|
ISPOR Comments on the Guidance for Industry - Patient-Reported Outcome Measures: Use in Medical
Product Development to Support Labeling Claims
Introduction by Judith Barr ScD, ISPOR Patient Reported Outcomes Special Interest Group Chair and Associate Professor and
Director, Northeastern University, National Education and Research Center for Outcomes Assessment, Boston, MA, USA
On February 2, 2006, the U.S. Food and Drug Administration
(FDA) issued guidance for industry on “Patient-reported Outcome
Measures: Use in Medical Product Development to Support Labeling
Claims”. The intent of this Guidance is to provide standards to
the pharmaceutical industry concerning the psychometric
properties and procedures that will be expected by the FDA as it
reviews submitted data to support PRO labeling claims. The
Guidance also describes a “Target Product Profile” process by
which the sponsor identifies the desired language for the label
and then, using psychometrically valid and reliable instruments,
links the target label language to a source of data.
ISPOR asked members to comment on this draft guidance. More
than 50 comments were received from 12 countries. Thirty-six
percent of the responses were from academia, 22% from research
institutions, and 42% from pharmaceutical companies. These
comments were summarized by members of the Patient- Reported
Outcomes Special Interest Group following a series of conference
calls.
The ISPOR PRO SIG response was in four parts. The first, the
cover letter, applauded the FDA’s definition of “treatment
benefit”, but cautions that the proposed “gold standards” for
all PRO instruments may paradoxically result in reliance on more
limited, uni-dimensional measures. The FDA was requested to
clarify its PRO standards, and to apply equal standards to
biologic markers and physician/caregiver reports. Additionally
the cover letter highlighted three major issue: 1) ideal versus
acceptable level of psychometric quality; 2) rigor of
requirements for modifying existing PRO instruments; and 3)
premature emphasis on minimal important difference, given the
current state of the science.
The second through fourth parts of the ISPOR PRO SIG comments
insured transparency of the ISPOR response. The second component
consists of the working group’s synthesis of all comments
organized by section; the third, the 50-plus verbatim responses,
anonymized, and organized by Guidance section and line; and the
fourth, the name and affiliation of all responders unlinked from
their comments.
To view the ISPOR comments, see:
http://www.ispor.org/workpaper/ISPOR%20Response%20to%20FDA%20PRO%20Guidance.pdf.
The response was coordinated by the Patient-Reported Outcomes
Special Interest Group Task Force Working Group Chairs and its
PRO/Quality of Life Information in Regulatory and Health
Decisions Working Group. Signers of this letter are the members
of this group who participated in the response effort. The
letter noted that the comments do not necessarily reflect the
views of the broader ISPOR membership.
Division of Dockets Management (HFA-305)
April 4, 2006
Food and Drug Administration
5630 Fishers Lane, Room 1061
Rockville, MD 20852
Re: Guidance for Industry - Patient-Reported Outcome
Measures: Use in Medical Product Development to Support
Labeling Claims [Docket No. 2006D-0044]
Dear Captain Burke:
Thank you for the opportunity to comment on the
proposed FDA Guidance for Industry - Patient-Reported
Outcome Measures: Use in Medical Product Development to
Support Labeling Claims.
The International Society for Pharmacoeconomics and
Outcomes Research (ISPOR) is an international organization
pro-moting the science of pharmacoeconomics and health
outcomes research and is organized to act as a scientific
leader rele-vant to research in pharmacoeconomics, health
outcomes assessment including patient-reported outcomes,
and related issuesof public policy. The Society represents
healthcare researchers and practitioners including
pharmacists, physicians, economists, nurses and
researchers from academia, pharmaceutical industry,
government, managed care, health research organizations,
and purchasers of healthcare.
More than 700 ISPOR members are interested in
patient-reported outcomes (PRO); and nearly 50 members
provided comments and suggestions to this Patient-Reported
Outcomes Guidance. The ISPOR members commenting on this
guidancecame from academia (36%), research organizations
(22%), and pharmaceutical/medical
device/diagnostic/biotech industry(42%). Comments were
received from 12 countries, indicating the world-wide
effect of this FDA PRO guidance on PROresearch. This
summary was developed by the ISPOR PRO Special Interest
Group (SIG).
Overall, The Guidance is a comprehensive document
reflecting the current state of the art of PRO
questionnaire development,validation, and implementation
in clinical research and analysis. The FDA PRO Guidance
makes a major policy statementconcerning the role of the
patient, and patient reported outcomes, in the drug
approval process. ISPOR applauds the FDA’sdefinition of
“treatment benefit” as “An improvement in how a patient
survives, feels, or functions as a result of
treatment.”This focus will provide a balance between the
traditional biologic and physiologic markers and outcomes
that directly measure impact to the patient, as reported
by the patient.
However, the “gold standard” proposed by the guidance
for questionnaire development may be very difficult to
achieve forevery instrument, particularly
multi-dimensional instruments. The desire to
comprehensively collect patient reported out-comes in
clinical trials, through the use of developed and
validated instruments, may be thwarted when the previous
valida-tion procedures do not meet the proposed FDA “gold
standard” level. The intent of the guidance to encourage
more compre-hensive and careful inclusion of PRO may
paradoxically result in a reliance on more limited patient
input through the use ofone-dimensional, symptomatic
measures.
As the FDA clarifies its standards for PRO measures,
ISPOR strongly suggests that it also examine its
requirements for the psy-chometric performance of other
endpoint measures such as clinical and biologic markers
and physician/caregiver reports. In this way consistency
of criteria will be ensured across measurement methods.
A summary of comments by section are given in the
attached document, which includes specific comments by
section, line orpage, and person commenting as well as
institution. The following are key issues to be addressed
in the FDA PRO Guidance:
- Issue One: the guidance states ideal
requirements which may be difficult to meet in every
case. Practical or non-consensual methodological issues
such as the determination of the MID are not discussed.
We strongly recommend inclusion ofadditional language
clarifying that review of PRO data and supporting
information about PRO measures will be based onthe body
of evidence provided and that the FDA recognizes that an
acceptable level of evidence may include less than
thefull set of criteria outlined as best practice in the
Guidance.
- Issue Two: considering any modification of an
instrument as a new instrument may be too strict in the
majority ofcases. Minor modifications or translation of
instruments should not be considered as new and
therefore should not requirea complete “revalidation. We
recommend inclusion in the Guidance of examples of the
types of evidence, short of full validation, that could
be acceptable for modifications of existing instruments
(e.g., cognitive interviews with patients).
- Issue Three: the determination of the MID is
given prominence in the discussion of interpretation of
results. The currentlack of consensus in the field
regarding the value of MID and the best methods for
establishing it make emphasis on MIDwithin the Guidance
inappropriate at this time. We recommend inclusion of
language to indicate that MID is just one ofmany methods
to aid interpretation and that choice of methods should
match individual submission requirements.
Finally, we would urge that FDA consider the addition
of language to emphasize that the document reflects ideal
requirements which have to be evaluated in light of
methodological and practical issues.
ISPOR would be interested in providing a forum to
address these specific issues. |
|
