How FDA is already requiring companies to study the safety and effectiveness of their approved drugs in diverse populations

Context: FDA’s recent policy development to increase diversity in clinical trials

• The FDA has long been interested in the diversity of patients enrolled in clinical studies used to support approval. The agency first issued guidance on demographic subgroup analyses in the 1980s. In the intervening years, FDA continued to update guidance in line with updates from the Office of Management and Budget (OMB) and Congressional directives. [ See AgencyIQ’s analysis for more detailed history of FDA’s policy and guidance on diversity in research studies].
• But in recent years, FDA’s efforts to promote and enforce clinical trial diversity have intensified. Most notably, in April 2022 the FDA issued a draft guidance document on diversity in clinical research programs. At a high level, the guidance laid out a policy whereby sponsors of certain products would voluntarily submit a Race and Ethnicity Diversity Plan as part of their development program that would outline and justify their approach to recruiting and retaining a “representative” research population. [ Read AgencyIQ’s extensive analysis of the draft guidance here.]
• One month before, the FDA also put industry on notice when it issued a Complete Response Letter (CRL) to Eli Lilly’s sintilimab after the regulator determined that the sponsor’s data – collected from trial sites in China – was not reflective of the U.S. population the drug was intended to treat. [ Read AgencyIQ’s analysis of the CRL here.]
• In late 2022, reforms within the 2023 Consolidated Appropriations Act gave FDA new statutory authority to require what the law now referred to as Diversity Action Plans (DAPs) for certain premarket clinical studies. Section 3601 of the law amended the Federal Food, Drug and Cosmetic (FD&C) Act to expressly allow FDA to require submission of a DAP for certain Phase 3 and pivotal studies of medical products, including drugs. According to the law, the DAP is a plan, developed by the product sponsor, that would need to include specific enrollment goals, the rationale by which these goals were developed, and “an explanation of how the sponsor intends to meet such goals.”
• The law also directed the FDA to “update or issue guidance” on DAPs by the end of December 2023 – a deadline that the FDA missed. However, the agency issued two new draft guidance documents related to diversity in the meantime in the interim period.
• First, an August 2023 new draft guidance on post-market approaches for obtaining data in underrepresented groups. In other words, if a company did not collect enough data on the use of their products in a specific demographic prior to approval, the FDA might require that it be collected later. The document clarifies that “there are various mechanisms for obtaining postmarketing data on underrepresented populations,” in particular, postmarketing requirements (PMRs), postmarketing commitments (PMCs), or a confirmatory trial for products granted accelerated approval. As AgencyIQ has previously discussed, PMRs are mandatory trials or studies levied by the FDA, while PMCs are post-market studies or trials that are agreed to by the sponsor but are not legally (statutorily or via regulation) required. The guidance explains that in this context, a PMR could be applied if there’s a safety signal, while a PMC could address underrepresentation more broadly. This aligns with the typical use of PMRs and PMCs, in which PMRs focus on known or potential risks, while PMCs can fill in data gaps or bolster safety/efficacy information. [ Read AgencyIQ’s analysis of the 2023 draft guidance here.]
• Second, in January 2024 the FDA updated a 2016 guidance on general collection of race and ethnicity data. The new draft offers recommendations to be used for collecting and reporting this information, which continues to be based on OMB Directive 15. Directive 15 defines the core set of minimum categories of race and ethnicity and offers recommendations on data capture. [ Read AgencyIQ’s full analysis of that draft guidance document here.] A few months later in late March 2024, OMB finalized some long-awaited updates to Directive 15. Following an “ initial set of recommended revisions” from the White House Office of the Chief Statistician in January 2023, Directive 15 has now been formally updated at the federal level. At a high level, this entailed three main changes: (1) race and ethnicity are now combined into one question; (2) Middle Eastern and North African (MENA) has been adopted as a new minimum category; and (3) the Directive now “requires the collection of more detail beyond the minimum required race and ethnicity reporting categories” in most situations. [ Read AgencyIQ’s full analysis of the changes here.]
• The FDA ultimately released its Diversity Action Plan guidance in June 2024. The document offers granular recommendations on how sponsors of both drug and device programs should comply with the new legal requirements, including how to design a DAP and set enrollment goals using data about the incidence and/or prevalence of a disease in the U.S. population. These goals and any adjustments should be explained through a rationale, and a plan must be developed with specific enrollment and retention strategies. Finally, sponsors must articulate a plan to measure progress toward meeting the goals. [Read AgencyIQ’s extensive initial analysis and follow-up here.] The comment period for the draft guidance document is currently open until September 25, 2024. Assuming that the comment period is not extended, and FDA meets its statutory timeline to finalize the guidance within nine months of the closing of the comment period, the guidance could theoretically be put into force by late 2025.

