FDA’s Office of Hematology and Oncology Products is taking steps to reverse the dearth of high-quality, patient-reported outcomes data in cancer drug labeling.

At an April 1 FDA workshop on clinical outcomes assessments, Office of Hematology and Oncology Products Acting Deputy Director Paul Kluetz highlighted some OHOP initiatives aimed at addressing challenges inherent in the use of patient-centered measurements in cancer clinical trials.

The hematology/oncology office’s goal is more detailed, consistent and proactive PRO advice.

The office’s efforts include building dedicated PRO expertise within the oncology and hematology review divisions, coordinating more closely with the agency’s Study Endpoints and Labeling Development (SEALD) staff and making a concerted effort to ensure that sponsors get consistent advice.

“My goal is more detailed, more consistent, more proactive PRO advice,” Kluetz said.

2009 Guidance A Hurdle …

PRO data are not always collected in oncology clinical trials and rarely included in product labeling, although there are a few exceptions to the rule, such as Incyte Corp.’s myelofibrosis drug Jakafi (ruxolitinib) and Janssen Biotech Inc.’s prostate cancer drug Zytiga (abiraterone) Also see "Patient-Reported Outcomes in Oncology: The Jakafi and Zytiga Exceptions" - Pink Sheet, 24 November, 2014..

Kluetz suggested that the reasons for this dearth of PRO labeling data are due to both general regulatory requirements and issues unique to the oncology and hematology field.

From a general standpoint, there has been widespread complaint within industry and other stakeholders that FDA’s 2009 guidance on use of PRO data to support labeling claims is overly stringent and has hindered development, validation and use of such instruments in drug development programs Also see "FDA's Final PRO Guidance Grounds Policy With More Details" - Pink Sheet, 8 December, 2009..

FDA is eyeing a compendium of clinical outcomes assessment tools as one approach to getting more PRO data into product labeling Also see "Patient-Reported Outcomes To Be Available For ‘One-Stop Shopping’ In FDA Compendium" - Pink Sheet, 6 April, 2015.

The 2009 PRO guidance, Kluetz said, “was needed at the time because there were a lot of people doing a lot of different things.”

“It is very stringent, and it is the gold standard for PRO development,” he said. However, “that has led to a lack of instruments, in oncology especially, that have been considered adequate by our colleagues and therefore it’s been difficult to get them in trials.”

… But So Is Nature Of Oncology

Oncology also presents its own set of challenges to measuring patients’ reports of how they feel and function Also see "Patient-Reported Outcomes Tools Should Be Symptom Specific, FDA Says" - Pink Sheet, 11 November, 2013..

Cancer patients can be very sick and are treated with drugs that can cause significant toxicities, including death, Kluetz noted. “We’re a little concerned to approve a drug … based on solely symptomatic evidence,” such as pain, with no other evidence of tumor control, he said.

Through the patient-focused drug development initiative, however, there is a cry both for more quality of life and PRO data and for approval to be based on benefits that patients value more than regulators – as evidenced by the April 2 meeting on metastatic breast cancer ([A#00150413002]).

Oncology trial designs can create problems for the interpretation of PRO data.

“If you have a great drug in oncology these days, you’re probably going to get approved on a single-arm trial because we want to get that to patients,” Kluetz said. “How do you interpret patient-reported outcomes in a single-arm trial? That is something that we all need to get our heads around.”

Even when trials are randomized against a comparator, they are frequently either open label because blinding is thought to be infeasible, or the blind is broken when drug-specific toxicities occur, he said.

Kluetz also pointed to the lack of standardization, from oncology PRO instrument selection to concepts measured, data capture and analysis, hierarchical structuring of endpoints and statistical testing.

In addition, there is less familiarity with PROs in oncology compared to other therapeutic areas.

“We have a very good endpoint in oncology – survival. And we can see our tumors. We can actually visualize our disease,” Kluetz said. “That is not so in many other therapeutic areas.”

