Asthma drug boomed despite risks and red flags

John Fauber / Milwaukee Journal Sentinel /

Published Nov 29, 2012 at 04:00AM

MILWAUKEE — The 2001 product launch of the asthma drug Advair was a lavish, Las Vegas affair.

Using images of a slot machine, top GlaxoSmithKline executives took the stage in front of thousands of assembled sales reps. “There are people in this room who are going to make an ungodly sum of money selling Advair,” one executive told the group.

He was right.

Millions of people with asthma, including many children, have gone on the drug and Advair sales have exceeded $4 billion every year since 2007.

Yet medical studies, independent doctors and court records indicate the drug can be dangerous, especially to children, and has been massively overused and often inappropriately prescribed.

Advair and other drugs that contain long-acting beta-agonists have been linked to 1,900 asthma deaths from 2004 through 2011, according to an estimate by AdverseEvents Inc., a private firm that analyzes incidents reported to the U.S. Food and Drug Administration. The incident reporting system is voluntary, so the estimate likely is low.

In a different analysis in 2008, an FDA researcher estimated the drugs contributed to 14,000 asthma deaths from 1994 through 2007. Advair, approved in 2000, is by far the biggest seller.

A Journal Sentinel/ MedPage Today investigation found the growth in Advair sales followed new asthma treatment recommendations that were written largely by doctors who received money from GlaxoSmithKline and other companies that market the drugs.

And the FDA sidestepped the concerns of some of its own doctors, who warned of the drugs' risks, especially among children and African-Americans, the investigation found. Those FDA doctors had urged that Advair and similar products not be prescribed to children and that other long-acting beta-agonists be taken off the market.

The picture of how a potentially dangerous drug became commonly prescribed comes from a Journal Sentinel/MedPage Today review of transcripts of FDA hearings, financial disclosures, medical journal articles and a U.S. Department of Justice complaint against GlaxoSmithKline.

Asthma, which affects an estimated 25 million Americans, is caused by inflammation and swelling of the airways in the lungs as well as constriction of those airways. Long-acting beta-agonists, which treat airway constriction, work by relieving routine, day-to-day symptoms. Inhaled corticosteroids treat inflammation.

Advair was considered a breakthrough in convenience because it combined the GlaxoSmithKline beta-agonist, Serevent, with an inhaled steroid.

Advair and similar drugs are considered appropriate treatment for more severe cases of asthma where the condition can't be controlled with just a steroid. But a 2010 study by Medco Health Solutions found that nearly two-thirds of mild asthma patients were taking the combination drug — often without trying a steroid first.

One theory from researchers for why Advair and similar drugs bring increased risk of deadly attacks is that the long-acting beta-agonists can mask worsening inflammation in the lungs. Patients feel better in the short term, but it may set them up for a catastrophic asthma attack down the road.

At a 2008 FDA advisory panel meeting reviewing the use of the drugs, agency physician David Graham questioned whether it was worth putting an entire population on the drugs just to find that occasional patient who gets better control of his or her asthma, according to a transcript of the meeting.

“We have no way of identifying who's going to have that miraculous response to Advair,” he said, “just as we have no way or predicting who's going to end up in the cemetery because of Advair.”

In 2010, when the agency approved more large-scale safety trials of the drugs, some of its own researchers called doing so unethical because it could put tens of thousands of patients at risk, according to a transcript.

Such trials would be especially dangerous for children, agency physician Andrew Mosholder warned, noting children “can't give consent themselves and one could argue the ethical burden is greater to protect.”

The trials, which are not expected to yield results until 2017, will involve 46,800 adults and adolescents aged 12 and older. A separate trial involving 6,200 4- to 11-year-olds also is being conducted.

FDA spokeswoman Morgan Liscinsky noted in an emailed statement that in 2010, the agency instituted new safety requirements for the drugs, including that they should be used for only the shortest duration necessary to achieve control of symptoms and then discontinued if possible.

In addition, the agency has said long-acting beta-agonists should only be used with a steroid. The agency also issued warnings in 2006 about the increased risk of deadly attacks.

Liscinsky said the FDA believes the benefits of the drugs outweigh the risks when they are used properly.