All of this raises an important question: Will FDA leverage its postmarket toolkit for issues related to diversity?

• But the DAP guidance document did not answer to a key question: What happens if targets are not met? As AgencyIQ has previously discussed, the intersection of DAP policy with FDA’s post-market tools is unclear. The 2022 version of the DAP guidance included the following footnote, which was the subject of much discussion in comments on the document: “In the event that recruitment goals are not met despite best efforts, sponsors should discuss with FDA a plan to collect this data in the post-marketing setting” (see footnote 24). However, the new 2024 version of the document does not include any information on the subject, and does not address the question of what happens if studies under DAPs are unable to reach their goals in the pre-market setting.
• However, there is recent precedence for the FDA using PMRs and PMCs to obtain data related to clinical trial diversity, especially for oncology products. For example, several oncology drugs granted accelerated approval in 2023 were required to conduct confirmatory studies and to enroll a study population that sufficiently reflects the racial and ethnic diversity of the U.S. patient population with the disease (See Jaypirca (pirtobrutinib) approval letter and Lunsumio (mosunetuzumab-axgb) approval letter).
• That said, these examples are anecdotal, and AgencyIQ wanted to know the extent to which FDA relied on this sort of approach more systematically. This led AgencyIQ to carry out a comprehensive assessment of the extent to which FDA has asked companies to address diversity-related concerns in a postmarket setting and what FDA wants from those same studies.

Before we talk about the results of our analysis on post-marketing diversity requirements, let’s first discuss our methods.

• The following analysis is based on FDA’s Downloadable Database of Postmarketing Requirements and Commitments, which was last updated July 31, 2024. Altogether, the database comprises over 2,500 unique PMRs and PMCs. According to FDA, the dataset contains all open PMRs and 506B PMCs (those with annual reporting requirements), as well as those have been closed within the last year due to fulfillment or release. This means that the following analysis excludes requirements or commitments closed prior to late July 2023. Of note, AgencyIQ also excluded entries for products with approvals that have been withdrawn.
• To curate this pool to PMRs and PMCs related to diversity, AgencyIQ carried out a two-step filtration process. First, the dataset was limited to entries for which the text of the requirement or commitment contained keywords related to racial and ethnic diversity or representativeness of the population to be studied. Specifically, these entries contained the words diverse, diversity, representative, underrepresented reflective, minority or demographic. In addition, a search was performed for the race and ethnicity categories within OMB Directive 15. The scope of this analysis did not include additional parameters of diversity often recognized by the agency as pertaining to diversity such as sex, age, and socioeconomic status.
• Second, these entries were narrowed down to those with an intentional element related to diversity. In other words, we only wanted to find cases where FDA asked the sponsors to go further than simply collecting demographic information.
• Specifically, PMR and PMC entries were included in the analysis set if the task had one of two of the following characteristics. First, FDA specifically asks the sponsor to ensure a representative population. For example, PMC 4248-1 for AstraZeneca’s Lynparza (Olaparib) asks the sponsor to do the following: “Conduct analyses of clinical trial data to characterize the safety and pharmacokinetics of olaparib across a diverse population. To support a comparative assessment across all U.S. race and ethnic populations, ensure that racial and ethnic minorities are sufficiently represented in the analysis. Include a tabular summary of the available pharmacokinetic data by each racial and ethnic group. Provide the specific racial ethnic group as well as the geographic location of each patient.” Second, FDA asks the sponsor to enrich data for underrepresented groups. For example, PMC 4074-3 for Janssen’s Darzalex Faspro (daratumumab and hyaluronidase-fihj) has the following ask: “Submit a final report containing data from clinical trials, post-marketing reports, compassionate use/expanded access programs, real-world evidence, and other sources to further characterize the safety and efficacy of daratumumab (SC) in combination with pomalidomide and dexamethasone among U.S. racial and ethnic minority patients with multiple myeloma.”
• Using this approach, AgencyIQ identified 55 total relevant PMRs and PMCs. The data signals that these PMRs and PMCs have become more common in recent years. The oldest active requirement was issued in 2018 related to the approval of Otsuka Pharmaceutical’s Jynarque. The newest requirement in this group comes from the March 2024 approval of a supplemental application for a new indication for Beigene’s Brukinska.
• The dataset also had a similar distribution of requirements (24) and commitments (31). In addition, there was no specific application type that was observed to be overwhelmingly more likely to receive this type of PMC or PMR. That said, there were more New Drug Applications (NDAs, 29) than Biologic License Applications (26), and more original applications (35) than supplementary applications (20).

