In 2007, an FDA advisory panel voted 22 to 1 to keep GlaxoSmithKline’s diabetes drug, Avandia (Rosiglitazone) on the market–albeit with a black box warning label. This week, that panel (now constituting 33 members) considered Avandia again, but voted differently.

Seven panelists wanted to add stiffer language to the current black box warning, and 10 wanted both stiffer language and restrictions on its use as a drug of last resort. A plurality, 12 members recommended pulling Avandia off the market altogether. Only three members supported leaving it on the market with its current labeling, and one panelist abstained.

Press headlines have been all over the place as the spin doctors try to figure out what it all means, but The New York Times nailed it:

"The vote is an enormous blow to Avandia and GlaxoSmithKline. The vast majority of panel members voted either to withdraw the drug or to allow continued sales only if strict controls are added. As the panel members explain their votes, the news just gets worse for GlaxoSmithKline. Of the first eight panel members who have spoken, nearly all who voted to keep the drug on the market said they nearly voted for the worse option."

As for GSK’s response…
Today, GSK suspended the Avandia TIDE trial in India after the Indian government requested a hold due to recent studies raising safety concerns.

Whatever the FDA and GSK may decide, the former blockbuster has a dim future as a therapy of last resort–if it remains on the market at all.  So, what changed between 2007 and 2010? 

There is little new scientific evidence per se–there is compelling confirmatory evidence showing that Avandia increased the risk of heart attacks by a third (Dr. Steven Nissen) and increase the risk of stroke by 27% (Dr. David Graham). And there were no big changes in the composition of the FDA panel.  But the drug review process under FDA’s new Commissioner, Dr. Margaret Hamburg, is undergoing a paradigm shift.

Numerous panelists were highly critical of the unsubstantial scientific evidence in support of Avandia’s approval in the first place.  When shown that company sponsored trials were poorly conducted, manipulated by GSK, and incompletely reported, they found the evidence of increased cardiovascular risk from independent meta-analyses and observational (epidemiological) studies (which have often been maligned in the past) more credible.

 
At this hearing FDA’s drug flawed approval process came under criticism from several panelists as it was apparent that Avandia’s approval (in 1999) had been made in the absence of scientific evidence of a positive risk / benefit. Instead, and as has been the norm, FDA’s approval was based on a surrogate endpoint, with insufficient scientific evidence of clinical benefit to support it. 

FDA’s modus operandi on safety reviews has been an orchestrated process by the same people who approved the drugs in the first place, the Office of New Drugs (OND).  OND has sought to suppress dissenting views emanating from the Office of Surveillance and Epidemiology (OSE).  Recall the antidepressant debacle:  When the FDA epidemiologist from OSE Dr. Andrew Mosholder, did a meta-analysis showing that antidepressants increased the risk of suicide, the all-powerful OND embargoed his research report from an advisory committee deliberating about the suicide risk in antidepressants. 

In sharp contrast, under Dr. Hamburg’s leadership, the Avandia safety review process –which was broadcast by CNN live–was notable for its dramatic democratic shift: the advisory panel heard not only from the company and the agency’s OND defenders of the drug, but critical analyses and divergent points of view from OSE and reviewers were presented to the panel.  For the first time, the internal split within the FDA between OSE and OND was laid bare for all to see–and the impact was palpable.

Panelists heard credible presentations puncturing GSK’s claims about the drug’s safety and benefit by independent scientists–including senior FDA officers, Dr. Thomas Marciniak and Dr. David Graham–as well Dr. Steven Nissen, whose 2007 analysis led to the 2007 advisory hearing. These scientists’ separate analyses showed that Avandia increased the risk of heart attacks without a demonstrable clinical benefit compared to placebo and compared to other available drugs.

 Like Vioxx, Avandia’s market success–and the success of numerous other hazardous blockbuster drugs that gained FDA approval without adequate scientific data, was shaped by aggressive marketing campaigns and corporate deception and manipulation of data. The panel’s vote tells the world that under the weight of open scientific scrutiny, Avandia’s success was a house of cards that has collapsed.  While GSK executives were applauding their blockbuster sales, patients were unaware of the risks they were being subjected to in pursuit of corporate profit.

