FasterCures Comments on the FDA’s Patient-Focused Drug Development Discussion Documents
February 14, 2018
Divisions of Dockets Management (HFA-305) Submitted electronically
Food and Drug Administration
5630 Fishers Lane, Room 1061
Rockville, Maryland 20852
RE: Federal Register Notice FDA-2017-N-5896
Patient-Focused Drug Development: Guidance 1 – Collecting Comprehensive and Representative Input; Public Workshop; Request for Comments
FasterCures, a Center of the Milken Institute, is a nonprofit, non-partisan think tank driven by a singular goal: to save lives by speeding up and improving the medical research system. We thank you for the opportunity to submit these comments in response to Federal Register Notice FDA-2017-N-5896, Patient-Focused Drug Development: Guidance 1 -- Collecting Comprehensive and Representative Input; Public Workshop; Request for Comments. This letter was prepared and authored by FasterCures staff and informed by discussions with representatives from patient organizations, life sciences companies, and other stakeholders in the biomedical ecosystem.
As part of its mission, FasterCures aims to improve health by driving adoption of methods by which patients’ perspectives shape processes for discovering, developing, and delivering medical products and services. Since its inception in 2003, FasterCures has worked to foster more frequent and effective patient engagement in the biomedical research enterprise.
FasterCures has long supported efforts at the U.S. Food and Drug Administration (FDA) to enhance patient-focused drug development (PFDD). For example, FasterCures participated in a number of the PFDD meetings; contributed to the FDA’s closing panel at the December 18, 2017 public workshop on PFDD; and actively participated in stakeholder meetings for the Prescription Drug User Fee Act (PDUFA) and Medical Device User Fee Amendments (MDUFA), where we advocated for the inclusion of commitments designed to advance the science of patient input.
FasterCures applauds the FDA for its efforts in developing PFDD guidance as outlined in the 21st Century Cures Act and as part of a performance goal in the sixth reauthorization of PDUFA. We see great value in the discussion document as it provides a shared vernacular for PFDD that has been lacking. We also appreciate it as a strong signal that the FDA is moving towards greater acceptance and use of patient experience data to inform regulatory decision-making and an acknowledgement that further technical guidance on how to collect these data is needed. However, we also see opportunities for the FDA to improve the document as well as the ensuing implementation of PFDD regulatory decision-making. Specific comments are as follows:
1. PFDD Guidance as a Shared Understanding in the Ecosystem
FasterCures views the discussion document as a valuable platform for advancing the PFDD conversation. The document affords the community a shared understanding of methods accepted by the FDA for collecting information on the patient experience, with the Standardized Nomenclature and Terminologies as an exceptionally useful section. It helps ensure that all members of the stakeholder community – industry, patient advocacy organizations, patients, and others – are clear about the FDA’s preferred methodological approaches to PFDD. This clarification is a critical step to encouraging investment in the emerging science of patient input and promoting use to benefit patients. It is worth noting that this guidance has a broader intended readership than many guidances that are directed at industry sponsors alone. Necessarily, this guidance may warrant additional considerations as we describe below.
In terms of opportunities for improvement, we recommend that the FDA vet the draft guidance to ensure the concepts are at a correct level for all members of the community. We have some concern about the readability level and how it could inadvertently exclude key members of the PFDD ecosystem, such as patients themselves. This could be remedied through involving patients already working with the FDA (e.g., members of the Patient Engagement Advisory Committee, Patient Representative Program) to comment on the draft guidance from their perspectives.
We also recommend that the FDA be mindful of the balance between prescriptiveness and innovation as it develops the common methods and practices. We recognize the challenge of including detail adequate to ensure a shared understanding while remaining open to the range of scientifically-valid methods as well as methodological innovation. We believe the current level of detail maintains an appropriate balance and structure, with summary information in the main document and details in the appendices. However, at this stage of experience, we are concerned that the more detailed content called for at the Public Workshop could be interpreted as overly prescriptive and chill the evolution of new methods and nontraditional ways of conducting PFFD. It could also make PFDD appear so complex and daunting that uptake of the methods, and potentially the enterprise of eliciting patient perspectives, will suffer.
Additionally, some current content in the discussion document could stymie the use of well-accepted methods by making prescriptive recommendations too soon. For example, the FDA recommendation to use grounded theory could be construed as a recommendation not to conduct PFDD using other well-accepted qualitative methods (e.g., ethnography). This interpretation could exclude methods that are more fit-for-purpose to the goals of PFDD, as well as cause key workforce issues that would slow development of the field, as described below. We encourage that the FDA not become so specific in methodological details that it stifles innovation and potentially excludes valuable, scientifically-valid ways of capturing the patient experience.
Finally, FasterCures appreciates the tremendous contribution of the Standardized Nomenclature and Terminologies to the emerging science of PFDD. We do note that “unmet medical need” is not in the current version. FasterCures recommends that the FDA include a definition of this phrase in the draft guidance, perhaps using the words found in Guidance of Industry: Expedited Programs for Serious Conditions – Drugs and Biologics (May 2014).