What does FDA ask for in its diversity-focused PMRs and PMCs? There are four distinct types of requirements.

• Through examining the text of each relevant PMR and PMC, AgencyIQ identified four distinct types of asks for sponsors (see the Appendix at the end of this piece for the full text of each requirement). In many cases, specific phrasing and descriptions were retained across different products, indicating that the agency leans on a few “template” PMRs and PMCs when seeking post-market study of diversity. In other words, sponsors can expect to see similar language in these types of requirements and commitments.
• First, FDA will frequently asked sponsors to conduct a new clinical trial. In many of these cases, the agency asks for a phase 3, randomized trial to verify and describe clinical benefit or better understand a safety signal. Most of these entries contain some variation of the sentence, “The trial should include sufficient numbers of racial and ethnic minority patients to better reflect the U.S. patient population and allow for interpretation of the results in these patient populations.” However, a few new clinical trials are asked to enrich certain populations. For example, PMR 4106-11 for Kadmon Pharmaceutical’s Rezurock (belumosudil) instructs the firm to, “Conduct a clinical trial in a sufficient number of Black patients with chronic graft versus host disease to assess the risk of cardiac toxicities and further characterize Grade 3 toxicities including gastrointestinal and vascular disorders associated with the use of belumosudil.”
• Second, FDA might ask sponsors to conduct additional clinical data analysis, which won’t necessarily mandate a new trial. For example, Helsinn Healthcare received an Accelerated Approval PMR for Truseltiq (Infigratinib) that instructed the firm to: “Submit the final progression-free survival (as assessed by blinded independent review) analysis and interim overall survival analysis at the time of final progression-free survival analysis, including datasets from a randomized clinical trial comparing infigratinib to chemotherapy to verify and describe the clinical benefit of infigratinib in patients with advanced or metastatic cholangiocarcinoma harboring an FGFR2 gene fusion or other rearrangement. Ensure that racial and ethnic minority subjects are adequately represented in the trial population, at a minimum, proportional to the prevalence of FGFR2 alterations in these subgroups in the US population.”
• Third, FDA often asked sponsors to conduct what it referred to as an “integrated analysis.” In general, these are PMCs that task the sponsor with using “data from clinical trials and other data sources such as post-marketing reports, real-world evidence and other sources” to answer a question in an underrepresented population with the disease or condition. These commitments also typically specify that the integrated analysis should “support comparative efficacy and safety analyses” between the underrepresented population and majority group.
• Finally, the agency asked for a prospective observational study in rare cases. For example, AbbVie’s Oriahnn (elagolix, estradiol and norethindrone acetate) received PMR 3837-3, which sought “a prospective observational study in premenopausal women receiving treatment with Oriahnn to assess the incidence rate, time to onset, pattern, extent, and reversibility of alopecia, as well as any racial/ethnic differences in developing alopecia.”

• According to AgencyIQ’s analysis, FDA allotted different timelines for sponsors to complete various types of PMRs and PMCs. When comparing the timeframes between assignment of a given requirement or commitment (not the original application approval) and the agency’s original final report due date, it becomes apparent that clinical data analysis tasks generally have a shorter time frame than the more intensive integrated analyses or new clinical trials. While only two such entries were in the analysis set, prospective observational trials were given additional time to complete.