GlaxoSmithKline spokeswoman Karen Collins said the company stands by the safety and its promotion of Advair. She said the remarks from the Las Vegas meeting — described in federal court documents — were taken out of context, and do not reflect “the company we are today.”

“It is absolutely against GSK's policies and practices to inappropriately influence prescribing decisions,” she said.

Criminal allegations

In July, GlaxoSmithKline agreed to pay a record $3 billion to settle criminal and civil allegations filed by the U.S. Department of Justice that it unlawfully promoted several drugs, including Advair.

The government alleged that the company fraudulently pushed Advair as a first-line therapy for mild asthma patients, though it was not approved or medically appropriate for such cases.

In addition, the company paid kickbacks to doctors to induce them to prescribe and recommend Advair and other drugs, according to the complaint. The payments included speaking and consulting fees; gifts, travel and entertainment costs; payments for serving on “sham advisory boards;” and continuing medical education programs.

Collins, of GlaxoSmithKline, said the company disputes the government's allegations and said there was not a corporate strategy to promote Advair for inappropriate patients.

“The company reached this settlement with the government to avoid the delay, expense, inconvenience and uncertainty of protracted litigation of the government's claims and to put behind us these long-standing investigations of what was, for the most part, very old conduct,” she said.

The financial links between the asthma drug companies, the medical associations that promoted the drugs and the doctors who wrote treatment guidelines that recommended the drugs followed a pattern identified with other drugs in earlier investigations by the Journal Sentinel/MedPage Today.

In 2007, the National Heart, Lung and Blood Institute issued new guidelines for treating asthma that included this recommendation: Long-acting beta-agonists should be the preferred add-on treatment to combine with inhaled steroids, both in adults and children aged 12 and older.

Of the 18 members of the panel that wrote the guidelines, 15 had financial ties at the time to GlaxoSmithKline or other companies that market beta-agonists, according to research by the Journal Sentinel/MedPage Today.

The chairman of the panel, William Busse, is a doctor and professor at the University of Wisconsin School of Medicine and Public Health, whose financial relationships with asthma drug makers include years of work as a paid adviser, speaker and consultant.

At the time the guidelines were issued, Busse disclosed that he worked as a speaker and adviser for GlaxoSmithKline and several other drug companies, though specific amounts of money were not listed.

Another UW professor who served on the panel, physician Robert Lemanske, also had financial ties to the companies. His financial conflicts required that he get a special waiver from the FDA so that he could speak at its 2008 hearing on the drugs.

In emails, both Busse and Lemanske said they followed the panel's rules for transparency and for managing conflicts of interest. Precisely how much they and the other panel members were paid at the time could not be established. Only recently, in the face of new rules passed by Congress, have some drug companies publicly listed their payments to doctors and academics.

From 2009 through 2011, companies that make long-acting beta-agonists paid more than $400,000 to nine doctors on the panel, according to a database operated by the watchdog organization ProPublica.

Conflicts of interest

The Milwaukee-based American Academy of Allergy, Asthma & Immunology, which serves doctors and patients, has played an advocacy role for the drugs. Since 2008, companies that market long-acting beta-agonists have provided $4.7 million to the group. About $2.6 million of that came from GlaxoSmithKline.

An example: When the FDA held a hearing on the drugs in 2008, Stanley Szefler spoke on behalf of the academy, testifying it would be a disaster if the drugs were taken off the market.

“I don't think you want to walk away from this meeting and have a headline that this panel decided to draw back asthma to 20 years ago,” Szefler, a doctor with the National Jewish Medical and Research Center in Denver, told the panel members.

Szefler, who did not reply to a request for comment, also was a member of the panel that issued the asthma treatment guidelines in 2007. At the time he listed financial relationships as a consultant and adviser to GlaxoSmithKline and several other asthma drug companies. The amount was not required to be listed.

Academy President Wesley Burks said the group has been active in disseminating balanced information about the drugs. On its website, he said, it clearly mentions the risks and proper use of the drugs.

“As with most medications,” Burks wrote in an email, “there are risks and benefits which must be carefully weighed in every situation.”

Eric Campbell, an associate professor at Harvard Medical School, said there is an abundant supply of experts without financial conflicts. He said doctors who have worked as speakers for drug companies should not be allowed to serve on treatment guideline panels.