FDA’s approval of new drugs is based on meager evidence of benefit. FDA’s dismissal of scientific methods such as observational research and meta-analyses required to show adverse clinical effects, shields industry but is untenable–as it ignores an important body of evidence. FDA officials, who lowered the standard for ensuring that only safe and effective drugs are approved, and who turned a blind eye to company concealment of serious risks, bear major responsibility for the harm produced by Vioxx, Avandia, Paxil, Zyprexa, Chantix, and too many others.  

Below, Gardiner Harris of The New York Times provides the flavor of the (sometimes dramatic) two-day proceedings.  His report in today’s Times contrasts with that of the rest of the press.

 Press coverage of the hearing shows that reporters mostly don’t grasp the sea change that the open review process brought about.  Most press reports stated that GlaxoSmithKline won and that Avandia lives after the devastating vote of no confidence in its safety or benefit: Wall Street Journal; MSNBC
  
But what does GSK win and how does Avandia live?

On the eve of the FDA hearing, GSK settled Avandia and Paxil lawsuits for $2.4 billion. Reuters reports that this will wipe out most of its earnings in the fiscal quarter ending June. Will the company, whose credibility has suffered irrevocably—first when the New York State Attorney General charged the company with fraud involving its (then) blockbuster antidepressant, Paxil—and now by the revelations that its Avandia studies were “inappropriate and biased,” and that the company concealed cardiac events rendering its ACCORD study not credible.

Twenty-nine of thirty-three committee members found the drug to increase the risk of heart attacks, calling for highest restricted marketing or withdrawal.  The majority of the committee indicated that the drug could be marketed only if the company completed the TIDE clinical trial designed to compare Avandia to another drug, Actos, to prove it is safe. 

In the 1990s, GSK had decided against conducting such a study fearing the results might hurt sales.  As Dr. Steven Nissen pointed out, “the drug has been on the market for 11 years. The company had every opportunity to do large-outcome trials, adequately powered and properly run, to answer this question. They didn’t do it." 

It is a matter of controversy whether–in light of the overwhelming consensus that Avandia increases the risk of heart attacks compared to Actos– the TIDE trial is ethical.

Indeed, the company was forced to suspend its critical TIDE study in India because the Indian government raised safety concerns for the subjects who would be exposed to Avandia. Might not other study sites follow suit? Who would knowingly volunteer for a study that tested a drug shown to increase the risk of heart attacks by about 30%?

How likely is it that the company will continue to market Avandia as a drug of last resort—where’s the profit?

But more to the point, what responsible physician adhering to the precautionary principle of “first do no harm” would prescribe it, knowing the extent of criticism against among the FDA advisory panel members?

Vera Hassner Sharav

The New York Times
July 14, 2010, 8:59 am
F.D.A. Panel Votes to Restrict Avandia
By GARDINER HARRIS

A majority of the advisory panel of the Food and Drug Administration voted today to restrict the sales of Avandia, a controversial diabetes drug, because of its potential risk for causing heart attacks. The 33-member advisory committee was deeply divided. Twelve voted to remove Avandia from the market altogether; 10 for continued sale but with new label revisions and possible restrictions; 7 to add more warnings and 3 for no change at all. But the votes can also be viewed as a decision by a majority, 21, to continue allowing sales of Avandia, with more restrictions. A final decision will be made by the F.D.A. at a later date.

The committee voted 20-to-10 in recommending continuation of the so-called “Tide” trial, the one intending to compare Avandia and its rival Actos.

But this vote almost certainly reflects wishful thinking on the panel members’ part. The “Tide” trial was already having difficulty getting patients to agree to take part in the trial because of worries about Avandia’s risks.

Now that the panel has decided those risks are substantial enough to require that the drug either be withdrawn or restricted in sales, those enrollment problems will certainly worsen.

Dr. Marvin Konstam explained that he voted to withdraw the drug but then voted to continue the Tide trial because, if Avandia stays on the market, it needs to be studied. Still, he said that the trial might not be ethical.

