Review divisions within the Food & Drug Administration are signaling a commitment to the development and use of patient-reported outcomes during the clinical development of new drugs and biologics – both in public statements from agency officials and in regulatory policy changes that would help develop new PRO tools and encourage drug sponsors to meet directly with FDA review teams to discuss new endpoints.

While that commitment has been slowly building within the agency over the past few years, a number of review managers from the Center for Drug Evaluation & Research used an April 1 FDA public workshop on clinical outcomes assessments to detail specific ways their divisions have become more active in the development of patient-reported endpoints.

The FDA workshop was not intended to be specific to patient-reported outcomes; the explicit objective of the meeting was to publicize the results of a one-year reassessment of the agency’s approach to reviewing and adopting patient-reported outcome measures and other clinical outcome assessment tools, and introduce a new website to help identify assessment tools in development.

But a clear message from the meeting was a growing receptiveness among senior drug review managers to better engage with sponsors in the creation of patient-reported endpoints – including some divisions that historically have not been out in front on PROs. For sponsors, that means the traditional route to getting patient-reported data accepted – working directly with the FDA review division – is once again becoming more attractive.

FDA’s Office of New Drugs has a central office that oversees the development and validation of patient-focused endpoints: the Study Endpoints and Labeling Development (SEALD) team. But some industry stakeholders have observed that SEALD has a tendency to get mired in process and become an obstacle to patient-reported outcomes, rather than a conduit to developing them. ([A#2015500002])

Although SEALD moderated the April 1 discussion, the central players at the workshop were senior managers from four drug review divisions: Paul Kluetz, MD, Acting Deputy Director of the Office of Hematology & Oncology Products; Ellis Unger, MD, Director of the Office of Drug Evaluation I; Andrew Mulberg, MD, Deputy Director of the Division of Gastroenterology & Inborn Error Products; and Janet Maynard, MD, a team leader from Pulmonary, Allergy & Rheumatology Division.

All four review managers expressed their buy-in to the effort to get more patient-described symptoms into the review efforts in their respective fields. Their comments demonstrate that they are accepting – at least in their public statements – the new ethos of listening directly to patients and working to translate those concerns into endpoints and/or issues to be addressed in clinical development and labeling.

The most enthusiastic support came from a CDER office not typically focused on patient-reported outcomes because of the nature of the diseases it regulates: the Office of Hematology & Oncology Products. The acting deputy director of that office, Paul Kluetz, has emerged as a champion of patient-reported outcomes, despite the challenges with developing PROs in oncology. Also see "Patient-Reported Outcomes in Oncology: The Jakafi and Zytiga Exceptions" - Pink Sheet, 24 November, 2014.

While Kluetz was the most effusive about efforts in the Office of Hematology & Oncology Products to build in more patient-reported data into reviews and labeling, the other FDA review managers at the workshop had similar endorsements. That indicates broad – although not uniform – acceptance among review division leadership of the call for patient perspectives in the regulatory process.

For sponsors interested in developing new patient-reported outcomes, that’s good news. There has been real movement over the past few years within the biopharma industry to include patient-reported outcomes earlier in clinical development. Engaging directly with review divisions is a more effective way to get that done.

Now the challenge is putting the right tools in place to inform those discussions.

PRO Interest Across Review Divisions

The commitment to developing PROs has not been uniform across the drug center. Certain CDER review divisions (like Neurology; Pulmonary & Allergy; and Anesthesia, Analgesia & Rheumatology) approve most products with PROs in labeling. Others (like Oncology, Cardiovascular, Antiviral and Dermatology) have lagged behind, either because those divisions deal in diseases that are not conducive to PROs, or perhaps, due to the proclivity of division management.

That tide seems to be shifting. Review managers at the workshop were eager to share ways their divisions have worked on patient-reported outcomes. Some examples were more significant than others, but in contrast to the SEALD staff, the reviewers were focused on ways to use new endpoints in drug development and regulation – with an emphasis on the endpoint, rather than the process of getting it accepted.

For example, ODE I Director Unger, who oversees the Cardio-Renal, Psychiatry and Neurology divisions, noted that he is “very optimistic” and enthusiastic about getting the patient voice into ODE I decisions.

Unger noted that the SEALD compendium of existing clinical outcome assessment tools has overlooked some used by his staff (specifically the psychiatric drugs division) in efficacy supplement reviews, suggesting that he feels current efforts by his review divisions may be underappreciated. He also suggested that some patient-reported tools collect too much information and as a result may drown out the key elements of the assessment with excess noise.

GI division deputy director Mulberg participated in a discussion on international coordination on clinical outcome assessment tools. He stressed the ongoing effort in Crohn’s disease and celiac disease endpoint development, noting that the agency held full-day meetings on March 30 and March 31. The “ongoing efforts” in Crohn’s is to a point where it “will most likely be published.”

