New technologies continue to emerge that improve care for patients taking prescription medications. On September 18, 2023, five years after the Food and Drug Administration sought initial public comment, the agency issued its draft guidance, “Regulatory Considerations for Prescription Drug Use-Related Software,” describing how FDA intends to apply its drug labeling authorities to certain software outputs that are utilized with prescription drugs, or drug-led combination products, by or on behalf of a drug sponsor.¹

The guidance defines prescription drug use-related software as “software that (1) is disseminated by or behalf of a drug sponsor; and (2) produces an end-user output that supplements, explains, or is otherwise textually related to one or more of the sponsor’s drug products.”² The guidance does not apply to software developers who are unaffiliated with the drug sponsor.

This Bulletin provides a high-level summary of key points covered in the guidance. The agency will accept comments on this guidance until December 18, 2023.

Highlights

• The guidance differentiates between “FDA-required labeling” and “FDA promotional labeling.” The former includes “labeling reviewed and approved by FDA as part of a New Drug Application, Abbreviated New Drug Application, or Biologics License Application, as well as supplemental applications.” The latter is labeling that is not required by FDA (e.g., printed, audio, or visual information that describes the drug).
• FDA will use many factors to evaluate whether the end-user output is promotional labeling or FDA-required labeling and whether the software function should be described in the Prescribing Information (“PI”), including:
  • whether the prescription drug use-related software provides a function that is essential to the safe and effective use of the product;
  • whether evidence is provided to support a clinical benefit when the prescription drug use-related software is used; and
  • whether the prescription drug use-related software relies on data directly transferred from the device constituent part of a combination product.
• The agency expects that many drug use-related software functions will provide information directly to a patient about the use of a prescription drug, such as helping a patient keep track of their own prescription drug use.
• The agency notes that end-user output is a type of prescription drug labeling.
• The sponsor should include a description of the prescription drug use-related software in the PI section if the software is essential for safe and effective use of the drug product.
  • Certain software functions may be included under applicable sections of the PI. For example, if the sponsor has evidence from adequate and well-controlled studies that use of the prescription drug use-related software results in a meaningful improvement in a clinical outcome as compared to drug products without such software, this may be applicable under the “CLINICAL STUDIES” section of the PI.
    • Where the sponsor can support the clinically meaningful benefit by sufficient evidence, the end-user output associated with that software function would generally constitute FDA-required labeling.
• If a software relies on data directly transferred from the device constituent part of a combination product, FDA recommends such inclusion in the PI.
  • Functions that are not directly transferring information from a device constituent part to software should not be described in the PI, unless there is an additional factor that necessitates including this information (e.g., the function is necessary for safe and effective use of the drug).
  • Promotional labeling should not be included in the PI.

Examples

• FDA provides two appendices that give examples and context of different considerations it uses when contemplating labeling of prescription drug use-related software.
• Appendix A provides examples of prescription drug use-related software functions and end-user output.
  • Example: A mobile application collects data from an oral tablet with an embedded sensor device system ingested by the patient, performs analysis of ingestion data, displays the data on a screen via the application, and wirelessly transfers data to a healthcare practitioner’s cloud-based web-application.
    • The software functions include:
      • The transfer of data from the embedded sensor to the mobile application
      • The analysis of ingestion data by the mobile application
      • The transfer of data from the mobile application to a cloud-based web-application
      • The display of ingestion data on the web-application
      • The display of ingestion data on the mobile application
    • All of the software functions are categorized as device-connected software functions because they depend on data that are directly transferred from the device constituent part of the product.
    • The two end-user outputs are: (1) the mobile application’s display of ingestion data; and (2) the display of ingestion data on the web-application.
  • Example: A mobile application prompts patients to document in a standardized way the manner and severity of symptoms of their condition and then transfers the self-reported information to a cloud-based application for their healthcare practitioner to review on a web-application.
    • The software functions include:
      • The capture of self-reported patient entry of information into the mobile application
      • The display of symptom incidence and the severity data through the mobile application display screen
      • The transfer of self-reported data to a cloud-based application
      • The display of data on the healthcare practitioner web-application
    • None of the software functions are considered device-connected software functions because they do not rely on data directly transferred from the device constituent part of a sponsor’s combination product.
    • The two end-user outputs are: (1) the mobile application’s display of symptom incidence and severity; and (2) the web-application’s display of self-reported information.
  • Appendix B provides examples of device software functions that are regulated by FDA and where the end-user output is considered promotional labeling.
    • Example: A mobile application is designed for use with an insulin product.
      • The software has a dose calculator function to help patients with mealtime insulin dose calculations. The device software function is subject to relevant device requirements, such as the submission of a new premarket notification.
      • The software output also mentions the insulin product; this output is promotional labeling for the prescription drug.

AGG Observations

• The guidance, as the name reflects, focuses on drug use-related software functions and end-user output. However, FDA makes clear that, in some cases, software functions may be “devices” subject to medical device regulation.
• FDA recognizes that digital health technologies continue to develop and, when used with prescription drugs, can improve patient care. However, with growth comes potential uncertainties and risks and, with many such products made available directly to consumers, the agency wants to ensure information is accurate and protects end-users.
• The guidance is new and the issue of prescription drug use-related software is evolving. Therefore, if a company wants to include this type of information in the PI, it should discuss with the appropriate FDA review division sooner rather than later.

AGG Food & Drug co-chair Alan Minsk would like to recognize AGG law clerk Priya Sinha as a co-author of this Bulletin.

[1] https://www.fda.gov/media/172165/download.

[2] According to the draft guidance, an end-user output is “any material (content) that the prescription drug use-related software presents to the end user (a patient, caregiver, or healthcare provider).”