“I voted yes, but I really don’t understand how you can do it,” he said.

This is the committee’s final vote.
4:32 p.m. |Panelists Reach Across the Divide

Dr. Marvin Konstam, a panel member, complimented the F.D.A. for holding the meeting, and he specifically cited Dr. Janet Woodcock, director of the agency’s drug center.

Dr. Woodcock decided in 2007 to keep Avandia on the market, and F.D.A. safety officers who have advocated for Avandia’s withdrawal blame her for resisting those efforts. But Dr. Konstam, who voted to withdraw Avandia, said that she had done a terrific job in organizing the advisory meeting.

“There’s been a lot of written in the last few days and weeks about the F.D.A. being broken. As someone who has worked with F.D.A. over the last 20 years, I don’t think it’s been at all broken,” Dr. Konstam said. And then to wide laughter in the audience, Dr. Konstam added: “I think some of the people within F.D.A. might work on getting along better.”

Dr. Woodcock nodded and smiled.
4:11 p.m. |Another Vote Coming Up on Research Trial

A final vote remains for the committee to decide whether to continue the “Tide” trial, which is intended to compare Avandia and Actos.

But given its decision to restrict how Avandia is sold, this trial will be all but impossible to conduct no matter what the committee votes.

Again, the votes have been a severe blow to Avandia and GlaxoSmithKline. The F.D.A. will make the final decision, but two-thirds of the panel members voted either to withdraw Avandia or severely restrict its sales.
3:37 p.m. |Panel Votes to Restrict Avandia

Twelve members voted to withdraw Avandia from the market, while 10 voted that it should continue to be sold but with serious revisions to its label as well as possible restrictions on its sale. Seven voted to simply add further warnings to the drug’s label. Three voted to allow further sales without change.

The vote is an enormous blow to Avandia and GlaxoSmithKline. The vast majority of panel members voted either to withdraw the drug or to allow continued sales only if strict controls are added.

As the panel members explain their votes, the news just gets worse for GlaxoSmithKline. Of the first eight panel members who have spoken, nearly all who voted to keep the drug on the market said they nearly voted for the worse option.

Even Dr. Sanjay Kaul, who voted to allow the drug to still be marketed with stronger warnings, said that the F.D.A. must make sure that Avandia is used far less regularly.

“Make sure this is available as second line and not as first line,” Dr. Kaul urged.

Dr. John Teerlink, a panel member, said he was on the advisory panel in 2007 that voted to continue to allow Avandia to be marketed. But he said that he made the earlier vote with the understanding that GlaxoSmithKline would use the added time to prove that its drug was safe.

“The burden of proof was on them to convincingly convince me that there was no risk here, and that didn’t occur,” Dr. Teerlink said.

Dr. David Capuzzi, a panel member, responded that he had been in the drug industry twice before and knew about drug development issues. He said that if Avandia has safety problems, then Actos, a very similar medicine, is likely to have them as well.

“I just don’t see that the evidence is there” to support a concern about Avandia’s safety, Dr. Capuzzi said.

A final decision will be made by the F.D.A., of course, at a later date.
3:33 p.m. |How Much Risk Is Too Much?

The committee is now in the midst of a discussion of just how risky a drug has to be before it gets withdrawn.

Dr. Robert Temple, the F.D.A.’s eminence gris, explained that most drugs have been withdrawn for rare but very serious side effects. And he explained that the analysis for whether to withdraw a drug should be the same as the analysis on whether to approve a drug. In such analyses, even for small hints at a significant safety problem will lead the agency not to approve a drug, he said.

These statements hint that the F.D.A. could be taking a fairly tough view of Avandia.

Dr. John Jenkins, director of the F.D.A.’s drug center, clarified Dr. Temple’s remarks. He has advocated for Avandia’s continued sale. And he pointed out that when the F.D.A. has removed drugs for cardiovascular risks, those risks have often been three to five times higher than placebos – risks far worse than any study involving Avandia has suggested.

