pharmaceutical manufacturing include:

    • Process design and scale up
    • Advanced process control
    • Process monitoring and fault detection
    • Trend monitoring

    The Agency is seeking answers to a list of eight specific questions when it comes to AI in drug manufacturing:

    1. What types of AI applications do you envision being used in pharmaceutical manufacturing?
    2. Are there additional aspects of the current regulatory framework (e.g., aspects not listed above) that may affect the implementation of AI in drug manufacturing and should be considered by FDA?
    3. Would guidance in the area of AI in drug manufacturing be beneficial? If so, what aspects of AI technology should be considered?
    4. What are the necessary elements for a manufacturer to implement AI-based models in a CGMP environment?
    5. What are common practices for validating and maintaining self-learning AI models and what steps need to be considered to establish best practices?
    6. What are the necessary mechanisms for managing the data used to generate AI models in pharmaceutical manufacturing?
    7. Are there other aspects of implementing models (including AI-based models) for pharmaceutical manufacturing where further guidance would be helpful?
    8. Are there aspects of the application of AI in pharmaceutical manufacturing not covered in this document that FDA should consider?

    Industry Stakeholders Have Been Quick to Respond

    The discussion paper focused on Artificial Intelligence in Drug Manufacturing (FDA-2023-N-0487) has received 35 comments since the date of this post.

    FDA posted comments it received on AI in Drug Manufacturing from The National Institute for Innovation in Manufacturing Biopharmaceuticals (NIIMBL). In response to Question 8 regarding aspects of the application of AI in pharmaceutical manufacturing not covered in this document that FDA should consider, NIIMBL suggests ‘FDA consider mechanisms to train reviewers and site assessors on the use of AI-based models and other AI applications in drug manufacturing. It might be helpful to establish an expert AI team in ORA to be available for AI CMC-related consults.’

    The Parenteral Drug Association (PDA) has also already submitted its comments. In response to Question 8, PDA suggested a total of nine aspects the agency should consider. PDA writes, ‘Guidance developed for AI applications should consider the risk to “overthink” decisions by trending more data than what the situation or process actually requires (“Analysis Paralysis”). This might be counterproductive to the desired process improvement as it could be better handled by a well-trained and experienced human operator.’

    The International Society for Pharmaceutical Engineering has also submitted its comments. In its opening statement ISPE suggests, “This subject document should serve as a starting point for engagement between industry and regulators to give orientation and guidance on the use of recent technologies in connection with digital transformation. Artificial Intelligence (AI) and Machine Learning (ML) are both evolving fields that may require specific guidance to be used effectively. International harmonization on guidance associated with Artificial Intelligence is strongly recommended.”

    Axendia’s Take

    These discussion papers serve as an excellent example of the FDA removing industry excuses for the inability to adopt modern technologies. We have warned life science companies about trying to catch up with regulators when it comes to the adoption of modern technology. 

    CDER established the Framework for Regulatory Advanced Manufacturing Evaluation (FRAME) initiative to prepare a regulatory framework to support the adoption of advanced manufacturing technologies that could bring benefits to patients.​  FRAME focuses on:

    • End-to-End Continuous Manufacturing (E2E CM): A fully integrated process into which raw materials are continuously fed and transformed and finished drug products are continuously removed.​
    • Distributed Manufacturing (DM): A decentralized manufacturing platform with manufacturing units that can be deployed to multiple locations.​
    • Point-of-Care Manufacturing (POC): A DM platform with manufacturing units in proximity to patient care; for example, at healthcare facilities.​
    • Artificial Intelligence (AI): Software and hardware systems that can perceive the environment, interpret data, and decide actions.

    Through FDA’s collaborative Emerging Technology Program, industry representatives can meet with Emerging Technology Team (ETT) members to discuss, identify, and resolve potential technical and regulatory issues regarding the development and implementation of a novel technology prior to filing a regulatory submission.

    Think the FDA isn’t already on the right side of the digital divide? See FDA’s Digital Transformation Journey to learn more.

    AI/ML are undeniably poised to revolutionize our industry. Connect with us for guidance on the appropriate use cases to meet your specific needs.

    Contact research@axendia.com to schedule an Analyst Inquiry on this topic.

    Resource Corner

    Both discussion papers can be accessed below:

    Using Artificial Intelligence and Machine Learning in the Development of Drug and Biological Products.   

    Artificial Intelligence in Drug Manufacturing.

    The opinions and analysis expressed in this post reflect the judgment of Axendia at the time of publication and are subject to change without notice. Information contained in this post is current as of publication date. Information cited is not warranted by Axendia but has been obtained through a valid research methodology. This post is not intended to endorse any company or product and should not be attributed as such.

    Scroll to Top
    Share via
    Copy link