Opening up the findings of drug trials
When a major medical journal raised questions about heart risks from a leading diabetes drug earlier this year, many patients began avoiding the drug. But concern turned to dismay when news emerged that both the Food and Drug Administration and the manufacturer had also picked up worrisome signals about the drug -- but had not sounded an alarm, pending further study.
Today, the drug, Avandia, remains controversial. But the larger issue is what kept warning signs about Avandia from reaching the public for months. The labyrinthine system for reporting on clinical trials of new drugs makes it hard even for scientists to tell what’s been found.
Now Congress, scientists and advocacy groups want to open up the world of these trials in hopes of heading off problems with new drugs sooner.
“Human subject research is only ethical if it is used to advance science, not the interests of a particular company,” said Sean Hennessy, an epidemiology professor at the University of Pennsylvania medical school. “The major reason why human subject research needs to be made public has to do with that ethical consideration, and the second reason has to do with the public health consequences of the information for other people.”
Although there is wide agreement that the system for reporting on clinical trials needs an overhaul, there are sharp disagreements on just how to do it. And the issue is in danger of getting short shrift as a Sept. 30 deadline approaches for broader House and Senate legislation to revamp the whole drug safety system.
Clinical trials, in which patients use a new drug and are watched closely for signs of trouble, are considered the gold standard to test safety and efficacy. But information about what such trials reveal is often scattered and difficult to study -- even after drugs have been approved for market.
For one thing, there is no central, easily searchable database to which drug makers are required to post all trial results. That makes it difficult for scientists outside the industry to gather data from different trials and make comprehensive studies of the findings, and that can lead to problems similar to those swirling around Avandia.
“The situation now is like delivering a wheelbarrow full of telephone books to somebody and telling them, ‘What you want to know is in here somewhere,’ ” said Dr. Jeffrey Drazen, editor of the New England Journal of Medicine, which published the Avandia article. “You need to have a system that requires registration [of clinical trials] and a succinct summary of the results, so you know what the box score is.”
In the case of Avandia, the controversy continues. FDA safety officers want the drug banned. Outside advisors recommend stronger risk warnings instead. Manufacturer GlaxoSmithKline says Avandia’s heart risks are comparable to those of other pills for treating type 2 diabetes. The FDA is still considering the matter.
A reason for encouraging independent study of clinical trial findings is that doing so can offer clues to risks and dangerous side effects that begin to show up in significant numbers only after a drug is on the market and being used by thousands of patients.
Another reason for such studies is that, once a drug is approved for one use, doctors often begin prescribing it for other conditions. Trial data can help determine whether such uses pose risks.
Also, since most decisions on drug safety involve balancing benefits against risks -- almost no drug is 100% safe -- making it easier to examine clinical trial data helps patients and doctors decide whether a drug is right for a particular situation.
What led to disclosure of the early warning signs about Avandia in the journal study was work by independent scientists who dug out and analyzed data from ongoing trials.
But there are unresolved disagreements in Congress and among experts and advocacy groups about how to improve the system. The dispute has become a sticking point as Congress tries to meet a Sept. 30 deadline for finishing the legislation, the main purpose of which is to provide funding for the FDA and revamp the drug safety system.
A bill passed by the Senate calls for studying how to create a new, easily used database that would make the results of all clinical trials available to the larger scientific community and the public. A parallel House bill spells out in detail how such a database would be configured. Some critics say the Senate plan would give industry too much say in shaping the scope of disclosure.
The issues are technical, but the stakes are large -- involving questions of medical ethics and public health, as well as billions in research investment by drug companies. One thing seems clear: The current hodgepodge of federal, state and industry policies has become an obstacle to disclosure.
The government’s current policy does not require full disclosure on all trials.
Clinical trials for drugs to treat serious or life-threatening conditions must be registered with the National Institutes of Health. Patients and doctors can search the online database at ClinicalTrials.gov to get an overview of ongoing research.
Drug companies do not have to disclose the results of such research, although many do so voluntarily on their own websites.
Both the House and Senate bills would require registration of clinical trials on all drugs -- not just drugs used to treat serious conditions. Only initial, small-scale trials to test if an investigational drug is too toxic for humans would remain exempt.
The dispute is over how to publish the results of trials.
The House bill directs the NIH to set up a database showing results in addition to the current clinical trials registry. The results database would include a nontechnical summary of each clinical trial and its outcome, as well as basic technical findings on effectiveness and safety that are of interest to researchers and regulators.
Results would have to be posted within 12 months after the research is finished.
The Senate bill does not lay out such a specific blueprint. Instead, it directs the NIH to conduct a feasibility study on how to set up the results database. The agency would have to seek input from all interested parties, including industry, the medical community and advocates for patients. Under the legislation, that process could take up to 2 1/2 years.
“The House version is likely to lead to a pretty good database quite quickly, whereas it’s much less clear with the Senate version,” said Peter Lurie, deputy director of the Public Citizen Health Research Group. “The concern is that the Senate version will lead to a kind of purgatory in which the database is promised but never emerges in a useful form.”
The leading drug industry lobbying group has criticized the House requirement for a nontechnical summary of clinical trial results.
“It could be very difficult or virtually impossible for companies to translate complex scientific language into easy-to-read language without omitting critical details of the clinical trial results,” Ken Johnson, top spokesman for the Pharmaceutical Research and Manufacturers of America, said after the House passed its bill in July.
--
ricardo.alonso-zaldivar@latimes.com