Again, signs of a divided F.D.A.

The voting is under way …
3:17 p.m. |Drum Roll, Please

The committee, just before it votes, is beginning to gel around whether Avandia should be withdrawn. Most of the clinicians seem to want to keep the drug on the market; many of the academics and statisticians are suggesting they want it withdrawn. The vote is starting …
2:58 p.m. |Practitioners Vs. Academics

Two of the 33 panel members said they treat diabetes patients, and they emphasized the need to have as many treatment options as possible for their patients. The class of medicines that includes Avandia “in some patients have profound effects in controlling blood sugar,” said Dr. Abraham Thomas, a panel member.

Dr. David Capuzzi, another panel member who treats diabetics, made a strong argument that clinicians need as many choices of drugs as possible. He said there are not that many good pills for diabetes.

“I think it would be a disaster to remove a useful agent like that,” Dr. Capuzzi said. “I don’t care what the controlled clinical trials show. Well, I do but I don’t think it’s knock down complete.”

The divide between clinicians and academics often drives such votes on F.D.A. advisory committee meetings.

Clinicians tend to analyze these questions based on what they would want at hand when treating the next patient who comes into their office, and they can often think of scenarios in which even a dangerous drug might be useful in those situations.

But academics, particularly those who spend time analyzing population level data, tend to point out that when drugs are sold broadly, small risks can lead to hundreds if not thousands of injuries. These experts have less confidence that doctors will always make the right decisions, so they tend to be far less accepting of safety problems.
2:52 p.m. |Weighing Drug Safety

The committee has moved on to discussing the benefit-to-risk profile of Avandia, which is a preview of how the panel will vote on whether to remove Avandia from the market.

Anyone who views the relative risks of this drug as being greater than its benefits will likely vote to withdraw the drug.

The debate begins.
2:46 p.m. |Correcting the Death Risk Vote

Because of a mistaken vote, the chairman of the committee corrected the vote on the fifth question. Here is the corrected explanation for that vote:

The next question has asked committee members whether Avandia increases the risks of death relative to older medicines. Only one panel member voted that Avandia increases the risks of death relative to older medicines; 20 said that Avandia did not increase these mortality risks; and 12 said that they did not know.
2:41 p.m. |Why They’re Voting This Way

The committee members are now explaining their most recent vote, with most saying that study data have failed to prove that Avandia increases the risks of death relative to older agents as well as relative to Actos.

The division between the members tends to split on how much confidence they have in retrospective observational studies, which are studies that look at medical records to see how well patients fared while taking drugs in the normal course of medical care. Such trials can be tricky to interpret compared to well-controlled, placebo-controlled trials.

Those who voted against Avandia on this issue said that such trials can be important when determining whether a drug is safe, although they are less useful to determine whether a drug is effective.

Dr. Howard Mann, a panel member, said repeatedly that he was far more willing to be wrong when deciding that a drug is unsafe.

On at least two votes, panel members have admitted that they pushed the wrong buttons.

“I meant to vote for B but I actually voted for Pat Buchanan,” Dr. Michael Proschan joked to laughter among the crowd of more than 200, (in a reference to problems with the ballots in the 2000 presidential election).
2:31 p.m. |Is Avandia Deadlier than Actos?

On the next vote, it’s 7-12-14.

Asked whether Avandia raised the risks of death relative to Actos, seven panel members voted that Avandia was deadlier, 12 voted it was not and 14 voted that they did not know.
2:25 p.m. |Next Vote: Few See Increased Death Risk

On the latest vote, it’s 2-20-11, on the question of whether Avandia increases the risks of death relative to older medicines. Only two panel members voted that Avandia increases the risks of death relative to older medicines; 20 said that Avandia did not increase these mortality risks; and 11 said that they did not know.
2:23 p.m. |Short Break

The committee has taken a brief intermission so that a particular set of questions can be updated.

It has already voted overwhelmingly (18-6-9) that Avandia increases the risks of heart attack relative to older medicines and that it increases the risks of heart attack relative to Actos, a similar medicine. They will vote later on whether to keep the drug on the market.

