Registration of Clinical Trials Benefits Everyone

Guest Editorial by Andrew P. Schachat, MD

A contemporaneous comparison group is required to provide clinicians knowledge of whether a management approach or treatment option is effective, unless the untreated natural history of the disease is very predictable. I hope it is no surprise to anyone reading my comments that we obtain the best efficacy data from randomized controlled clinical trials. Clinical trials are hard to plan, difficult to conduct, usually are very expensive, and typically rely on the cooperation and collaboration of large numbers of researchers. Of course, society also owes a debt and a thank you to the study participants. Much is invested in these trials, and the data that derives from them is precious.

�Positive� clinical trials, which show that treatment A is better than B, are very useful. Assuming treatment A has better efficacy and comparable safety, clinicians and patients generally will select treatment A unless it is very expensive or inconvenient. We all like the comfort of knowing which treatment is superior. However, there is also value in equivalence, having confidence that A equals B. Treatment A may be much simpler for patients, or less expensive, and if we know it is as good as treatment B, we will recommend it.

�Negative� studies are valuable as well. Treatment with new high-tech treatment C which allows, for example, �small-incision surgery with rapid recovery� may sound great compared to standard approach D, but if there are data from a well-conducted large randomized trial showing that new treatment C in fact does not achieve comparable or better outcomes than old fashioned (and perhaps less expensive) treatment D, we would not recommend the new treatment. The study result was �negative.� If the study is published, we can all learn about the outcomes of treatment C.

Transparency � Key for Ethical Reporting

Sponsors have usually invested a good deal of time, effort, and money developing drug or device or treatment C to get it to the point of human clinical trials. The sponsor of the study with the negative result may choose not to publish the findings. Perhaps they are raising money for other studies and release of a negative result may hinder their ability to raise funds. If a sponsor controls the release of study data, there is a potential conflict of interest. Society usually has an interest in seeing the data and having it released whereas the sponsor may not. The ethical arguments are complex but when research is performed on humans, most medical professionals support transparency and public reporting. I will not list detailed pro and con arguments for release of study data, but if a trial is performed and nothing is ever reported, many of us assume that the outcomes were not wonderful, and if only positive studies are reported, bias seeps into the literature.

The journal I edit, Ophthalmology, supports the registration of clinical trials. When a trial is registered, there is a public announcement that the study exists with a synopsis of the study design and other important information. To encourage registration, many journals, ours included, will not accept manuscripts from trials that have not been registered. It is possible that a private author may forget about this requirement, and even faintly possible that a large sponsor, for example, a drug company, could forget to do this. Since the data may have public health importance, I may make an exception to the policy on a case-by-case basis, but I hope this will not be necessary. Registration has little or no cost and is quick and easy. Disclosure that the trial is being conducted can be accomplished a number of ways, for example, at the http://clinicaltrials.gov Web site.

Andrew P. Schachat, MD, is vice chair for Clinical Affairs at the Cole Eye Institute, Cleveland Clinic, in Ohio and is the editor of "Ophthalmology". Dr. Schachat in the last year has been a consultant for Eli Lilly, OSI/Eyetech, and Neovista.