- BIOETHICS FORUM ESSAY
Hidden Data at the FDA
When a doctor makes a treatment recommendation to a patient, he should do so on the basis of information about what works, and what doesn’t. To do this, he should integrate a basic understanding of medical science with the results of ongoing research, including the results of clinical trials that evaluate the benefits and harms of medications and medical devices. Physicians may get summary results of such trials at medical meetings, but typically they rely on the full reports published in medical journals. Yet there is another repository of trial results that remains inaccessible: the FDA.
When a pharmaceutical company initiates the development process of an investigational drug, it does so under a FDA-approved Investigational New Drug (IND) application. Many related trials, starting with a small group of healthy volunteers, may subsequently be completed under one IND. When a particular development process fails to produce an approvable drug (which happens much more often than not) the results – and even the existence – of these completed trials are not generally disclosed. In other cases, only the results of positive trials proffered by the Sponsor to the FDA in support of a marketing application may be made public – and typically published in leading medical journals with much hype – while the results of completed negative trials are withheld.
The ostensible reason why the FDA does not ultimately make all the trial results in its possession public may be found in a recent report in Slate by Jeanne Lenzer, who described an unsuccessful attempt to get information in the possession of the FDA about serious adverse effects, including suicidality, possibly caused by anti-depressant medications. The claim is that such information is considered proprietary, protected from disclosure by statutes concerned with confidential commercial information and trade secrets. In an interview with Lenzer, Robert Temple, acting director of the FDA’s Office of Medical Policy, articulated a version of the oft-repeated reason: “My hands are tied. This is something only Congress can change.”
The claim is disingenuous in the face of a detailed argument to the contrary made by the FDA itself. On January 18, 2001, the FDA proposed to amend the regulations to make a wide variety of information usually included in an IND publicly available for two categories of clinical research: xenotransplantation (involving animal tissues) and genetic transfer. The FDA discussed the familiar concerns about releasing confidential commercial information, but then claimed it could act to enable the release of such information under provisions contained in the Trade Secrets Act. Its argument appeared in the Federal Register:
That statute prohibits the disclosure of confidential commercial or trade secret information, except as “authorized by law.” Because agency regulations that specifically provide for the disclosure of such information can supply the requisite legal authorization for release of the information for purposes of the Trade Secrets Act, that statute would not present a bar to any of the disclosure contemplated by this proposed rule. The broad rulemaking authority conferred on FDA by Congress under the act (21 U.S.C. 201 et seq.) permits the agency to amend its regulations as contemplated by this proposed rule. Section 505 (i) of the act (21 U.S.C. 355(i)) gives FDA the authority to issue regulations imposing conditions on the investigation of new drugs. In addition to prescribing certain mandatory conditions, that section may impose “other conditions” as necessary “relating to the protection of the public health.” (21 U.S.C. 355(i)).
Nothing has come of this proposal since its publication, notwithstanding the FDA’s candid comment in the notice of the proposed rule that “The public health often is served not only by the collection of research data and information, but also by disclosure of such information.”
The extent of such undisclosed information held by the FDA was recently chronicled by Erick Turner in the Public Library of Science. Turner, a former reviewer of psychotropic drugs at the FDA, said that the FDA maintains its own registries of completed trials, and noted that the details of some of these trials are disclosed when public meetings of the FDA’s Advisory Committees are held. He went on to propose that all internal FDA reviews of approved new drugs be made public. But this does not go nearly far enough. Much of the relevant data for these drugs would remain undisclosed, along with all of the data presented in applications for drugs that the FDA does not approve.
The potentially harmful effects of keeping trial information hidden are considerable. Because the amount of published research is voluminous, physicians typically rely on summary reviews or practice guidelines published by colleagues deemed “experts” in a particular medical realm. The credibility and informativeness of such reviews and guidelines are critically dependent on a rigorous appraisal of all the research results relevant to the topic at hand. When evidence from clinical trials of the inefficacy or harmful effects of interventions is not made publicly accessible, a distorted perspective of the merits of medical interventions is inevitable.
Nondisclosure of research data may even adversely affect the research of investigators not involved in commercial drug development. Consider the development of resistance to the antiretroviral drugs used to treat AIDS. Stanford University researchers maintain an online database of information concerning the development of resistance to antiretroviral drugs. The database is intended to facilitate research performed in the broad community of scientists. In an open letter to the FDA, Dr. Robert Shafer, principal investigator and scientific curator of the project, calls for the public release of drug resistance data submitted by companies in support of applications to market new agents. He writes that “HIV-1 drug resistance – like immunological escape from vaccine-induced immunity – looms as a problem that cannot be solved by any single study or clinical trial but only by the effective synthesis of data from multiple studies and clinical trials.”
Commercial sponsors will always maintain that the compelled public release of all trial results will inhibit the conduct of drug research and development by eliminating the competitive advantage over its rivals that secrecy purportedly affords a company. No evidence in support of this argument is offered. And since companies typically obtain protective patents very early in the development process, the assertion is very much open to dispute.
The FDA needs to reassess how it is discharging its responsibilities to the American public. Many already believe that the FDA exhibits undue deference to the commercial interests of pharmaceutical and biotechnology companies. “Regulatory capture” of the FDA would undermine a public good and threaten a pervasive loss of the public’s trust.
The FDA should amend the federal regulations to ensure that the raw results of all clinical research conducted by industry, and submitted to the FDA, are released to the public at an appropriate time in the development process. This might be immediately after approval of an investigational drug, or it might be after a drug company has stopped sponsoring trials of a particular drug. The industry is likely to litigate the new regulations vigorously. But this should not deter a committed federal agency. And on the upside, for the FDA: If the FDA acts on the problem, it might be able to allay the unrelenting calls that Congress step in to correct this ongoing threat to the public’s health.
Published on: June 15, 2006
Published in: Clinical Trials and Human Subjects Research