Considering the number of panel members who have said that they want more information about how these drugs operate, the vote on whether to keep Avandia on the market is likely to be much more favorable to Avandia, since withdrawal will mean an almost total end to more information about how Avandia functions.
2:04 p.m. |Setting the Table on Votes

The committee is voting on six of nine different questions, including whether Avandia contributes to heart attacks and deaths, how it compares with older drugs and how it compares with Actos, and whether it should remain on the market and, if so, whether its sales should be limited.

The set of questions being asked of this advisory committee are uniquely complicated, and they are possible only because the F.D.A. has installed a new voting system that allows for multiple choices. In the past, the committees were forced to choose between two stark choices – to approve, say, or not approve the drug.

The medical experts have complained for years that such stark choices do not allow for the sort of nuanced thoughts that they can have when examining complex medical trials. The F.D.A. has responded in the past that its own decisions tend to be binary – to approve or not approve a drug; withdraw or not withdraw a drug – so nuance is less useful.

But the agency has begun to make more nuanced decisions as well, using new tools to allow sales that are far more limited than in the past.

This complex environment makes for lengthy and complicated voting sessions, and the Avandia advisory committee is scheduled to vote on at least six questions and to take all afternoon doing so. The most important of these questions – whether Avandia should remain on the market – will come at the end of the day.
2:00 p.m. |Gauging the Panel’s Leanings

The advisory committee voted in 2007 overwhelmingly that Avandia raised the risks of heart attacks, but then voted overwhelmingly that the drug should remain on the market.

And as the members explain their votes, it seems likely that a considerable number of those who voted that Avandia raised heart attack risks may vote to keep the drug on the market. Several of the members who voted against Avandia said that they would like more data on the issue. If Avandia were withdrawn, there would be no need for more evidence and more evidence would be difficult to obtain.
1:40 p.m. |The First Vote Tally Signals Concerns Over Risk

Eighteen panel members voted that they were concerned that Avandia could cause heart attacks; 6 said they were not concerned about this risk and 9 voted that they were not sure.
1:29 p.m. |One Vote Leans to Avandia

Rebecca Killion made clear that she was likely to support the continued marketing of Avandia. Ms. Killion is a patient advocate, and such advocates often argue that patients need as many options as possible.

“This drug is not for everybody, but they may not mean it’s not for anybody,” she said.
1:27 p.m. |Divisions Apparent

A brief discussion by the committee over the quality of the evidence gives a telling glimpse that this panel is likely to split sharply over the issue of whether to withdraw Avandia. And for many, their views on one research aspect, the Record study is likely to serve as a surrogate for the later votes.

Some committee members said that they were not particularly concerned about the problems with the Record study. Dr. John Teerlink said that the problems with the study were not terrible.

But Dr. Michael Proschan said the problems with the Record study “scares the hell out of me.”

“It was frightening to see how one-sided it was,” he said.
1:17 p.m. | Pitting One Against the Other

Dr. Lamont Weide, a panel member, said he was concerned that the committee was being asked to compare Avandia and Actos, which he said was unusual for the F.D.A.

“This is really predecent-setting,” he said. “That they want us to decide that one agent is better than another agent.”

Dr. Janet Woodcock, the director of the F.D.A.’s drug center, answered that Dr. Weide was right. “But we are asking you to talk about the relative safety issues of this drug relative to other drugs.”

And then Dr. Robert Temple, one of the veteran members of the F.D.A., said that the F.D.A. has long required that a drug be withdrawn if it’s nearly identical with another drug that doesn’t have such a serious safety problem. He recalled, for instance, that Seldane, an antihistamine, was withdrawn because it caused heart rhythm problems after a similar antihistamine without such problems came on the market.
1:14 p.m. |The Timetable for the Vote

Dr. Kenneth D. Burman, the committee’s chairman, started the afternoon’s sessions by reading over the questions on which the committee must vote at the end of its meeting.

But he emphasized that time was very limited and that the day’s discussion must end at 5 p.m., a point that led to a murmur of pleased agreement from the scrum of reporters covering the meeting.