He contrasted cancer with therapeutic areas such as psychiatry and irritable bowel syndrome. “They have all sorts of experience looking at patient-reported outcomes because that’s the only endpoint they can measure,” Kluetz said. “It’s a heavy lift to get all the oncologists to change their attitude on patient-reported outcomes and make it an important thing.”

Working Together Toward Consistent Advice

Despite these challenges, Kluetz is hopeful that some recent OHOP initiatives will lead to more high-quality PRO data being included in cancer drug labeling.

The effort starts with ensuring that sponsors receive consistent advice from FDA.

In the past year, OHOP has integrated PRO leads in each of the office’s three clinical divisions.

“These are reviewers whose job it is to understand patient-reported outcomes, to understand the diseases in their division, and to provide advice to all of the reviewers in that division,” Kluetz said. “When they have a question they kick it up to me, who is trying to provide consistency across the entire office.”

The addition of a dedicated PRO expert within each review division was among the internal process changes suggested by clinical drug reviewers in a recent survey of attitudes toward patient input in the regulatory process (see sidebar).

[Sidebar#00150413007]

In addition, more attention is being paid to ensuring that the oncology/hematology clinical review divisions and SEALD reviewers work together in their dealings with sponsors.

Instead of giving sponsors PRO-related advice on separate but parallel tracks, “today we are all talking together – SEALD, the clinical review divisions, our PRO leads,” Kluetz said. “We also have labeling deputy directors in each of our clinical divisions that understand labeling and their implications” and are involved in PRO-related discussions.

An OHOP/SEALD working group also meets on a monthly basis.

And, in a bid to educate oncology/hematology reviewers about PROs, “we’ve integrated a monthly case series where we talk about PRO cases at our office meeting where everyone’s involved,” Kluetz said.

Looking Ahead

While Kluetz is hopeful that these internal changes will boost the use of PRO instruments in trials and, ultimately, the amount of PRO data in labeling, there are also external factors that need to be addressed.

For example, he cited the need not only for better PRO instruments but also for standardization in cancer clinical trials. Certain core concepts – such as disease-related symptoms, treatment-related symptoms and impact on activities of daily living – should be measured in every cancer patient in every clinical trial, Kluetz suggested.

“If you want to use a PRO as a very important part of the efficacy of your drug, you have to be very, very thoughtful in how to use that endpoint,” Kluetz said.

Kluetz also had some clear advice for sponsors looking to use PRO instruments in their clinical trials with an eye toward labeling claims.

“If you want to use a PRO as a very important part of the efficacy of your drug, you have to be very, very thoughtful in how to use that endpoint,” Kluetz said.

“What we see frequently is that everyone tries to put everything into one trial, so you have your [progression-free survival] primary endpoint … but then …also they’re looking to find this one PRO endpoint and several others. Typically that’s not a good way to optimize your patient-reported outcomes.”

“What might be a better strategy is to look for a smaller, second trial that is enriched for those who have symptoms of the disease, and therefore you can look for a symptom palliation type of endpoint rather than a time-to-symptom-progression endpoint, which is harder to show and requires a lot more sensitivity.”

Kluetz also urged a re-thinking in the questions PRO instruments ask.

“It’s not ‘is my drug effective,’ it’s ‘what’s the patient feeling while they’re taking my drug.’ That actually can add value to your proposition as a drug developer because you can show a more clear picture of not only the antitumor activity of the drug but also how it’s affecting the symptoms and the function of the patient.”

Regardless of whether a sponsor wants to answer an efficacy question or a descriptive question with PRO measures, “the data needs to be as sound as possible for either of those, because they’re both critically important,” Kluetz said. “All of that PRO data that comes in with your registration trial we’re going to look at very carefully, and we’re currently looking at ways to best analyze that data and inform our overall risk/benefit calculation on the type of regulatory action that we would take with that drug.”

FDA Oncology Office Hopes Better Advice Will Lead To More Patient-Reported Outcomes Data

More from United States

More from North America