By Mark D. Uehling
May 12, 2005 | Beginning in July, several powerful medical journals – including JAMA, the New England Journal of Medicine, the Lancet, and Annals of Internal Medicine — have promised to only publish articles about clinical trials that have been logged into a public database upon the start of patient recruitment.
The pharmaceutical industry, however, has its own ideas about what it is comfortable divulging. A Bio-IT World investigation reveals that the registry visions of the industry and the medical editors only partly overlap – see chart.
In fairness, two drug companies — Glaxo and Lilly — have found religion. After taking a public relations beating over unpublished data relating to antidepressants given to adolescents, the two companies erected their own exemplary, public-spirited registries in an attempt to dampen anti-industry rhetoric that only the most rosy, marketing-friendly clinical trial data get published. (Glaxo and Lilly declined to comment.)
Christine Laine, senior deputy editor of the Annals, supports industry efforts, but she’s not budging — at least in public. Laine notes that registration long before publication is a central demand that would aid scientific understanding of the trial. Only by knowing the trial’s key milestones and endpoints, she says, can outsiders be confident that disappointing results have not been sugar-coated — or dumped in some obscure clinical journal. “There are many trials that never see the light of day,” notes Laine. “It’s a great competitive advantage to decide which results you want to make public and which you don’t.”
At press time, the industry’s preference is to have a “lock box.” (The phrase was borrowed from the imaginary federal account into which Social Security taxes are supposed to be deposited.) Laine and colleagues appear unenthused about a clinical lock box that would allow a company to register a study — and divulge the details later. “It depends on what’s in the lockbox,” says Laine. “If the basic elements are locked and not publicly available, then the trial registration is not achieving the objective of public accessibility.”
Laine declines to say what sort of conversations may be under way with the industry: “I think we have heard the concerns of industry. There is not a lot of movement on what we want. We still want those data elements at the time the first patient is enrolled in a publicly searchable, non-profit registry.”
Ironically, both sides agree that more than a handful of major registries will create a mess for patients and researchers. “It’s going to be a problem if there is a proliferation of trials and registries,” says Laine. “We hope every trial will have a unique identification number.” The NIH’s ClinicalTrials.gov can supply that, but so can Current Controlled Trials out of London.
In anticipation of future controversies over what pharma chooses to publish, the industry’s lobbying arm has set up its own Potemkin village trial registry to which companies besides Lilly and Glaxo can funnel a few scraps of data. That site will be useless to medical editors and the public. But in an era in which appearances matter, it could be just enough to deflect the next round of criticism. l