Comments on FDA Transparency
July 20, 2010
Division of Dockets Management
Food and Drug Administration
5630 Fishers Lane, Room 1061
Rockville, Maryland 20852
To Whom It May Concern:
We welcome the opportunity to comment on FDA Transparency Initiative: Draft Proposals for Public Comment Regarding Disclosure Policies of the U.S. Food and Drug Administration. Public Citizen commends the FDA’s proposals, which represent meaningful progress toward a more open regulatory process at the FDA. At the same time, we urge the agency to go further, as additional transparency will do more to protect public health and safety not only by alerting the public to particular dangers, but also by promoting scientific innovation that has the potential to improve the public welfare. To that end, these comments focus on how the FDA should implement stronger requirements concerning pre-approval data disclosure, data disclosure for abandoned, withdrawn or denied marketing applications, more consistent post-approval review disclosure, and better affirmative disclosure practices as a way of reducing the backlog in processing Freedom of Information Act (FOIA) requests.
I. Strengthen Proposed Release of Pre-Approval Data
To date, FDA has generally declined even to acknowledge that a drug or device product application (that is, an application for authorization to conduct clinical trials) or a marketing application (that is, an application for authorization to sell a drug or device) has been filed, much less to divulge pre-approval safety or efficacy information. In the few instances where disclosure of such information is required, the combination of regulations prohibiting disclosure and regulations requiring disclosure makes the disclosure “difficult to implement,” as noted by the Transparency Task Force. Recommendations at 37.
Public Citizen applauds the Task Force’s recommendations that the existence of and basic information about product and marketing applications be made public. Public Citizen also strongly supports the Task Force’s recommendation that FDA release summary safety and effectiveness information from product and marketing applications.
Public Citizen urges the FDA not to limit the mandated release of summary safety and efficacy information from product and marketing applications to instances where the agency “concludes that disclosure is in the interest of public health.” Recommendations at 48 (Draft Proposal 16). Release of all summary safety and efficacy data is in the interest of public health, because it enhances scientific innovation and discovery. First, scientists may be unaware that prior research has detected a lack of either safety or efficacy and may continue researching related products, squandering time and money and going down roads already proved to be dead ends. Second, progress in science is based on the free publication of study results and on the public release of data, allowing scientists to build on the experience of others.
Releasing all summary data for product and marketing applications would not competitively harm industry firms. Because this recommendation only concerns release of summary data, as the Task Force found, “industry has not demonstrated that blanket protection of aggregate information is warranted to maintain incentives for innovation.” Moreover, industry may realize many benefits from the release of information, including spurring new innovation and curbing duplicative efforts.
Public Citizen also urges the FDA to disclose raw data from clinical trials. While the advent of clinicaltrials.gov, born out of the 2007 FDA Amendments Act, provides some basic information about clinical trials, the information made available on the website does not reduce the scientific utility of providing public access to the underlying patient-level data. As it stands, companies only have the incentive to publish favorable trials, and therefore can have a misleading effect on researchers and the public at large.
The clinical data from one drug will rarely enable another firm to take any short-cuts or realize any savings on clinical trials of a different drug. Therefore, the release of clinical data will not cause the sponsor competitive harm. In particular, the safety data, which is of the highest public concern, will rarely have any bearing on work on another drug, even if related. Absent a particularized showing about how another firm could use the data to gain a competitive advantage, (as opposed to, for example, fear that the data would provoke public criticism), courts generally have found that clinical data submitted to FDA is not confidential commercial information.
In addition, the FDA should post advisory committee materials online at least one week in advance of each FDA advisory committee meeting. Currently, the FDA posts the documents a two days before each meeting. However, the material has been compiled and sent to advisory committee members three to four weeks earlier. Even allowing time for FDA to review the documents so as to redact portions that the agency believes to be confidential, the online posting could and should occur sooner. The late posting makes meaningful public participation in advisory committee very difficult, as the time in which to review and evaluate the materials is very short. Accordingly, posting the materials online one week in advance of each meeting would be an important step in enabling public participation in advisory committee meetings and in increasing public confidence in the FDA review process.
II. Release Clinical Data From Abandoned, Withdrawn, or Denied Marketing Applications and Supplemental Marketing Applications
Public Citizen supports the Transparency Initiative’s draft proposal 11 concerning the release of any safety concern associated with abandoned, withdrawn, or denied marketing applications. Simply releasing a summary of safety concerns, however, stops short of the disclosure best tailored to serve the public interest. Summary safety and efficacy data should be available for all abandoned, withdrawn, or denied marketing applications to enable researchers to build off the work of one another. Clinical data pertaining to denied, abandoned, or withdrawn supplemental marketing applications is also critical to doctors prescribing off-label uses of drugs. Finally, for applications that have been denied, detailed patient-level data should be available to researchers to promote scientific discovery that could save patient lives. The benefits described above with respect to pre-approval data are equally true for these categories of applications.
