Wrangling Over Research Results

The clinicaltrials.gov database is growing fast as research sponsors post more trial listings and begin to add information on the results of studies, as is required by law. The FDA Amendments Act of 2007 (FDAAA) established stiff penalties for failing to comply with posting requirements, resulting in a surge of data on more than 71,000 studies in the system operated by the National Library of Medicine (NLM) at the National Institutes of Health (NIH).

Sponsors also have begun submitting results on listed studies, creating a database that now has some 700 records, with more than half coming fromfrom industry sources. Currently, submissions consist of tables listing basic results, but the system is slated to add narrative summaries of results next year—a proposal that has given rise to objections about summaries being biased as well as promotion of unapproved drugs.

At an NIH public forum in April, pharma companies and other stakeholders discussed a range of concerns, such as whether results must be filed for discontinued studies, how to harmonize reporting standards with other countries, and who should write narrative summaries. Consumer advocates and medical journal editors argued that sponsor summaries would allow marketers to publicize not-yet-approved treatments. The Consumers Union proposed that NIH collect user fees to support write ups of objective trial summaries. If pharma companies are to do the job, though, they want to follow the International Conference on Harmonization E3 study synopsis format, and get assurance that FDA will not bring charges of illegal promotional activity based on required results disclosure.

A related issue is whether providing summaries reduces the need to post full clinical trial protocols. Sponsors fear that publication of complete protocols would reveal proprietary information on how data is assessed, and also raise privacy issues by disclosing the names of investigators and subinvestigators. Pharma companies also want leeway to determine which discontinued studies warrant reporting of results—preferably only those dropped for safety reasons.

Meanwhile, sponsors have to start filing reports on serious and frequent adverse events in clinical trials beginning in September 2009. FDAAA calls for providing this data in a tabular format, but it’s not yet obvious that the information will be useful to patients or researchers. Sponsors seek to clarify what events are considered “serious,” and want time frames for submitting AE reports as well as common AE reporting requirements for NIH, FDA, and other entities.

Industry moved to defuse some of the opposition by unveiling a revised version of PhRMA’s voluntary clinical trial disclosure policy, which now supports “timely registration of all interventional clinical trials involving patients.” That still would leave out most Phase 1 safety studies, which primarily enroll healthy volunteers. PhRMA also would limit reporting on Phase IV trials to those that are interventional, as opposed to the vast spectrum of observational studies.