Stakeholders want consistency, more examples in last FDA draft guidance on patient-focused drug development

The last draft guidance on using clinical outcomes assessments (COA) as endpoints in patient-focused drug development (PFDD) should use terminology more consistently, align more closely to other PFDD guidance documents in the series, and provide more high-level examples of concepts outlined in the guidance, according to parties who commented on the US Food and Drug Administration (FDA) final draft guidance in their series.

The draft guidance describes a number of considerations for COA-based endpoints in PFDD, such as how to select an endpoint, recommendations for analyzing endpoint results, how to handle missing data, and other study design considerations (RELATED: FDA issues last guidance in patient-focused drug development series, Regulatory Focus 05 April 2023).

As with other documents in the PFDD series, stakeholders generally expressed support for FDA’s efforts to include patient voices in drug development. Some commenters of previous draft guidance documents in the series were unclear about whether the PFDD series was intended to replace the agency’s 2009 document, “Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims,” and said they appreciated that FDA clarified this point for stakeholders in the draft guidance on COA-based endpoints in PFDD (RELATED: Stakeholders want more clarity on concepts introduced in third PFDD draft guidance, Regulatory Focus 03 October 2022).

Consistency across guidances, FDA centers

In their comment, the Alliance for Regenerative Medicine (ARM) said it is important for the PFDD draft guidances to complement the others, including referencing previous guidance documents in the series. “We recommend additional referencing of the many existing FDA guidance documents that are relevant to the concepts within the current one,” they wrote. “ARM particularly suggests referencing the third guidance document in the series on the topics of Item Response Theory and computerized adaptive testing where applicable within this guidance document.”

While the Biotechnology Innovation Organization (BIO) praised FDA in their comment for seeking input on the guidance series from the Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER) and Center for Devices and Radiological Health (CDRH), they shared their experience with inconsistent implementation of recommendations regarding patient experience data (PED) and COA endpoints.

“We suggest FDA increase COA and statistical methods expertise across review divisions and/or utilize the Division of Clinical Outcome Assessment (DCOA) and other biostatistics teams with psychometric expertise in the review of COA data to ensure consistency of feedback with the recent guidance documents and across divisions, they said. “We also recommend FDA have processes in place to share knowledge relevant to the guidance series across Centers as well as more broadly with interested stakeholders.”

The Pharmaceutical Research and Manufacturers of America (PhRMA) suggested in their comment that the PFDD guidance series should also align with other PED guidances, such as the Core Patient-Reported Outcomes in Cancer Clinical Trials draft guidance describing various patient-reported outcome (PROs) measures and currently “does not recommend sponsors seek evidence supporting the fitness of those PROs across cancer populations.”

New terminology

FDA introduced meaningful score differences (MSD) or meaningful score regions (MSR) as new terms in the draft guidance. In their comment, AstraZeneca said the agency should avoid introducing new terms like MSD and MSR, “instead using specific terminology that makes it clear how the threshold should be applied.”

“If the Agency decides to retain the term MSD, AstraZeneca recommends explicitly stating that MSD is being used as a generic term encompassing thresholds applied to within-person change, within-group change and between-group difference,” the company wrote. “To avoid confusion, we recommend that the Agency acknowledge that thresholds for each application could be different and derived using different methods.”

In their comment, the Pharmaceutical Research and Manufacturers of America (PhRMA) told FDA that they “should leverage existing literature to focus on the need for advancement of existing evidence-based approaches, rather than introduce new methodologies and terminologies as the draft guidance does.”

“We believe that any new concepts should have a clear foundation within existing literature,” they wrote, citing the ISPOR Good Practice Guidelines as an example. In addition, any new terms should be consistent across other guidances in the PFDD series and updated in the PFDD glossary.

More examples

BIO said that generally, the draft guidance should have more specific examples and better communicate concepts for all stakeholders, including patients. A summary version of the guidance accessible to patients may help these stakeholders understand the main highlights and visuals of the draft guidance while avoiding more technical language, they noted.
PhRMA and AstraZeneca also requested more high-level examples, use cases, and considerations in the guidance. Areas where this would be useful include the section on creating endpoints for a fixed time point related to a recall period and selecting an assessment for evaluating baseline changes when in a trial with multiple other baseline assessments, they said.

AstraZeneca also asked the agency for more examples of when the use of certain endpoints is appropriate. “To aid in decision-making regarding which type of endpoint to assess, AstraZeneca recommends incorporating further considerations and examples illustrating when change from baseline may be preferable to difference at a fixed time point (in addition to the single arm trial example),” they wrote.

Selecting, assessing COAs

Several stakeholders wanted more information from FDA and how to select and assess COAs as endpoints.

BIO noted that more information on how COAs can be used to assess safety and tolerability, particularly with regard to the agency’s comment that “additional considerations” are needed when COA informs risk.

“We recommend the guidance specifically state whether a COA to inform risk is in scope of this guidance. If in scope, we suggest providing some reference on what these considerations are and evidentiary needs for inclusion in label. and/or regulatory decision-making,” they said.

ARM asked FDA to acknowledge that PROs for some rare diseases may be challenging to validate, and that there is a need for flexibility in these cases.

In their comment, the National Organization for Rare Disorders (NORD) said discussions of missing data that can impact evidence from COA-based endpoints should include additional encouragement to “utilize the practical and methodological experience that patient groups may have in securing timely feedback, encouraging participant compliance, factors underlying drop out, and ways to mitigate challenges and barriers that underlie missing data.”

NORD said FDA should also think of the limitations of rare diseases with COA endpoint selection, “as it is not uncommon for a diagnosis to impact multiple aspects of feeling and functioning.”

“By adding engagement for rare diseases, this can better inform the design of patient-focused trials for more patients, but also improve the collection of data overall,” they said.

Patient considerations

In their comment, the Patient Focused Medicines Development (PFMD) said there is a “need for clear expectations to work with patients as partners from the design stage.”

“We strongly advocate incorporating the patient's voice, priorities, and experiences in developing COAs,” PFMD said. This includes involving the patient throughout, in COA selection, study design, generation, and analysis.

The National Multiple Sclerosis Society also encouraged patient engagement early in the process. In their comment, they said that meeting with a diverse patient group aids in considering various disease manifestations and presentations.

“When a disease has multiple manifestations/presentations, these meetings may assist drug sponsors/FDA with identifying a specific aspect of the disease for evaluating treatment benefits. Only by understanding what is most meaningful to patients will the goal of patient-focused drug development be achieved,” the National Multiple Sclerosis Society wrote. “Additionally, these meetings may uncover the need to construct multiple endpoints if disease manifestations vary greatly from patient to patient.”

NORD said that personalized endpoints for rare disease drug development requires more research and selecting examples of rare disease patient-centered outcome measures “would assist the audiences in better understanding what is acceptable and desirable from a regulatory standpoint.”

NORD also said additional guidance is needed to understand how patient experience data is associated with COA scores. “By providing this transparency to patients, they will be able to understand how their experiences can be turned into data. NORD encourages the Agency to provide additional examples that are disease-specific or disease group-specific, which would assist sponsors and investigators in understanding COAs and endpoints that are viewed as capturing meaningful outcomes for specific conditions or diseases,” they wrote.

Draft guidance

Stakeholders want consistency, more examples in last FDA draft guidance on patient-focused drug development

Related topics

Recommended content

Welcome to the new RAPS Digital Experience