F.D.A. advisory committee meetings are famous for extending well beyond their schedules, making for very long days.

“It’s already 1:07, making only 23 minutes for discussion,” Dr. Burman said, keeping strictly to the schedule.
12:27 p.m. | Point and Counterpoint

Just before the committee broke for lunch, there was a brief set of presentations pitting GlaxoSmithKline against one of its toughest critics at the F.D.A.

Dr. Murray Stewart, a vice president at GlaxoSmithKline, was given five minutes by the committee’s chairman to clarify certain facts about the company’s controversial clinical trial, Record. The trial has been criticized repeatedly by the F.D.A. and independent clinicians as having been poorly designed and conducted. But Dr. Stewart said long term trials involving diabetics were challenging to conduct because diabetics must at some point be given the option of insulin, which cannot be given in a blind or secret fashion.

And he defended the conduct of the trial as appropriate.

He was quickly followed by Dr. Thomas A. Marciniak of the F.D.A., who on Tuesday had picked apart aspects of the Record trial. And while Dr. Marciniak had been careful on Tuesday to avoid saying that the many mistakes he found in the Record study were the result of deliberate manipulation by GlaxoSmithKline, he cast aside that caution today.

“It’s not a supposition that GlaxoSmithKline intervened with the adjudication — it’s documented,” he said.

He maintained that doctors in the study correctly took note of problems that patients in the study experienced only to have those initial judgments later overruled or questioned by the company. He said sick patients were dropped from the study instead being allowed to tarnish Avandia.

“This is a great way to improve your mortality statistics, which everybody tells me can’t be biased,” he said.

And then he provided the committee with the F.D.A.’s criteria for withdrawing a drug for safety reasons.

Dr. Stewart raised his hand to respond, but the committee’s chairman cut him off.

“We will break for lunch,” said the chairman, Dr. Kenneth D. Burman, who is also chief of the endocrine section at Washington Hospital Center.
12:24 p.m. | Documents and Research

While we’re waiting for more action by the F.D.A. committee, we’ll point you to a set of documents released Tuesday by the Senate Finance Committee that chronicle discussions within GlaxoSmithKline about the controversies over research on the drug.
11:20 a.m. | Other Risks From Actos?

Dr. Elaine Morrato, a panel member and an assistant professor at the University of Colorado, asked about concerns that Actos might increase the risk of bladder cancer.

Dr. Mary H. Parks of the F.D.A. answered that animal trials had suggested there might be a cancer risk in the bladder. And she said that large trials of Actos did not dismiss that risk. But she said similar drugs in the same class have all suggested the drugs might pose a cancer risk.
11:18 a.m. | Debating Risks

Rebecca Killion, a panel member, asked whether Avandia might still be appropriate for patients with few cardiovascular risks.

Younger patients, for instance, are not nearly as susceptible to heart attacks as older ones. Her question led to a small debate between Dr. Steven Nissen, the Cleveland Clinic cardiologist who has advocated Avandia’s withdrawal, and Dr. Philip Home, who led the Avandia Record trial and has said that his research was financed by GlaxoSmithKline.

Dr. Nissen maintained that no diabetic is without risk for cardiovascular disease.

“Not only are diabetics getting younger, but diabetics with coronary disease are getting younger,” Dr. Nissen said. “To say that there’s some diabetic that’s not at risk, is probably not something we can parse.”

Dr. Home responded that he routinely made such decisions in his clinical practice. For instance, since Avandia and Actos increase the risks of fracture, he would not give either drug to patients with osteoporosis, he said. “We would certainly target these drugs to certain groups of people,” Dr. Home said.
11:14 a.m. | Listening for Clues

After a brief debate about whether the questions they will be asked to vote upon are appropriate, the committee’s members have moved into an hour of discussion and questions.

This is when we will begin to see how the committee is moving; these sorts of discussions can sway uncertain panelists.
10:57 a.m. | Deliberations Begin

After only a handful of speakers, the public hearing ended.

