Avandia tale shows weaknesses in regulatory system

[Update: A plurality (12) but not a majority of the 33-member FDA advisory panel voted to withdraw Avandia from the market. Most of the others voted to allow continued marketing of the drug, but with strict warnings and restrictions. Does this mean the FDA will keep the drug on the market or not? Unclear.]

A panel of Food and Drug Administration experts will decide today if Avandia, the controversial diabetes medicine made by the drug giant GlaxoSmithKline, should be withdrawn from the market due to concerns about its safety.

Avandia (known generically as rosiglitazone) has been linked to a 43 percent increase in relative risk of heart attacks among its users. That finding has cut sales of the drug (which was once the most popular diabetes medication in the world) by more than half — from $2.5 billion in 2006 to $1.2 billion last year, according to the Wall Street Journal.

How the mighty have fallen. On Tuesday, as the contentious FDA hearing was getting under way, GlaxoSmithKline announced that it had settled — for $460 million — a large lawsuit regarding Avandia. Plaintiff lawyers claimed that the drug had caused heart attacks and strokes.

Most observers are betting that the FDA panel will decide against Avandia today, despite the fierce internal battle within the agency regarding the drug’s safety and the trustworthiness of GlaxoSmithKline’s studies. (“It’s not surprising that they [some FDA staffers] don’t want to reverse their decision,” said an FDA physician-reviewer who criticized GlaxoSmithKline’s studies at the hearing Tuesday. “… If I approve a drug, my bosses say, ‘Good job,’ and my work is done. If I don’t approve a drug, it’s endless meetings.”)

Even if you are not among the 23 million Americans with type 2 diabetes, this FDA hearing — and the decision that comes out of it today — is important. As reporter and Pharmalot blogger Ed Silverman wrote Monday, the hearing “is about much more than any one drug or even the larger issue of drug safety — it’s also a referendum on the ability of a much-maligned institution to fulfill its mission as a public health agency.”

I’ll be following the hearing today via the intrepid Cardiobrief blog, and I’ll update this post as soon as a decision is announced. In the meantime, I thought I’d point out some of the seamier aspects of this story that have emerged this week. I wish I could say I was surprised by these revelations, but, sadly, I’m not.

From Monday’s New York Times:

The heart risks from Avandia first became public in May 2007, with a study from a cardiologist at the Cleveland Clinic who used data the company was forced by a lawsuit to post on its own Web site. In the ensuing months, GlaxoSmithKline officials conceded that they had known of the drug’s potential heart attack risks since at least 2005.

But the latest documents demonstrate that the company had data hinting at Avandia’s extensive heart problems almost as soon as the drug was introduced in 1999, and sought intensively to keep those risks from becoming public. In one document, the company sought to quantify the lost sales that would result if Avandia’s cardiovascular safety risk “intensifies.” The cost: $600 million from 2002 to 2004 alone, the document stated.

From Tuesday’s Wall Street Journal:

[In a deposition, a] former Food and Drug Administration official said the maker of diabetes drug Avandia withheld from regulators information suggesting the drug posed an increased risk for serious heart problems, according to people familiar with her statements. … The deposition was given by Dr. Rosemary Johann-Liang, a former manager in FDA’s drug-safety office. … In a 2009 interview with The Wall Street Journal, Dr. Johann-Liang said her FDA team had reviewed early reports from company researchers showing its scientists were concerned about heart-attack risks with their drug. She said she left the FDA after she was unable to convince her superiors about the need for stronger warnings.

[In fact, as Pharmalot’s Ed Silverman reports, Johann-Liang was excluded from the FDA’s safety meetings regarding Avandia. When asked in her deposition if she had felt vindicated when the FDA finally put a black box warning on Avandia’s label in 2007 — years after she had sounded the alarm — Johann-Liang responded, “I did not feel vindicated … Between the time it could have expeditiously gone to label, I believe that many people could have been — could be hurt by this adverse event that possibly would not have been hurt if they were alerted to this problem earlier.”]

From Tuesday’s hearing (via MedPage Today):

[FDA staff reviewer Dr. David Graham] slammed [GlaxoSmithKline’s] TIDE trial, an ongoing head-to-head postmarketing outcomes trial comparing rosiglitazone with pioglitazone.

Graham explained that TIDE — ordered by the FDA — was always meant to explore whether rosiglitazone was more harmful than pioglitazone and that enrolling patients in a trial designed to detect harm is unethical.

“When you go into a trial, you expect to get something good out of it,” Graham said. “[In the case of TIDE] the best [patient volunteers] can hope for is to not get a drug that causes a problem.”

The company requires consent forms for TIDE participants, but Graham said the form is misleading. After explaining that the trial will compare adding a thiazolidinedione (either rosiglitazone to pioglitazone) to placebo, the consent form mentions that vitamin D’s effect on fighting cancer would also be examined. The consent form goes on to mention vitamin D 18 times, which Graham said was meant to mislead patients.

“It was probably intended to mask the bad deal nature of the study,” Graham said.

Conflicts of interest abound in the Avandia story. Here’s one example from a Tuesday entry in Silverman’s Pharmalot blog:

Steve Haffner played a relatively modest role in the Avandia scandal, but until now, he was known primarily for leaking to GlaxoSmithKline the damaging meta-analysis of the diabetes pill that was published in The New England Journal of Medicine three years ago. Haffner later acknowledged breaching his responsibility as a peer reviewer, but blamed “bad judgment” because he wasn’t “feeling well.” In doing so, Glaxo was able to respond very quickly to the disclosure.

As it happens, Haffner felt just fine several years earlier when he allegedly had an Avandia study ghostwritten for him. Documents released today by the US Senate Finance Committee appear to indicate Haffner, a University of Texas Health Science Center at San Antonio researcher, was to be the author of a paper that was targeted at the American Journal of Cardiology, according to an internal email from 2001. … “Please find attached the Haffner manuscript,” David Harrison, the Avandia publication strategy manager writes. “The manuscript is currently in rough form and has not been sent to author yet.”

For the record: GlaxoSmithKline denies it ghosted the article (which appeared in the journal Circulation) and insists that it was authored by Haffner and reflects his “substantial input and his views.” (As Silverman notes: “How is ‘substantial input’ defined?”)

GlaxoSmithKline also claims that any documents that suggest it tried to hide negative study data were “cherry picked” from 14 million pages of documents that the company has released to plaintiffs’ lawyers — and, thus, present “an incomplete and misleading picture of the safety of Avandia and company conduct.”

For a summary of the first day of the FDA hearing, I recommend Emily Walker’s article at MedPage Today, as well as those in the New York Times and the Washington Post.