CLC comments on FDA plan for sharing
masked and de-identified data
(October 31, 2013)
October
31, 2013
Margaret A. Hamburg, M.D.
Commissioner
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, Maryland 20993
RE: Docket No.
FDA-2013-N-0271, Availability of Masked and De-Identified
Non-Summary Safety and Efficacy Data
Dear Dr. Hamburg:
The undersigned organizations representing cancer
patients, health professionals, and researchers appreciate
the opportunity to comment regarding a Food and Drug
Administration (FDA) proposal to make de-identified and
masked data derived from medical product applications
available to interested parties. We commend FDA for
advancing this proposal and for reopening the comment
period for additional public input. In the comments
below, we offer recommendations for ensuring the best use
of such data as well protecting the privacy of clinical
trial participants whose data might be shared. We
also identify concerns about the potential cost of the
plan for sharing preclinical and clinical data.
Potential Benefits of Sharing Preclinical and Clinical
Data
We agree with the agency that the proposal for sharing
preclinical and clinical data from product applications,
which holds the promise of generating new knowledge beyond
the initial clinical trial, honors the commitment and
altruism of clinical trials participants. It is our
hope that the pooling of preclinical and clinical data
from multiple trials and multiple applications might lead
to additional information about the safety and efficacy of
products and even classes of drugs. Cancer patients
have historically tolerated significant toxicities for
treatment benefit. However, cancer survivors are
increasingly concerned about the long-term and sometimes
delayed significant side effects of treatment. The
data-sharing plan may hold the potential for pooling of
data that would reveal more information about such late
effects of treatment and contribute to strategies to
mitigate the adverse impact of cancer treatments.
Protecting Patient Confidentiality
The agency has stated that any preclinical and clinical
data that are available for sharing would be stripped of
any information that could identify patients or research
subjects. FDA identifies the standards of the
Privacy Rule as those that would guide de-identification,
while at the same time noting that the agency is not a
covered entity under the Privacy Rule.
We recommend that the agency consider standards that go
beyond the Privacy Rule criteria for de-identification in
certain cases or that certain data not be made available
for additional review and research. For example,
trials for some rare cancers and most pediatric cancer
trials have such limited enrollment that de-identification
according to the Privacy Rule standards is inadequate and
individual trial enrollees could still be
identified. We recommend against sharing of data
from these trials, unless FDA can develop a standard for
de-identification that will truly protect the
confidentiality of pediatric cancer patients and others
with rare cancers. We understand that such a
prohibition against sharing these data might diminish the
impact and benefits of the proposal, but in the cases
described above protecting the privacy of clinical trial
participants should be honored.
Consent to Share Preclinical and Clinical Data
For those participants whose trials have already been
completed but whose data are subject to sharing, we
recommend that the terms of the consent for clinical trial
participation govern the availability of the data.
In other words, the data should only be shared for
research efforts that are related to the original
consent. This is the most appropriate
way to honor and respect the consent granted by clinical
trial participants.
For data sharing on a prospective basis, we anticipate
that clinical trial sponsors will amend their consent
standards to address potential opportunities for sharing
of data and as a result may be able to obtain informed
consent for such practices.
The matter of patient consent for data sharing is one that
deserves more attention from FDA and possible new
approaches.
Standards for Granting Access to Data
In its original notice regarding the proposal for
availability of preclinical and clinical data (published
in the Federal Register of June 4, 2013), FDA did not
provide details about the procedures that it will follow
for making preclinical and clinical data from product
applications available to researchers. We recommend
that the agency develop specific standards to ensure the
appropriate use of patient-specific data and that the
agency retain responsibility for determining those
entities that should be granted data access. These
standards should advance two goals; they should: 1)
maximize the research benefit related to availability of
trials data, and 2) protect the interests of innovative
research companies. Data should not be made
available to entities that will use them in support of
marketing applications in other regulatory systems.
Costs of Data Availability
The system for managing the data-sharing opportunities,
which we reference above, will be accompanied by some
costs to the agency. We believe that the standards
for data sharing are critical to the success of the
effort, and as a result the costs cannot be
eliminated. We also understand that the management
of the preclinical and clinical data will result in some
financial burden to FDA. Although the proposal
related to availability of data is an important one with
potential to advance research and foster new knowledge, it
should not move forward without a clear financing plan
that does not adversely affect the speed and quality of
new product review.
As advocates for transparency in research and research
data sharing, we have been strong supporters of
clinicaltrials.gov and legislative efforts to refine and
strengthen data sharing through that system.
However, the system is not a user-friendly or efficient
one. As a result, many other organizations have
attempted to address the weaknesses of clinicaltrials.gov
through efforts that are duplicative and expensive.
Our enthusiasm for development of a system of availability
for preclinical and clinical data is tempered by our
concern that all details of the system – including cost –
be addressed before the program is launched.
We appreciate the opportunity to offer these comments and
look forward to working with FDA on its plan for data
sharing.
Sincerely,
Cancer Leadership Council
American Society for Radiation Oncology
Bladder Cancer Advocacy Network
CancerCare
The Children's Cause for Cancer Advocacy
Fight Colorectal Cancer
International Myeloma Foundation
Kidney Cancer Association
The Leukemia & Lymphoma Society
LIVESTRONG Foundation
Lymphoma Research Foundation
National Coalition for Cancer Survivorship
National Lung Cancer Partnership
Prevent Cancer Foundation
Us TOO International Prostate Cancer Education and Support
Network