January 1, 2005 (Vol. 25, No. 1)
A Bandwagon in the making
There has been a substantial amount of attention paid in recent months to the issue of making information on clinical studies of marketed pharmaceuticals publicly available.
The original catalyst was a highly publicized settlement of fraud charges between New York state and GlaxoSmithKline (Middlesex, U.K.), which involved Glaxos alleged selective publication of studies involving the use of its antidepressant, Paxil, in a manner that obscured data showing an increased tendency for patients in younger age groups to demonstrate suicide-linked behaviors.
Following that settlement, a similar one was reached shortly afterward by New York with Forest Laboratories (New York City), regarding alleged off-label uses of its antidepressant drugs Lexapro and Celexa.
Both companies agreed as part of their respective settlements to publish online registries of studies involving their drugs. The publicity surrounding these settlements quickly evolved into the more global issue of the publication of negative drug studies in general.
Interest in what was a fairly newsworthy issue in its own right, however, became quickly eclipsed by Mercks announcement to remove from the market its huge-selling COX-2 anti-inflammatory drug, Vioxx, due to clinical trial results showing a significantly high rate of associated cardiovascular events following long term use.
In light of a decades long environment in which approval times of drugs have been under intense pressure to be shortened, these episodes generated interest in the issue of whether safety-related results of clinical drug trials are getting appropriate exposure and interpretation.
Rush to Registries
Remedies for this perceived problem have taken several forms, the most conspicuous are recommendations for the creation of registries of planned, ongoing, and completed clinical drug trials. Prominent medical journals have announced a plan that would require the registration of all effectiveness studies at their inception beginning in mid-2005 as a condition for the results of those studies to be published when completed.
In addition, certain legislators, notably Senator Charles Grassley of IA, announced plans to introduce federal legislation for a web-based registry. Moreover, a representative of the New York Attorney Generals office announced recently that further investigations can be expected.
Pharmaceutical companies have introduced their own voluntary registries. In addition to Glaxo and Forest, Eli Lilly, Pfizer, Bristol-Myers Squibb, and AstraZeneca, as well as the drug industry association PhRMA, have all begun some sort of voluntary, web-based public drug study registry. Even the FDA has been said to ensure that lists of trials it reports on the existing government website are more comprehensive than before.
The publicity surrounding the issue has led to questions regarding what the present requirements, if any, for the listing of drug studies are. Some confusion was generated by an FDA reaction at the time of the Glaxo settlement, suggesting that requirements for the public listing of such trials were already in effect.
The truth is, as a general matter, they are not, though there are some existing requirements that apply to a more limited universe of studies. Enforcement of these requirements to date, however, has been limited, if not nonexistent. What follows is a brief discussion of the history and nature of these requirements.
Section 113 of the FDA Modernization Act of 1997 (codified at 28 USC 282) required the Department of Health and Human Services to establish a data bank of clinical trials for effectiveness of drugs for serious or life threatening diseases or conditions.
Paraphrasing the statute somewhat, the data bank is required to include a registry of clinical trials of experimental treatments, providing a description of the purpose of the experimental drug, when the trial begins, eligibility information, trial sites, a point of contact, and other information.
Similar information is required for treatment INDs for similar types of drugs and Group C cancer drugs. The criteria for determining whether a disease or condition was serious or life threatening was defined by FDA regulations (21 CFR 312.81), a principal purpose of which was for determining whether drugs are eligible for certain expedited or fast track review procedures.
The fairly obvious purpose of this statutory section is to advise members of the public about the existence of planned or ongoing clinical trials for serious or life-threatening diseases in order that they may be aware of them, and to facilitate enrollment when appropriate. This is quite evident from the registry itself, which is found at www.clinicaltrials.gov. The FDA has published the following guidance on this subject, found at www.fda.gov/cder/ guidance/4856FNL.PDF.
While it states that deciding which drug studies qualify for inclusion is ultimately a matter of judgment, it identifies relevant criteria for study inclusion in detail, and includes, by way of example, drug studies for diseases such as AIDS, HIV, Alzheimers disease, angina pectoris, heart failure, cancer, and others.
In addition, as the criteria include diseases which can be well managed using existing therapy, but which can have serious outcomes if unmanaged, the list also includes studies for diseases such as inflammatory bowel disease, asthma, rheumatoid arthritis, diabetes, depression, and others, including any drug obtaining a fast track designation.
Companies in the process of conducting effectiveness trials for the diseases identified above, or which fit the fairly expansive definition of serious or life threatening diseases, should examine whether those studies should be listed on the registry.
These are the only known governmental requirements of general applicability regarding the registration of clinical trials to date. Given the impetus created by the New York state settlements and the subsequent furor created by the Vioxx market withdrawal, expanded requirements, most significantly for studies of marketed drugs such as those involved in the New York litigation which the companies themselves are now voluntarily including in company-based registries, are essentially a certainty.
These developments would seem on the surface to be very desirable. But underneath the surface lies a potentially significant political issue that bears at least some consideration: will there be an effect on the FDAs traditional role of interpreter of such studies for the medical community?
In drafting new product labeling, and in revising it during a products marketing life, the FDA is responsible for weighing the significance of available information and drawing the most appropriate medical conclusions from it. Not every new potential risk is accepted, as initial presumptions need to be overcome by evidence whose weight is largely cumulative.
Moreover, evaluating this type of dynamic information flow is not readily left to individual practitioners or others who are not experienced in doing so. Perhaps the issue would be less troublesome were it one of keeping information from the agency. But it is not.
In the episodes described above, the FDA was not in the dark at the time the issue became one of public interest. While companies may be selective in the studies they publish, agency rules do not permit companies to be nearly as selective in the studies they bring to the attention of the agency, either prior to or after approval.
Both the association of antidepressants with suicide based-behaviors in children and Vioxx link to cardiovascular issues were known to the FDA before they became front-page events. In both cases, the question was not so much of what was known, but whether the institutional response to what was known was appropriate.
It is too soon to know exactly what the responses to these issues will be, though required publicly-available study registries for both marketed and under-development drugs are almost a certainly. Other suggestions have included a more prominent drug safety office within the FDA to the establishment of an agency outside the FDA, a sort of NTSB for drug safety, to monitor drug safety.
It remains to be seen whether what does occur will impact the FDAs traditional role as the sole regulatory filter for drug safety information, and whether such a result is a positive one.