Why We Need Full Disclosure of Drug Trials

The latest news that the type 2 diabetes drug Avandia increased the risk of cardiovascular disease (heart attack, stroke) came into the public domain when researchers at the esteemed Cleveland Clinic published its findings in The New England Journal of Medicine last week.

Somewhat less publicized was the fact that the investigation occurred largely because of an unusual settlement the manufacturer agreed to in 2004 with the former State of New York Attorney General Eliot Spitzer (now the Governor) over the antidepressant drug Paxil. In that settlement, GlaxoSmithKline agreed to publicly disclose the results of its clinical trials for all of its other drugs, including Avandia. It was because of that public disclosure that investigators discovered information that was not widely disseminated.

Apparently, a cardiologist at the Cleveland Clinic, Dr. Steven Nissen, stumbled onto the GlaxoSmithKline web site while researching Avandia last April. He and a colleague quickly analyzed the data, and last week, The New England Journal of Medicine released its finding that Avandia posed a heightened cardiac risk.

"It was a treasure trove," Dr. Nissen said about the GlaxoSmithKline web site.

As our compatriots at Close Concerns (note: you may need to hit the "refresh" button on your browser for access) recently wrote:

"Companies are growing increasingly savvy about conducting clinical trials and designing a publications strategy for their products, as evidence-based medicine has become paramount for both medical professionals and coverage."

I agree completely, but the dark side to this business-as-usual story is that pharmaceutical companies have also become increasingly savvy about positioning their products in the most favorable light for the FDA reviewers, often omitting information that might cause reviewers to investigate further. This also means that doctors don't always receive the full picture about a drug's risks and benefits because they tend to hear or read about only those trials in which the medication shows a benefit.

I call this "selective disclosure of the truth," its not exactly falsifying new drug applications, but its nevertheless questionable when it comes to ethics. However, as long as regulators at the FDA do not require full disclosure, then our regulators are the guilty parties, not the executives running these companies. After all, the manufacturers have a fiduciary responsibility to their shareholders, but regulators are supposed to be representing the public interest.

Many of my readers have type 1 diabetes, therefore news about another diabetes pill generates kind of a yawn, but its worth noting that some convincing arguments have been made about synthetic insulin -- in fact, an entire book (published in 1987) about it called "The Galenics of Insulin" by Jens Brange, the former Director of Diabetes Pharmaceutical Research for Novo Industri A/S (now Novo Nordisk A/S), who is now on the Board of Scientific Advisors for SmartCells, Inc. -- highlights exactly how much variation can occur in the potency of insulin made via rDNA processes. It can be managed effectively, but we are taking a blind leap of faith in the manufacturers to govern their manufacturing processes themselves.

Although the FDA does have "Good Manufacturing Practices" outlined, these apply mainly to chemical drugs which are governed by the Federal Food, Drug & Cosmetic Act. Because insulin was one of the first rDNA manufactured medicines on the market, that act also governs insulin manufacture. More stringent requirements for vaccines and most other biotech drugs are outlined separately under the Public Health Services Act. But as the Avandia incident highlights, insulin users are not immune from the lack of FDA monitoring of medicines after they are approved.

As one of my readers noted on my previous post recently wrote:

FDA biologics overseers should set up standards and guidelines for batch testing these (all) insulin products since their manufacturing technology is much more critical in end-product identity/replication than in, for instance, products labeled "aspirin" that roll off a conveyor belt. Currently there are no batch testing requirements for insulin products. (emphasis mine)

rDNA insulin was the first genetically-engineered product to gain FDA approval. Because it was NEW, there were no guidelines in place to address issues that continue to present themselves. By remaining quiet, and allowing the belief that insulin is tightly scrutinized under biologic guidelines, the insulin manufacturers have escaped compliance with more stringent regulations that have been enacted SINCE its introduction.

The net effect is that we need the brand-name drug manufacturers to live by the same standards they want to hold generics manufacturers up to. As for the Avandia issue, the following article from The New York Times calls attention to several issues, including exactly what full disclosure could mean for the pharmaceutical industry (which explains why they fight it), as well as issues the FDA has in trying to regulate the pharmaceutical industry. The real issue is that the expectations for the FDA are huge, yet the agency's source of funding comes largely from the user fees (see the following articles from The Boston Globe and The Washington Post for details), so we have a few conflicts of interest that need to be resolved. However, as we have seen in the case of Avandia, full disclosure has benefits to society as a whole, now we need the regulators at the FDA to mandate it.


For Drug Makers, a Downside to Full Disclosure
By Barry Meier, The New York Times
May 23, 2007

When GlaxoSmithKline settled a lawsuit three years ago with the State of New York over the antidepressant medication Paxil, the company agreed to take an unusual step: publicly disclosing the results of its clinical trials for Paxil and other drugs.

The company, which was criticized at the time for failing to publicize all pediatric trials of Paxil, not just the positive ones, made good on its promise. The first posting on a new Web site was about 65 studies involving its popular diabetes drug, Avandia.

This week, GlaxoSmithKline learned what that greater disclosure could mean.

A cardiologist at the Cleveland Clinic, Dr. Steven Nissen, stumbled onto the Glaxo Web site while researching Avandia last April. He and a colleague quickly analyzed the data, and on Monday, The New England Journal of Medicine released its finding that Avandia posed a heightened cardiac risk.

"It was a treasure trove," Dr. Nissen said about the Web site.

