Pharmas move towards full disclosure

The 2005 Clinical Trial Congress explored what may be the most private area of any pharmaceutical company's anatomy: its unpublished clinical trials. Must the pharmaceutical industry reveal more? Or should it ride out the cries of journal editors who demand that it disclose the designs and results of trials? That question will roil government regulators, journal editors, and companies of all sizes this year.

As the meeting made clear, most companies now elect to publish clinical trial data only when those facts can be used in advertising and marketing. The vast majority of clinical trials are consigned to a dusty file cabinet or perhaps read by a regulatory official -- and then consigned to a dusty file cabinet.

In a panel on trial registries at the Clinical Trials Congress conference, it was clear some of the thought leaders in the industry do not fear more disclosure of unpublished trials per se. They're already undertaking the selective dissemination of trial designs both via their own websites and the federal government.

What does concern some of the leading companies is a proliferation of clinical trial registries around the world. To head that off, physicians from two major companies -- Eli Lilly and GlaxoSmithKline -- took the Philadelphia podium and explained their efforts to date. A third company doctor, from Merck, offered his sympathy but explained his employer's strategy, which is more synchronised with that of the industry as a whole.

Ronald Krall, senior VP of worldwide development at Glaxo said there are two intertwined crises of confidence -- one in the industry in general, and another for the clinical trial endeavour. In one poll, he noted, "60 percent of the people were not very or not at all sure that drug companies publish timely information on side effects."

Clearly, Krall said, confidence in the industry is in short supply at the moment, despite the precaution and thought that goes into clinical research. Glaxo has launched an online database of its clinical trials to begin to restore that confidence.

Taking such a route was much debated inside GSK, Krall suggested. But at the end of the day, the company believed there was an implicit ethical contract with patients. Those who elect to participate in clinical trials, Krall said, "expect that the results of the trial in which they're participating will be available and contribute to the knowledge that informs medical practice."

At GSK, he said, the company will publish most trials that meet nonbonding International Conference on Harmonisation (ICH) guidelines. The company will describe its finished Phase I-IV trials begun since 2000, though some trials prior to 2000 may also be included. Ongoing research is first deposited in ClinicalTrials.gov, an NIH effort with a tiny fraction of all industry-sponsored trials.

For tracking purposes, ClinicalTrials.gov will assign each trial a unique numerical identifier. The information to be published at the GSK website includes type of trial, rationale, methods, study design, patient population demographics, primary and secondary endpoints, adverse events, and references to published citations in the peer-reviewed medical literature. That's most of the laundry list of requests from editors of a handful of elite medical journals, who have made a so-far empty threat to block articles about trials that have not been put in a searchable public registry.

Needless to say, there is little agreement between what journal editors, trial sponsors, the ICH, and regulators believe should be included in an online registry of clinical trials. Much of the private bargaining seems to be turning on which database fields will need to be supplied by the industry.

For its part, GSK is choosing not to publish the results of a trial. It's unclear how the medical editors will respond to this. In some ways, a clinical trial registry without any results -- without stating if a drug worked or not -- is like a topless bar in which the entertainers remain clothed. There are other places (such as peer-reviewed journals and medical conferences) where trial results may be bared. But a no-results strategy could disappoint community and academic physicians attempting to advise their patients and move beyond the cursory information in either the scientific literature or the FDA-approved label.

Will Glaxo's approach be enough to silence the critics of the industry? It's not clear. Private discussions with top medical editors appear to be ongoing.

His real worry, Krall confessed, is the proliferation of different regulatory obligations around the world. As it is, there are several dozen people working on this area in GSK. "The cost as long as it's one register is pretty marginal," said Krall. "We really fear the proliferation of national registries. It could turn out to be a big burden on us."

GSK appears to have sufficient confidence in the primacy of its science to allow it to endure short-term pain that may accompany its registry.

Indeed, Krall has warned the company's high-level managers, lawyers, and communications specialists about a few bumps along the way. "As we populate registers, there is going to be information that wasn't available before," said Krall. "People are going to interpret it. We are going to take some hits for that, as individual companies and as an industry. We know it will happen."

That displayed a level of organisational bravery that, for now, only one company in the industry is willing to match. In another presentation at the Clinical Trials Congress, Eli Lilly's global medical director, Per Cantor, outlined that company's online registry. "What we are discussing today is only the first chapter," said Cantor.

At Lilly, Cantor said, there has been a philosophical shift about registering trials and publishing selective facts about them. "We had the philosophy of saying, 'Why disclose?' Is it needed that we share information that was traditionally considered confidential?" Cantor asked. "That has shifted to a consensus of, 'Why not disclose?' "

Rather than wait to be further pummelled by a sceptical public and a skittish regulatory community, Lilly decided to get out in front of a coming shift in the minimal level of disclosure about many routine trials. It too seems worried about registries multiplying like rabbits. "There are 300 registries right now," Cantor noted. "It's very important that the industry takes the lead."

Cantor's explanation of exactly what is published at Lilly's site was not as explicit as the Glaxo presentation. "We will post (trials) within one year after the completion of the trial," said Cantor. "That's what we have concluded is an appropriate measure. There may be situations where we cannot meet the one-year deadline."

Cantor urged other companies in the industry to begin determining their own comfort level with greater disclosure.

Merck, for example, spoke briefly in the same session in Philadelphia. Laurence Hirsch, Merck's executive director of medical communications, said Merck prefers to participate in a scientific dialogue with physicians using the peer-reviewed literature. "We do not have a Merck clinical trial registry," said Hirsch. "We don't have a trial results database on Merck.com. The company's thinking at this point is that it is not intending to go down that road."

Hirsch seemed saddened and wearied by the perception of both his company and the industry as being inclined to manipulate unflattering trial data. "We have published more negative trials in the last decade than anybody," said Hirsch. "Some of these are large and important trials. They're there."