FDA Issues Artificial Intelligence/Machine Learning (AI/ML)-Enabled Device Software Functions Draft Guidance

The U.S. Food and Drug Administration recently issued its draft guidance entitled “Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence/Machine Learning (AI/ML)-Enabled Device Software Functions.” The draft guidance follows the passage of the Food and Drug Omnibus Reform Act of 2022 (FDORA), which explicitly authorized the Agency to approve or clear Predetermined Change Control Plans (PCCPs).

We summarize some of the key takeaways from FDA’s draft guidance.  Read the client alert here.

 

 

 




The European Commission Proposes First Major Overhaul of the EU Medicines Regulatory Framework in 20 Years

On 26 April 2023, the European Commission published two legislative proposals – a new Regulation 2023/0131 and a new Directive 2023/0132 – to replace the current EU regulatory framework for all medicines (including those for rare diseases and for children).

The Directive contains all the requirements for authorisation, monitoring, labelling and regulatory protection, placing on the market and other regulatory procedures for all medicines authorised at the EU and national level. The Regulation sets specific rules (on top of the ones in the Directive) for medicines authorised at the EU level, in particular the most innovative ones.

The proposals aim to reduce costs, expedite the introduction of new medicines and prevent medicine shortages.

Read the key points in the client alert here.




FDA’s Final Q&A Guidance on Risk-Based Monitoring of Clinical Trials Provides Additional Recommendations for Sponsors

The U.S. Food and Drug Administration (FDA) recently finalized its guidance, “A Risk-Based Approach to Monitoring of Clinical Investigations” (the “2023 RBM Guidance”) which follows up on the Agency’s March 2019 draft guidance (the “Draft Guidance”) of the same name and expands on (but does not supersede) the FDA’s August 2013 guidance, “Oversight of Clinical Investigations – A Risk-Based Approach to Monitoring” (the “2013 RBM Guidance”), with new recommendations summarized below to aid sponsors in implementing an effective and efficient risk-based approach to monitoring both risks to participants and to data integrity throughout all stages of clinical investigations of human drug and biological products, medical devices, and combination products.

(1) Approach: Identify, assess and re-assess risks. Create a plan to manage, mitigate, and/or eliminate those risks, including those risks that are newly identified or may not have been anticipated.

  • Risk assessments should inform clinical trial protocol design, investigational plans, and monitoring plans and should be reevaluated and revised throughout the investigation. The monitoring plan should be comprehensive in highlighting identified risks, even those less likely to occur but that could have a significant impact on trial quality or subject safety, and should note how risks will be managed, mitigated, or eliminated.
  • Consider how easily detectable the identified risks are, and the severity and consequences of those risks to human subject welfare and data quality if not detected and addressed.
  • Assess systemic risks, as well as site-specific risks, and consider whether site-specific risks have the potential to become systemic risks.
  • Determine an approach to on-site monitoring visits by taking into account the risks identified and the complexity and intensity of a clinical investigation. Monitoring activities should evolve based on risks identified during trials and should be proportionate to the risks to participants’ rights or safety or to data integrity.
  • Implement a centralized monitoring approach to help minimize missing data and protocol deviations in real-time, such as through the use of electronic data capture systems.
  • The risk assessment should guide how and to what extent source data verification (SDV) will be utilized during on-site monitoring visits.
  • Establish processes to ensure appropriate blinding is maintained. Identify and monitor deviations which could result in unintentional unblinding.
  • Be prepared during an FDA inspection to furnish documentation of the sponsor’s initial risk assessment, if requested.

(2) Content: Components of the monitoring plan should help explain how the sponsor intends to address the risks that could affect the investigation.

  • Include the following components (in addition to those recommended in the 2013 RBM Guidance) in the monitoring plan:
    • Overall investigation design, including blinding and randomization procedures and processes for confirming randomization is performed according to the protocol and investigational plan
    • Sample plan(s), including rationale for, and approach to, identifying the records and data that will be monitored
    • Description of particular issues that would trigger immediate escalation
    • Approach for assessing and addressing a site issue that could escalate into a systemic issue that may warrant protocol or investigation plan changes
  • Reference other clinical investigation management plans in the monitoring plan rather than repeating the information in the current monitoring plan to avoid inconsistencies.

(3) Communicate: Promptly address and communicate monitoring results to the appropriate parties to mitigate and eliminate risk.

  • Perform monitoring in accordance with the pre-established monitoring plan and address issues as the monitor identifies them, including escalation, if needed.
  • Perform a root-cause analysis of issues and promptly implement corrective and preventive actions (CAPAs).
  • Consider amendments or revisions to the protocol or the investigational plan.
  • Communicate and document significant issues to the relevant parties involved at the sponsor and site level, which may also include institutional review boards, data monitoring committees, and/or regulatory agencies, such as the FDA.
  • Provide reports of monitoring activities in a timely manner to the site and discuss the findings with the clinical investigator and site staff. Reports should follow the 2013 RBM Guidance.

While the FDA’s regulations require sponsors to monitor the conduct and progress of their clinical investigations, there are no specifics on how sponsors are to conduct such monitoring. FDA’s guidance provides helpful direction on clinical trial monitoring while recognizing that a monitoring approach should evolve over the course of a trial as risk assessments evolve. Sponsors with upcoming or ongoing clinical trials should consider FDA’s recommendations in monitoring plan development and execution of monitoring activities throughout a trial.

 




The Long (Un)Winding Road Part 2: FDA’s Final Transition Guidances for COVID-19 Devices

On March 24, 2023, the FDA’s Center for Devices and Radiological Health announced the issuance of two much anticipated final guidances that describe the Agency’s transition plans for medical devices that fall within certain COVID-19 enforcement policies or that were issued emergency use authorizations (“EUA”s):

The guidances follow the announcement in early 2023 that the Biden Administration plans to wind-down a number of pandemic-related programs and to allow the COVID-19 public health emergency (“PHE”) declaration, which has been in effect since January 2020, to expire on May 11, 2023.

We summarize some of the key takeaways from FDA’s finalized transition plans.  Read the client alert here.




US Artificial Intelligence Regulations: Watch List for 2023

Companies are developing, deploying, and interacting with artificial intelligence (AI) technologies more than ever. At Goodwin, we are keeping a close eye on any regulations that may affect companies operating in this cutting-edge space.

For companies operating in Europe, the landscape is governed by a number of in force and pending EU legislative acts, most notably the EU AI Act, which is expected to be passed later this year; it was covered in our prior client alert here: EU Technology Regulation: Watch List for 2023 and Beyond. The United Kingdom has recently indicated that it may take a different approach, as discussed in our client alert on the proposed framework for AI regulation in the United Kingdom here: Overview of the UK Government’s AI White Paper.

For companies operating in the United States, the landscape of AI regulation remains less clear. To date, there has been no serious consideration of a US analog to the EU AI Act or any sweeping federal legislation to govern the use of AI, nor is there any substantial state legislation in force (although there are state privacy laws that may extend to AI systems that process certain types of personal data).

Read the client alert here.