“Physicians should not be part-time drug salesmen,” he said.

An elusive number

Putting a reliable number on asthma deaths and severe attacks caused by Advair and other long-acting beta-agonist drugs has evaded researchers and regulators, in part because overall asthma deaths have been declining since the late 1990s to about 3,300 a year.

Estimates of deaths related to long-acting beta-agonists have varied widely.

In addition to the estimate by AdverseEvents Inc. of nearly 1,900 deaths between 2004 and 2011, long-acting beta-agonist products were the primary suspect in 3,500 hospitalizations in that period, according to the firm. Advair was the drug used in a vast majority of the cases.

The firm uses its own software to analyze reports of drug side effects to the FDA's Adverse Event Reporting System. Firm President Brian Overstreet said the FDA's voluntary system is known for capturing as few as 10 percent of such cases.

The estimate of 14,000 excess asthma deaths between 1994 and 2007 came from Graham, the FDA researcher who warned that Advair might be putting people in the cemetery.

That number is based partly on estimates using findings from two large clinical trials, both of which looked at Serevent, GlaxoSmithKline's beta-agonist drug, as a single agent.

Those estimates were disputed by Collins, the Glaxo-SmithKline spokeswoman, who said there were no such deaths in other research involving 18,000 Advair patients.

But the 18,000 number comes from 86 separate clinical trials sponsored by the company.

The fact that so many separate, smaller trials were involved can make a critical difference, especially when looking for unusual occurrences such as asthma deaths, said Sanjay Kaul, a physician who has served on FDA advisory panels and is considered an expert on drug safety data and clinical trial design.

In large clinical trials, patients are typically divided into groups with similar risk factors such as age, the severity of their disease and the other drugs they are taking — all of which can affect mortality. That can vary significantly across so many trials.

“Zero deaths reported from 18,000 patients in 86 different trials doesn't supply any evidence of reassurance,” said Kaul, a Los Angeles physician who also serves as director of the vascular physiology and thrombosis research laboratory at Cedars-Sinai Heart Institute.

1990s trial provided early warning signs for asthma drugs

MILWAUKEE — The troubled history of asthma drugs known as long-acting beta-agonists goes back to the 1990s when a large clinical trial in Great Britain of the GlaxoSmithKline drug Serevent produced disturbing results.

In that 16-week trial of 25,000 asthma patients, 12 people getting the drug died of asthma and respiratory causes, compared with only two in the control group who took a standard medication, according to a 1993 paper.

In addition, there were 193 hospitalizations or life-threatening events related to asthma and respiratory causes in the users of the new drug, compared with 102 in the other group.

The serious concerns raised in the first trial of Serevent, also known as salmeterol, led to a second large study of the drug several years later in the United States.

That trial was halted by GlaxoSmithKline, in part because of excess deaths among African-Americans.

Researchers had hoped to study 60,000 people in the trial, but only 26,000 people had been enrolled when it was stopped in 2003.

A 2006 paper that detailed the results of the U.S. trial noted there were 37 asthma- and respiratory-related deaths in those receiving salmeterol, compared with 14 among those on usual asthma medication who also got a placebo.

Among African-Americans getting salmeterol, 15 died of asthma and respiratory causes during the 28-week study period, compared with only three who got a placebo.

It is not known why African-Americans may be more vulnerable to the effects of salmeterol. One theory involves genetic variations in cells in the lungs.

Long-acting beta-agonists are inhaled drugs that are sold as single agents. They also can be combined in an inhaler with a steroid.

The leading beta-agonist/steroid combination product is Advair, sales of which have topped $4 billion a year since 2007, according to the market research firm IMS Health.

Another long-acting beta-agonist, Foradil (formoterol) also is available as a single-agent and as the combined product Symbicort, made by AstraZeneca.

In an email, GlaxoSmithKline spokeswoman Karen Collins said studies consistently have shown that Advair provides better overall asthma control than just an inhaled corticosteroid, including improvements in lung function, symptom control and less use of short-acting rescue inhalers.

She said both Advair and Serevent have improved asthma care.

“These medicines are used by millions of people worldwide and in the vast majority of people treated, they are well tolerated,” she said.

— John Fauber, Milwaukee Journal Sentinel