With a sound methodological foundation in place, investment in PFDD should grow. This will result in (i) a greater need for academic researchers skilled in the requisite qualitative and quantitative methods; and (ii) more PFDD studies for the FDA to review. Capacity-building on both dimensions will be required to ensure the advancement of PFDD for regulatory decision-making. We discuss recommendations for each in turn.
First, the current supply of researchers with the requisite methodological skills for PFDD, especially qualitative research, is limited. Further, the career path and publication opportunities for researchers with these skills are sometimes uncertain. Building capacity in the research community for conducting PFDD studies is a key component of enabling biopharmaceutical companies, patient advocacy organizations, and others to use the methods outlined in the discussion document.
FasterCures recommends that the FDA explore avenues for building capacity in the research community for PFDD. This could include opportunities through the Centers of Excellence in Regulatory Science and Innovation. It could also include inter-agency partnerships to create grants for institutional training programs for building research capacity for PFDD. These partnerships could model the Agency for Healthcare Research and Quality (AHRQ) Institutional Health Services Research Training Program (T32) grants. The FDA could also explore partnerships to create individual dissertation and early-career awards to support the continued development of career paths. These could, for example, be modeled after the Ruth L. Kirschstein National Research Service Award Individual Predoctoral Fellowships (F31) at the National Institutes of Health (NIH). Potential partners for grant creation and administration could span the governmental and quasi-governmental medical ecosystem, such as AHRQ, NIH, and Patient-Centered Outcomes Research Institute. Further, potential partners could include industry, especially grant programs for mid-career and otherwise established social scientists. As research support is obtained throughout the scholarly lifecycle, academic institutions should begin to engage in internal capacity-building to support the PFDD ecosystem – to include training as well as promotion and tenure systems. As more researchers are developed and supported, publication opportunities should begin to flourish alongside, a key condition for the success of an emerging science.
Second, the FDA requires the capacity to evaluate the plethora of PFDD studies that could result from this and other planned guidance documents. Capacity includes the expertise to evaluate the quality of studies using quantitative and qualitative methods that may be new to the FDA. For example, many submissions going forward may require FDA judgments about the quality of a grounded theory analysis. Capacity also includes hiring the appropriate number of social scientists with this expertise. Failure to provision for the requisite expertise, at the appropriate levels, could result in delayed reviews – and therefore slower drug development. As with much of the FDA’s work, a small pool of qualified researchers combined with high demand from industry may affect the FDA’s ability to recruit. FasterCures has a longstanding interest in ensuring appropriate FDA workforce levels and stands ready to partner with the FDA and other stakeholders to develop possible solutions to this and other workforce challenges.
FasterCures is also concerned that the FDA’s prescriptive recommendations about research methods in the discussion document could limit the current talent pool even further, making the workforce challenges even more pronounced. Academic disciplines using qualitative methods tend to gravitate to certain analytic methods. By recommending grounded theory to the exclusion of other well-accepted qualitative methods, the FDA could be restricting PFDD expertise to only the disciplines that favor grounded theory. This could result in further contractions of the workforce qualified to conduct and evaluate PFDD studies. With a scarce supply of researchers with the requisite expertise, FasterCures recommends that the draft guidance encourage other established qualitative methods as well.
3. Collecting and Analyzing Data
FasterCures appreciates the discussions on sampling and representativeness, especially as applied to quantitative, hypothesis-testing research. However, it appears to convey an interest in deductive approaches designed to explain the average patient experience to the exclusion of inductive, exploratory approaches designed to generate new constructs about the patient experience and/or to describe atypical patient experience. This is likely to hamper the development of valuable qualitative studies using methods enumerated elsewhere in the document. It also creates a situation in which the document unintentionally contradicts itself. In contrast to our recommendations above to not add information on representativeness, this is an area in which relatively simple additions and changes could increase consistency within the document, fidelity to qualitative methods, and innovation.
For example, the document notes that “the purpose of this document is to present methods for collecting information on the patient experience that is representative of the intended population” (lines 128-129); “an important goal is obtaining patient experience data that are … representative of the target population” (lines 416-417); and “regardless of how the study sample is constructed, it is important to try and ensure that patients in the study sample represent the target population” (lines 515 – 517). It could be construed from these and other similar statements that the FDA does not endorse studies for which representativeness is not a primary objective, such as many qualitative studies. It is appropriate for qualitative studies to be non-generalizable when the studies are hypothesis-generating as opposed to hypothesis-testing. For example, grounded theory (endorsed by the FDA in Appendix 6, lines 230-231) uses theoretical sampling on concepts as opposed to statistical sampling on a representative population (Glaser and Strauss 1967). Thus, a focus on representativeness is in many ways incongruent with the FDA’s endorsement of grounded theory. It is also incongruent with other descriptions of qualitative research methods outlined but not endorsed in the discussion document (e.g., lines 531-533; Appendix 6, Section 2). Of note, the document references “studies in which generalization to the target population is not the primary objective” and notes they often use non-probability sampling schemes (lines 433-434); however, the goal of the guidance remain representativeness.