The status: most diversity-focused PMRs and PMCs are currently pending

• FDA’s database contains information pertaining to the status of the open PMRs and PMCs at the time of the update (for our analysis, late July 2024). The status is reported in the following categories, as defined on FDA’s webpage.

• Overall, 60% of the diversity-focused PMRs and PMCs are currently pending, meaning they have yet to be initiated. Just one entry in the dataset has been submitted to the agency, with the rest proceeding either on or off the original study schedule.

The assigners: Just one FDA review office has issued the overwhelming majority of diversity-focused PMRs and PMCs

• The majority of diversity-related PMRs/PMCs were issued by FDA’s Office of Oncologic Diseases (OOD). Other offices that issued these types of PMCs or PMRs were the Office of Neuroscience, Office of Therapeutic Products, Office of Office of Rare Diseases, Pediatrics, Urologic and Reproductive Medicine, and the now-reorganized Office of Drug Evaluation.

• OOD works in partnership with the agency’s Oncology Center of Excellence (OCE). This is important because OCE has functioned as a catalyst for many of FDA’s diversity-oriented policy activities. The Center wrote FDA’s original 2022 draft guidance on voluntary diversity plans. As AgencyIQ has previously discussed, it appears OCE also took the lead with drafting the new DAP guidance document; LOLA FASHOYIN-AJE (previously with OCE, and now at CBER’s Office of Therapeutic Products) is listed as the top point of contact on the guidance, along with TAMY KIM, who is OCE’s director for regulatory affairs and policy.
• Within OOD, most assignments came from the Division of Hematologic Malignancies 2 (DHM2). According to our analysis, requirements and commitments issued by DHM2 make up nearly half (47.37%) of the entire dataset. FDA’s webpage states that DHM2 is charged with reviewing regulatory submissions for products proposed to treat for lymphoma, chronic lymphocytic leukemia, multiple myeloma, and other plasma cell malignancies. The division is led by Director NICOLE GORMLEY. Gormley has had a leadership role in the agency’s diversity initiatives. For example, in May 2024, OCE announced that Gormley would be a co-director of Project Equity’s expansion from “project” to “program” [ Read AgencyIQ’s full analysis here.].

Within OOD, most assignments came from the Division of Hematologic Malignancies 2

Analysis

• These data indicate how the FDA is already making use of postmarketing commitments, and what drug companies and public stakeholders can expect. This data represents a starting point; as FDA’s new policies for clinical study diversity are implemented in coming years, the volume and makeup of these assignments is likely to change, especially in the wake of the agency’s 2023 draft guidance on obtaining data on underrepresented groups in postmarket settings. They are also likely to change as they are used more outside of the field of oncology.
• It’s worth noting that for “integrated analysis” types of PMCs, the FDA has said that real-world evidence (RWE) is an acceptable source of supportive data. This is notable it may make it easier for some companies to collect additional data without necessarily running clinical studies.
• While FDA’s authority to assign postmarketing studies is solid, industry’s performance with fulfilling these assignments has been historically lackluster. As required under the 1997 Food and Drug Administration Modernization Act (FDAMA), the agency creates a public report that provides a snapshot of sponsor performance on PMRs and PMCs each federal fiscal year (FY). The most recent report covered FY 2022 and was released in late April 2024. In that year, 30% of all required postmarketing studies had not yet been completed and were off schedule for future completion. In addition, just half of all PMRs established in FY 2016 had been fulfilled. Recent legislative reforms to FDA oversight of PMR completion status for accelerated approvals may make an impact on future reports, though it will likely take some time to see an effect reflected in the data.

Featuring previous research by Laura DiAngelo.
To contact the author of this item, please email Amanda Conti ( aconti@agencyiq.com).
To contact the editor of this item, please email Alexander Gaffney ( agaffney@agencyiq.com)

Key Documents and Dates

• Postmarketing Requirements and Commitments: Downloadable Database File

Appendix: Postmarketing requirements and commitments referencing trial diversity factors