Dr. Gerald Dal Pan, director of the Office of Surveillance and Epidemiology at the F.D.A., is explaining to the committee what the advisers are supposed to do.

His message is a bit like the charge that a judge gives a jury before they consider a case. But the fact that Dr. Dal Pan is making this speech is interesting: He leads a group of review officials who have advocated forcefully for Avandia’s removal from the market. During the meeting, he has been sitting next to Drs. John Jenkins and Janet Woodcock, two F.D.A. officials who have defended Avandia’s continued sales.

The agency’s internal conflicts have been on stark display in this advisory committee meeting. But Dr. Dal Pan’s calm and even-handed presentation may be intended to signal that agency officials can work well together despite their differences.

Dr. Dal Pan assured the committee that its views were important to the agency.

“A transcript is made of these meetings, and we actually do go back and read them,” Dr. Dal Pan said. “We are very interested in the rationale for your vote.”
10:25 a.m. | Letter From Investigator

Jackie Bosch is reading a letter written by Dr. Salim Yusuf of McMaster University, who is a principal investigator of the Tide trial, the test comparing Avandia with Actos.

Public testimony is generally given directly, but this hearing has been unusual for several reasons.

In his letter, Dr. Yusuf bemoans the discussion in the news media and in medical literature about Avandia’s risks, saying it has been based on poor science. He argued that only trials like the one he is conducting can definitively answer questions about drug safety.
9:45 a.m. | Public Speakers Take the Stage

The public part of the advisory meeting has begun, and like so much about this meeting, it is unusual.

In many of these meetings, the public part includes a collection of unsophisticated patients and doctors who tell poignant and personal stories about their experiences with a drug. But the first several public speakers for this meeting came with PowerPoint slides and sophisticated arguments about the underlying trials.

Dr. Christopher McCoy, a hospitalist at the Mayo Clinic in Rochester, Minn., and a representative of the National Physicians Alliance, spoke about the effects that financial connections with drug makers can have on interpretations of studies. He was followed by Dr. Hal M. Roseman, who consults for GlaxoSmithKline, and amid a set of slides with pictures of the comedian David Letterman, offered a complicated set of slides defending Avandia’s continued sales.

Then came Diana Zuckerman, president of the National Research Center for Women and Families, whose slides emphasized Avandia’s dangers.

Many more speakers are coming.
9:25 a.m. | Drug Maker Under Fire

GlaxoSmithKline’s reputation has been battered throughout this advisory committee hearing, and it took another hit a moment ago.

Dr. William Knowler, a panel member who is chief of diabetes epidemiology at the National Institute of Diabetes and Digestive and Kidney Diseases, described a key analysis by GlaxoSmithKline as “totally incorrect and deceptive.”

Dr. Dean Follman, a mathematician at the National Institute of Allergy and Infectious Diseases, agreed that the company’s analysis was incorrect.

“I don’t know if I would term it as deceptive,” Dr. Follman said. But since the company failed to explain why it conducted its analysis so poorly, “I would ignore it, basically,” he said.
8:59 a.m. | Update

The Food and Drug Administration hearing on Avandia, the controversial diabetes drug, is under way. A complex set of votes is expected later today on whether to remove the drug from the market because of the risk for heart attacks.

Dr. Hertzel Gerstein, a professor at McMaster University who is in charge of an ongoing study comparing the safety of Avandia and Actos, started today’s hearing with a passionate defense of ethics of the study, which has been dubbed the “Tide trial.”

Nearly shouting into a microphone, Dr. Gerstein said that experts who called the trial unethical on Tuesday were mistaken.

“In the next 20 minutes, I hope to correct the misperceptions that were repeated yesterday and to show you that Tide is both appropriate and needed,” he said.

One presenter on Tuesday claimed that the Tide trial was largely being conducted in the Third World because doctors in the United States are not comfortable putting patients in the trial.

“Contrary to what was stated yesterday, this study is mainly being conducted in the developed Western world,” Dr. Gerstein said.

We’ll be updating this post throughout the hearing with the latest developments. See the  related article on the first day of hearings on Tuesday.