FasterCures recommends that the FDA (a) qualify current representativeness goals as applying to hypothesis-testing research studies only; (b) enumerate separate sampling goals (e.g., theoretical sampling and theoretical saturation) for hypothesis-generating research studies (including but not limited to those using grounded theory) and highlight regulatory acceptance of non-generalizability as an integral part of the study purpose and method for these types of studies; and (c) convene an external panel of qualitative research experts to review the document. Implementing these recommendations will ensure internal consistency within the document and adherence to best qualitative research practice as the FDA forges into new and tremendously useful research methods. The panel could also advise on the workforce issues described previously and help the FDA develop recommendations for research methods.
FasterCures also appreciates the FDA’s acknowledgement of the need to sample from persons across relevant socioeconomic and cultural backgrounds and levels of literacy (Section 2.6). This practice is crucial to achieving sufficient representation in hypothesis-testing research and to attaining inductive insights in hypothesis-generating research. Once included in research, issues of trust may deserve special attention for eliciting a response useful to research. For example, it may be difficult to establish rapport with some populations historically excluded from or mistreated during research studies, and this could lead to interview or survey data that does not adequately represent the patient experience. Approaches such as community-based participatory research could help greatly in obtaining and maintaining trust in research. FasterCures recommends that the FDA include an acknowledgement of these issues and approaches to handling them.
Additionally, if researchers solicit patient perspectives, patients should see the results of the research. Too often, this does not occur. FasterCures recommends that the FDA acknowledge that communicating the results of PFDD studies to patients is good research practice.
Finally, the discussion document notes that methods for collecting and analyzing patient preference information is addressed in two separate guidances: (1) Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims, and (2) Patient Preference Information – Voluntary Submission, Review in Premarket Approval Applications, Humanitarian Device Exemption Applications, and De Novo Requests, and Inclusion in Decision Summaries and Device Labeling. The second guidance did not include the FDA Center for Drug Evaluation and Research (CDER). FasterCures recommends that the discussion document clarify if the second guidance applies to patient preference information for medical products reviewed by CDER.
4. Specific Illustrations for Use in PFDD Meetings
There is high interest among patient organizations to conduct PFDD meetings under FDA’s guidelines for externally-led meetings. It would be valuable to include within the guidance a set of information that pertains specifically to the collection of patient perspective/experience data for this purpose. Past examples could be used to illustrate acceptable practices so that hosting organizations understand the FDA’s expectations.
5. Clarity on Repository of Patient Experience Data
At the December 18, 2017 public meeting, the FDA noted that it plans to develop a repository of patient experience data to house submissions to the Agency. FasterCures supports the development of a repository and applauds the FDA for taking steps to reduce duplication by increasing transparency of submissions. There could be some concern in the stakeholder community about the repository related to minimum quality thresholds, as well as requirements to participate in the effort. For this reason, FasterCures suggests that the FDA issue additional information on whether the repository is voluntary or mandated, and on any quality standards the FDA will use to determine inclusion of studies in the repository.
6. Use of Guidance in Regulatory Decision-Making
Understanding the use of patient experience data in regulatory decision-making is a crucial, anticipated element of future guidances. We look forward to future FDA documents outlining when companies can engage the FDA to obtain input on patient experience studies, as well as an exemplary regulatory pathway and/or roadmap for the FDA’s use of patient experience data. Guidance on these topics will encourage investment in the emerging science of patient input.
Additionally, FasterCures notes that patient experience data are crucial to informing benefit-risk assessments, but they are not generally made available in Advisory Committee briefing materials. In future guidances, FasterCures recommends that the FDA stipulate how patient experience and patient-centered benefit and risk assessments will be communicated to Advisory Committees evaluating evidence for medical product applications.
Once the FDA issues draft guidance, a comprehensive communications plan is the key to ensuring maximum impact. Because PFDD is in its early stages, exploring how to reach a full complement of audiences is paramount -- from traditional drug developers to patient advocacy organizations to academic researchers. FasterCures recommends that the Agency incorporate draft guidance into traditional FDA communications platforms as well as conduct outreach through non-traditional means, such as emails to participants in the series of 24 PFDD meetings that may not receive regular FDA updates. We also recommend that the FDA include targeted communications to different functions within organizations to ensure broad reach (e.g., government affairs as well as research and development). Finally, public-private partnerships could serve as a valuable conduit for communications, and we encourage the FDA to explore public-private opportunities to extend the Agency’s reach using limited public resources.
FasterCures is committed to partnering with the FDA and other stakeholders to ensure that the patient perspective is reflected in medical product development and regulatory decision-making. We appreciate the FDA’s continued efforts to solidify and extend the accepted methodology for PFDD. In the process, we hope the FDA will take our considerations into account and refine a well thought-out document while planning for future workforce and communications needs.
We welcome the opportunity to explore these comments in greater detail with FDA staff.
Colleen Rye, PhD
Director, Research and Regulatory Policy
FasterCures, a Center of the Milken Institute