Maynard, the team leader in Pulmonary, Allergy & Rheumatology, reported on the work in her division to restart development in chronic fatigue syndrome and myalgic encephalomyelitis. Those conditions were the subject of the first patient feedback meetings held by FDA under the Prescription Drug User Fee Act reauthorization signed into law in July 2012. Also see "Confluence of Symptoms and Treatments at FDA Patient Meetings" - Pink Sheet, 27 September, 2013.

She highlighted the role of the April 2013 patient-focused meeting on the disease and the publications from that meeting (a “Voice of the Patient” report, which led to a draft guidance on drug development). “We have also formed a working group discussing PROs for CFS/ME,” Maynard added. “We hope that the guidance clearly articulates FDA’s expectations for drug approval for this indication but also emphasizes the importance of assessing patient symptoms.”

At one point, Maynard was asked what FDA would need to adapt fatigue endpoints from other conditions (like multiple sclerosis and arthritis) to chronic fatigue syndrome and myalgic encephalomyelitis. She returned to the importance of patient responses: “We really need to get that patient input piece to see how much applicability an existing measure might have applied across conditions.”

Oncology’s Kluetz was arguably the most effective of the review managers on the topic at the FDA workshop. He participated in the meeting all day, and began his remarks by declaring that he is “very passionate” about the topic of patient discussion of disease experience, and by sharing a personal experience from his medical practice.

During his once-a-month hospital oncology duty over the last weekend of March, he encountered one young woman diagnosed with carcinomatosis. The medical interns who conducted the first morning meeting with the patient (what Kluetz described essentially as a “clinician-related outcome”) reported her as doing fine overnight. “I go in there and talk to her,” Kluetz said, and she has “pain that is out of control. We need to titrate her meds. She has nausea and vomited five times.”

That real-life example illustrates the disconnect that can occur between what the physician sees and what the patient is experiencing, Kluetz suggested. The physician who comes in and sits down and looks in a patient’s eyes, Kluetz said, “is going to get a totally different read than just running through and asking ‘Are you okay?’” The “translation of the clinician to the patient” is a factor, “I saw it this weekend.”

Oncology Office Adds PRO Leads In Each Review Division

Kluetz said that academics, companies and regulators are starting to align on the issue. “We are at this critical point where I think we can really get something done in the next couple of years.”

One needed step, however, is revising FDA’s final guidance on Industry Patient-Reported Outcome Measures from 2009. The guidance has served as a “gold standard” on PROs, yet it is “very stringent,” Kluetz said. He emphasized that he is “not a guideline basher,” but he was clearly unafraid to question that approach and urge more flexibility at an open FDA meeting.

Kluetz’ leadership may have helped influence the oncology office – which historically has approved few products with PROs in labeling – to invest in the area. The office has added “PRO leads” to the staff, and associate directors of labeling in each of its clinical divisions to provide PRO and labeling expertise, and to work with the SEALD staff, he said. OHOP has also instituted a monthly hematology/oncology PRO case series to bring patient outcomes issues more clearly before the review staff.

Toward the end of the meeting, Kluetz reiterated that he is willing to work with patient reported data to try to make it useful to development and labeling. “We take exploratory patient-reported outcome data seriously....We look at it very carefully if it goes the right way and very carefully if it goes the wrong way.”

“Just because a patient-reported outcome that was measured was not statistically in the hierarchy of primary or secondary endpoint,” the oncology division will not rule out those data. “We look at it at great length.”

Using pain or time to pain progression as examples, Kluetz said he would examine the results carefully: “I will test it; I will beat the result up from all sorts of angles to see if it holds up to scrutiny – looking at different definitions of time to pain progression, how much missing data.” He noted that exploratory data on pain progression in one trial was reflected in labeling as supportive data to another endpoint: time to opiate progression. (He was likely referring to Janssen Inc.’s Zytiga for metastatic prostate cancer.)

Kluetz also gave advice on how to make PRO data more convincing in the oncology area. He suggested small secondary studies to focus narrowly on the patient-reported characteristic instead of trying to build that feature into the original, larger studies.

“What we see frequently,” Kluetz said, “is that everyone tries to put everything into one trial — so you have your [progression free survival] primary endpoint, but there is also a PRO endpoint and several others. Typically, that is not a good way to optimize patient reported outcomes.”

“What might be a better strategy is to look for a smaller second trial that is enriched with those who have symptoms of the disease so that you could look for a symptom,” he said. For example, sponsors could use a “palliation endpoint rather than a time to symptom progression endpoint, which is harder to show and requires a lot more sensitivity.”

Kluetz also suggested that sponsors attempt to ask more descriptive questions to patients to find out “what the patient is feeling while taking the drug.

A New Set of PRO Meetings with FDA

The showcasing of different PRO activities across FDA was not the actual purpose of the meeting: main objective was to announce the result of the one-year review of the clinical outcomes assessments qualification process and unveil the key first “product”: a planned new FDA website page with a compendium of COA tools.

That website would be a de facto way for both FDA and drug developers to get wider and earlier attention to some assessment tools – and offer an FDA-controlled way for publicizing tools without going through the long processes for qualifying them completely.

