FDA Publishes Paper Asking Manufacturers for Feedback on AI and Regulation
On March 1, 2023, the US Food and Drug Administration (FDA) released a discussion paper on the use of artificial intelligence in pharmaceutical manufacturing.
In the paper, the FDA outlined its commitment to the promotion of advanced manufacturing techniques. The regulator also admitted that part of this commitment may mean that the regulatory framework surrounding such technologies would need to be updated.
- Three Ways AI and Machine Learning Will Enhance Drug Discovery and Drug Candidate Identification
- Long-Acting Injectables: Machine Learning in Drug Formulation and Therapeutic Development
- “Stability Can be a Problem” — Dealing with Nanoparticle Stability Challenges in Drug Formulation
In a section titled ‘Areas of Consideration Associated with AI’, the FDA outlined some aspects of pharma manufacturing and the regulation that may need to evolve with the development of AI. These areas were identified by the Center for Drug Evaluation and Research (CDER).
Cloud and Edge Computing
Firstly, cloud and edge computing may affect the location of the different software used to control manufacturing equipment, with critical execution software close to the machinery, and supplementary software stored elsewhere on the cloud.
On this issue, the FDA says:
“Data integrity and data quality must be ensured in these environments. While FDA allows the use of third parties for CGMP functions under appropriate oversight by the manufacturer, existing quality agreements between the manufacturer and a third party (e.g., for cloud data management) may have gaps with respect to managing the risks of AI in the context of manufacturing monitoring and control. During inspections, this may lead to challenges in ensuring that the third-party creates and updates AI software with appropriate safeguards for data safety and security.”
The Internet of Things
Secondly, the FDA considers the ‘internet of things’, which has the potential to generate a vast amount of data, previously unforeseen to the regulator.
Current guidance does concern the amount of data and metadata that should be stored by the manufacturer for each batch of drug product. But the paper says that with increased data, “there may be a need to balance data integrity and retention with the logistics of data management.”
Regulatory Oversight
The FDA also suggested that “applicants may need clarity about whether and how the application of AI in pharmaceutical manufacturing is subject to regulatory oversight.”
The paper suggested that AI could be used in monitoring and maintaining equipment, supply chain logistics, and characterising raw materials. Herein, the FDA listed the areas of these applications that would be subject to regulatory oversight.
The regulator also wrote:
“Applicants may need standards for developing and validating AI models used for process control and to support release testing.” They said that there were limited industry standards and FDA guidance available for the development and validation of models that impact product quality. This is problematic when establishing the credibility of a model in particular situations.
Problems with Continuous Evolution of AI Models
Another concern was that AI models can continuously learn, evolving over time. This is contrary to classical models used for manufacturing that are “developed, validated, implemented, and updated as needed through the change control process within the pharmaceutical quality system.”
Here the FDA said that this could produce difficulties in understanding when to notify regulators that the model has changed. It said:
“Applicants may need clarity on: (a) the expectations for verification of model lifecycle strategy and the approach for FDA’s examination of continuously updated AI control models during a site inspection, and (b) expectations for establishing product comparability after changes to manufacturing conditions introduced by the AI model, especially for biological products.”
Public Feedback
The paper concluded with a section that asked manufacturers for feedback on regulatory guidance and AI with regards to the manufacturing, development, and production of drugs. Public comments on the discussion paper will be taken until 1 May 2023.
Get your weekly dose of industry news and announcements here, or head over to our Formulation portal to catch up with the latest advances in manufacture and therapeutic delivery.