For example, when the FDA (CDER), in accordance with its practice, released the medical officer’s review and other material after approving three of four uses in the NDA for Valdecoxib (marketed as Bextra), it redacted information discussing the fourth use (treatment of acute pain), the approval of which it had denied. After Bextra came on the market, it was touted in a journal article for the unapproved use, and many physicians prescribed it for that use. In response to a lawsuit filed by Public Citizen, the FDA released some of the information redacted from the medical officer’s review, which showed that the FDA had denied approval for use to treat acute pain because of safety concerns. Proactive release of the portions of the medical officer’s review addressing the unapproved use would have protected patients put at risk by this dangerous drug (sine withdrawn from the market entirely, for safety reasons).
III. Consistently and Promptly Provide FDA Reviews of Approved Drugs Online
Several components of the FDA, including CBER and CDER, provide reviews by categories in an easily accessible online format, which can be accessed by looking up any particular drug. By contrast, other components, such as CVM, do not provide the reviews of approved drugs online. Forced to request the reviews under FOIA, members of the public interested in the reviews may wait for years to receive them, and even then, may not know if they have received all of the reviews.
Every component that reviews and approves the marketing of products should affirmatively post all reviews by drug, so that consumers can search by drug and access all FDA reviews, organized by category, such as medical officer reviews, clinical pharmacology reviews, and statistical reviews.
IV. Reduce FOIA Backlogs
Public Citizen supports the Transparency Task Force’s observation that affirmative disclosure can “free up Agency resources for other activities that improve public health.” Recommendations at 20. In the context of FOIA, the FDA should prioritize proactive release of information that might help realize these efficiencies. At the end of fiscal year 2009, the FDA had a backlog of 4,818 FOIA requests. Although the FDA has reduced its backlog in recent years, the backlog remains significant and delays the release of important information to the public. Moreover, the FDA spent more than $21 million in FOIA processing costs in fiscal year 2009, representing by far the highest expenditures of any component of the Department of Health and Human Services.
Public Citizen encourages the FDA to adopt a routine practice of affirmatively disclosing as much data as possible as a way of cutting back on the FDA’s FOIA backlog and freeing up agency resources. The recommendations of the Transparency Initiative in this regard, in conjunction with Public Citizen’s recommendations above, would go a long way toward meeting this goal and will serve the public’s interest by reducing the delay associated with FOIA requests and with receiving vital information.
Thank you for your consideration.
Sidney Wolfe, M.D.
Public Citizen Health Research Group
Public Citizen Litigation Group
 The draft proposals define “product application” to include Investigational New Drug Application (IND), Investigational Device Exemption (IDE), Notice of Claimed Investigational Exemption for a New Animal Drug (NCIE), and Investigational New Animal Drug (INAD) files.
 The draft proposals define “marketing application” to include New Drug Application (NDA), Abbreviated New Drug Application (ANDA), Biologics License Application (BLA) New Animal Drug Application (NADA), Abbreviated New Animal Drug Application (ANADA), Premarket Approval Application (PMA) and 510(k) notification.
 Meier B., Contracts keep drug research out of reach, New York Times A1 (Nov. 29, 2004); Whittington CJ, Kendall T, Fonagy P, Cottrell D, Cotgrove A, Boddington E., Selective serotonin reuptake inhibitors in childhood depression: systematic review of published versus unpublished data, 363 Lancet 1344-45 (2004); Natanson C., Kern S, Lurie P, Banks SM, Wolfe SM, Cell-free hemoglobin-based blood substitutes and risk of myocardial infraction and death: a meta-analysis, 299 Journal of American Medical Association 2304-12 (2008).
 Aaron Kesselheim and Michelle M. Mello, Confidentiality Law and Secrecy in Medical Research: Improving Public Access to Data on Drug Safety, 26 Health Affairs 483, 489 (2007).
 Public Citizen Health Research Group v. Food and Drug Administration, 185 F.3d 898 (D.C. Cir. 1999); Public Citizen Health Research Group v. Food and Drug Administration, 2000 U.S. Dist. LEXIS 4108 (D.D.C. 2000); Public Citizen Health Research Group v. Food and Drug Administration, 964 F. Supp. 413 (D.D.C. 1997).
 As an example, the reviews for Orlistat are available here: http://www.accessdata.fda.gov/drugsatfda_docs/nda/99/020766a.cfm
 Public Citizen has recently had precisely this experience in requesting the reviews for Slentrol from CVM.