The next concrete step toward accelerating new types of patient-oriented endpoints may be to create a formal mechanism for sponsors to meet with FDA review divisions about new assessment tool development. That is likely to be a topic for the Prescription Drug User Fee Act reauthorization negotiations this year.

One of the industry panelists at the April 1 workshop, Katarina Halling (PRO lead at AstraZeneca PLC), noted that “currently there is no precedent for FDA to review our efficacy and safety endpoints before the end of Phase IIb.” Halling said that is “simply too late if we need to develop new COA endpoints or if we need to modify existing ones.”

She noted that there has been discussion of creating a new, voluntary meeting for the sponsor to request earlier in development. “We believe that could be a really important meeting. In addition, there were suggestions for increased opportunities for follow-up interactions with the FDA after formal discussions, in order to clarify or have follow-up conversations.”

The interest in new endpoint meetings – which would presumably engage representatives of the specific review divisions to handle planned filings – reflects a movement by sponsors away from the centralized SEALD team as the primary point of contact for discussion of new clinical outcome assessments.

If a new set of meetings is built into the regulatory development process, it will be resource-intensive for FDA, which is likely to put it directly on the agenda for the next user fee reauthorization discussions. That, in turn, makes it an action item that can be accomplished between industry and FDA without relying on separate legislative authority beyond ratifying the PDUFA VI agreement.

Pfizer Inc. Senior Director Global Patient Affairs Roslyn Schneider suggested adding even another round of meetings on the endpoint development topic: follow-up sessions with sponsors, review divisions and patients to act on the voice-of-the-patient comments collected by FDA from its patient-focused meetings. That has already been flagged by both FDA and industry as a topic for PDUFA VI and the “21^st Century Cures” process already under way in the House.

The April 1 workshop already fulfilled several political objectives for the agency: completing an obligation under PDUFA V, answering a request from Congressional appropriators and serving as a good opportunity for the agency to show mockups of the type of listings of assessment tools it already has to underscore the argument that further legislation in the area is unnecessary.

Web Compendium to Reflect Continuum of Endpoint Development

For now, FDA is focusing on making more public the extensive work over the past decade on assessing and using tools for patient report outcomes.

At the April 1 meeting, FDA said an assessment tool compendium webpage could be ready in pilot form before the end of this fiscal year (September 30); it will be announced via the Federal Register. A final version of the compendium will take much longer to achieve and will be accompanied by a draft guidance.

The web compendium itself grew out of discussions in other settings. The Engelberg Center at Brookings convened two meetings in 2014 at FDA’s behest with broad groups of stakeholders (industry, academia, patient advocacy groups and government) where the website idea was first presented.

At the Brookings sessions, the website was described as a way to “identify endpoints that, while not yet fully qualified, are currently considered potentially acceptable for use in particular therapeutic areas.” A short-form name for the list, the potentially acceptable endpoints (PAE) list, was developed.

FDA has backed off that “potentially acceptable” name to the more neutral “compendium” and also taken out some of the information from a mockup chart created at the second Brookings meeting in October.

SEALD staff at the April 1 workshop explained that the agency did not want to include “too much detail” about outcome assessment tools listed in the compendium because FDA did not want to cut off interaction with tool developers. One of the purposes of the listings, FDA said, is to encourage companies to come into the agency to talk and also to encourage collaborative work.

SEALD staffers also described the compendium as a way to portray the “continuum” of projects under way on clinical assessment tools in each area instead of the current focus just on fully qualified products.

One panelist at the April workshop, Jean Paty (Quintiles Inc.) said that FDA’s slide depicting the recognition of the clinical outcome assessment development process as a continuum approach was a “cool slide”, indicating a significant change in thinking by the agency.

Stephen Coons from the PRO Consortium noted that one of the suggestions when the PRO effort was getting underway in 2006 was for FDA to compile and publicize a list of clinical outcome assessment projects that had been completed or were in development. The current effort essentially shows FDA returning to an earlier idea.

FDA’s plan to use a website to get information out without following the formal, centralized qualification process developed by the agency matches some other recent web workaround approaches by the agency and indicates an emerging quasi-regulatory tool. In another such example, the agency announced the creation of a biomarker website wish-list catalogue in mid-February of this year.

The clinical outcome assessment and biomarker projects are being handled by the same group in the agency, so the similarity is not unexpected. The goals are also similar: highlighting areas of progress –not just the all-too-short list of fully qualified tools.

Having a pilot version of the compendium publicly and electronically available by late this year would be important for FDA in the upcoming PDUFA VI discussions and as a factor in the ongoing Congressional efforts to stimulate biomedical innovation.

Opening up labeling to more patient-reported outcomes data could act as a relief valve for the pressure FDA continues to feel to permit broader commercial discussion of real-world data. This project may also give FDA a viable alternative to head off legislative calls for changing the evidence standards for data in labeling and promotions.

PRO-ving Ground: FDA Review Teams Commit to Patient-Reported Outcomes

More from Approval Standards

More from Pathways & Standards