GlaxoSmithKline has disputed the journal’s interpretation. Officials with the Food and Drug Administration said they were reviewing whether to take any action on Avandia.

Whatever the drug's fate, the episode is likely to fuel efforts by some medical experts, including Dr. Nissen, to persuade lawmakers to require makers of drugs and medical devices to disclose study results publicly. Currently, producers are not required to do so, but Congress is considering legislating a requirement.

Many companies besides GlaxoSmithKline already post results from some studies or trials on their Web sites, or one operated by the Pharmaceutical Research and Manufacturers Association, a trade group in Washington.

Dr. Bruce M. Psaty, a cardiologist at the University of Washington, said that having such information can play a critical role, as the case of Avandia suggests, in spotting signals of a drug's possible dangers.

Other experts have argued that the relative efficacy or cost of competing drugs can be compared only when all study results, rather simply those that a company chooses to publicize, are available.

Studies have found that the vast majority of drug and medical device studies are never published in medical journals.

"The more information, the better," Dr. Psaty said.

Dr. Ronald L. Krall, chief medical officer for Glaxo, said his company sharply disputed the methodology of Dr. Nissen’s study, and a top F.D.A. official said that the agency had previously informed doctors about Avandia’s heart risks.

Dr. Krall said his company was aware when it created its database of study results a few years ago that it might lead to controversy. Other scientists might look at its data or choose to analyze it differently than company officials did, he said.

"We are committed to the principle of transparency," Dr. Krall added. "But we knew that when starting this, by putting the data in the public, many things could happen, some of which could be trouble."

Some experts also believe that releasing the results of hundreds of studies involving drugs or medical devices might create confusion and anxiety for patients who are typically not well prepared to understand the studies or to put them in context.

"I would be very concerned about wholesale posting of thousands of clinical trials leading to mass confusion," said Dr. Steven Galson, the director for the Center for Drug Evaluation and Research at the F.D.A.

Roughly a decade ago, some experts raised concerns that doctors were not getting the full picture about a drug's risks and benefits because they tended to hear or read about only those trials in which the medication showed a benefit.

Companies and researchers typically did not seek publication of studies that showed that a drug had little benefit or might even cause harm. In some cases, trials that were started and stopped before completion were not disclosed.

As a result, outside researchers could not learn what trials of a drug had been performed so they could put findings in context or compare studies of competing drugs.

That issue caught the public’s attention after it was disclosed that Glaxo had not publicized trials of Paxil in children in which the drug showed little, if any, benefit. The company was subsequently sued by Eliot Spitzer, who was then the attorney general and who is now the governor of New York. The drug maker, as part of the lawsuit's settlement, created a public Web site for trial results. Glaxo was by no means the only drug company that came under scrutiny. In late 2004, a group of leading medical journals, including The New England Journal of Medicine, said that they would no longer publish articles about study results unless producers publicly registered the tests on Web sites like ClinicalTrials.gov, which is run by the National Library of Medicine.

As a result, the number of drug trials registered on that site has sharply increased, said Dr. Deborah Zarin, its director. (Currently, drug manufacturers are required to register trials of new drugs for serious or life-threatening conditions).

But even before the recent Avandia episode, advocates for greater study transparency like Dr. Nissen were pushing lawmakers to take the next step by requiring that producers of drugs and makers of devices not only register trials but also publicly disclose study findings.

"It is critical, but this raises the question of how many other drug safety issues are out there," Dr. Nissen said. Recently, the Senate passed an F.D.A.-related bill that would set up a process for developing a mechanism that experts expect would result in a government-run database where companies and others would post the results of clinical trials. The House is currently considering a bill that has somewhat different provisions.

Dr. Alan Goldhammer, a senior executive at the Pharmaceutical Research and Manufacturers Association, said the organization supported the disclosure provision in the Senate bill that had passed.

He said the group, however, was concerned that some states may be trying to get ahead of the federal government on the issue; for instance, Maine recently passed a bill that mandates the release of study findings.

"We want to make sure it is done in a reasonable way," Dr. Goldhammer said.

Recently, a report issued by the Institute of Medicine, a part of the National Academy of Sciences, recommended that the F.D.A. release all summaries of study data it had collected in the process of approving new drugs as well as all post-marketing studies of those products.

The F.D.A. rejected the first recommendation as overly burdensome and Dr. Galson, the director of the F.D.A.'s drug evaluation and research, said that the agency already released much of this information. "It is not that we are philosophically opposed to it, but the work would be enormous," he said.

Even those supporting mandatory results disclosure acknowledge that finding uniform ways to disclose complex scientific information would prove difficult and time-consuming. For example, Dr. Zarin of ClinicalTrials.gov said that reviewing a study's results to make sure that it was free of any biases interjected by researchers involved in a study or by its sponsor was a major undertaking.

Then, there is also the question of who the audience for such information should be — scientists, consumers or both?

Dr. Zarin said that there had been significant discussion among experts over the last year about that issue. Most have agreed that data is best understood by experts, a view that might not prove popular with patients.

Dr. Krall of Glaxo agreed, saying the drug maker had considered providing summaries of its studies for patients, but then dropped the efforts after deciding it would require making subjective decisions about trial results.

"There is not a uniform view about how to interpret results," he said. "It is quite problematic to go that next step."

Stephanie Saul contributed reporting.

URL for this article:
http://www.nytimes.com/2007/05/23/business/23drug.html