Last month, the U.S. Food and Drug Administration convened a committee of medical experts to weigh new evidence concerning the potential dangers of drospirenone, a synthetic hormone contained in popular birth control pills including Bayer AG’s Yaz and Yasmin. In a decision that helped ensure the continued presence of these drugs on American pharmacy shelves, the committee concluded by a four-vote margin that the benefits of drugs with drospirenone outweigh the risks. However, an investigation by the Washington Monthly and the British medical journal BMJ has found that at least four members of the committee have either done work for the drugs’ manufacturers or licensees or received research funding from them. The FDA made none of those financial ties public.
Yaz and Yasmin are among Bayer’s top selling pharmaceutical products. According to the German drug manufacturer, over 4 million women worldwide use Yasmin alone. But the drug has also sparked growing controversy in recent months. A series of studies published in BMJ have shown that users of pills containing drospirenone have an increased risk of blood clots, which can cause deep vein thrombosis, pulmonary embolism, stroke, heart attack and death. And thousands of women have filed a lawsuit against Bayer, saying they were injured by Yaz or Yasmin.
Other Bayer products containing the hormone drospirenone are Beyaz and Safyral. In addition, four companies—Bayer, Teva, Watson and Sandoz—sell five generic versions of the drugs evaluated by the committee last month.
The FDA’s decision not to reveal its advisors’ relationships with the drugs’ manufacturers and Bayer raises serious questions about the agency’s treatment of potential conflicts of interest, a historically problematic area for the department. In 2004, controversy erupted over another FDA advisory hearing—involving the painkiller Vioxx—when it became clear that advisors with financial conflicts of interest were more likely than those without conflicts to vote that Vioxx was safe and should be allowed to stay on the market. The drug was ultimately withdrawn after a study showed it was associated with approximately 60,000 deaths. In 2008, partly in response to this uproar, the agency issued a new guidance on how to handle conflicts of interest among its advisors, stating that it wanted to reduce “bias” and to be more “transparent.”
The agency’s handling of the drospirenone hearings casts major doubt on how far it has gone to attain those goals.
What’s more, interviews and records also show that the drospirenone advisory committee, when it voted last month, lacked significant information about increased risks to the health of women using contraceptives containing drospirenone.
The FDA ordered the safety review of drugs containing the hormone after three articles published in the BMJ found an increased incidence of blood clots among users of the pills. The advisors were given these studies prior to the December 8 meeting. However, the panel was not shown documents from a report by David Kessler, a former FDA commissioner and an expert witness in the lawsuit filed against Bayer on behalf the women who say they were injured by Yaz and Yasmin. Kessler’s report cites internal Bayer corporate documents that suggest the company had withheld safety data from the FDA.
The documents show that in 2004, Bayer scientists reported in a draft analysis that Yasmin incurs a “several-fold increase” in reporting rates for blood clots compared to three other oral contraceptives, and that Yasmin’s rate of all serious adverse events was “10 fold higher” than that of other products.
In his report, Kessler writes that Bayer failed to report safety data to the FDA, engaged in off-label promotion of the pills, and paid $450,000 to a high profile gynecologist to sponsor her book tour in which she promoted such off-label uses.
Kessler’s report, originally under seal, was not released until December 6, two days prior to the advisory meeting. The FDA said it could not give the advisors the information because the date to submit documents had passed.
The FDA relies on some 50 advisory committees to obtain “independent” expert advice and “maintain the public trust.” Recommendations on evidence for safety and efficacy can lead to drug approvals and withdrawals; while nonbinding, recommendations by such committees can move stock prices and are generally regarded as indicators of official FDA action.
Under FDA guidelines, when advisory committee members have a financial relationship with a drug company whose product is under review, the agency can issue waivers that acknowledge the advisors’ links to industry but allow them to vote if their expertise is needed and no other specialists are available. Such waivers often serve as the only public signal that advisors have financial conflicts of interest. In the hearings on drospirenone, the FDA issued none.
And yet, according to public records, recently unsealed court documents, and interviews with some of the FDA advisors, at least four of the advisory committee members in this case have served as paid researchers, consultants, “key opinion leaders,” or speakers for Bayer or other manufacturers or licensees of drospirenone. A fifth advisor agreed to serve as a consultant but never executed the agreement. A sixth received consulting fees from a law firm representing Bayer.
In a 15-to-11 vote, the committee decided that the benefits of drospirenone outweighed its risks.Had those members with undisclosed conflicts of interest voted the other way, the committee would have come to the opposite conclusion. The committee also voted 21 to 5 to change the drug labels to warn that the hormone can cause blood clots. But the panel stopped short of recommending that the labels should warn that they are more likely than other contraceptive pills to cause blood clots. Instead the experts suggested that the labels say that the evidence about blood clots is “conflicting.”
Ironically, while the FDA allowed voting by advisors with business connections to drospirenone, the agency did bar another advisor—Sidney M. Wolfe, the panel’s lone consumer advocate—on the grounds that he has an “intellectual conflict of interest.” Wolfe, a former researcher, is a frequent critic of FDA practices on behalf of the advocacy group Public Citizen. Wolfe, the author of the consumer guide Good Pills, Bad Pills, had marked Yaz with a “do not take” advisory to readers. Jerome Hoffman, professor emeritus of medicine at UCLA, questioned Wolfe’s exclusion from the panel: “The idea that because he has an opinion on data that already exist he should be prevented from voting is nutty.”
Each of the advisors with ties to manufacturers told the Washington Monthly and BMJ that they fully disclosed their ties to the FDA, and there are no indications that the advisors failed to make required disclosures to the agency. The FDA, however, has declined to make public advisors’ financial conflict of interest forms, explaining that the forms are “confidential” and no information from them can be shared under a section of the Ethics in Government Act.
When asked whether the agency was aware of any financial ties between its advisors and manufacturers or distributors of drospirenone, FDA spokeswoman Morgan Liscinsky said, “No waivers were issued.”
But the committee’s financial ties to Bayer begin with no less than its acting chair. Julia Johnson, a professor of obstetrics and gynecology at the University of Massachusetts Medical School, conducted four clinical trials—including one of drospirenone as hormone replacement—for Bayer or its subsidiary Berlex, a manufacturer of drospirenone. She said in an e-mail, “The FDA is very vigilant on examining potential conflicts of interest and was aware of all my research.” When asked for an interview, she said, “The U.S. FDA states that I cannot speak about the meeting.”
Another advisor with ties to industry is Paula Hillard, a professor of obstetrics and gynecology at Stanford School of Medicine who has served as a paid consultant to Bayer Schering. Dr Hillard said she had fully complied with all FDA disclosures and referred the Washington Monthly and BMJ to the FDA for further questions. The FDA declined to respond to any questions, again citing the confidentiality clause of the Ethics in Government Act.
Elizabeth Raymond, a senior medical associate at Gynuity Health Projects in New York and another member of the committee, conducted studies funded by Barr, which has a licensing agreement with Bayer for generic versions of Yaz. Dr Raymond said that the FDA was “fully aware of all of my relevant current and past activities.”
Anne E. Burke, an assistant professor of gynecology and obstetrics at Johns Hopkins University in Baltimore, has received research funding from Bayer-Berlex and Duramed, which has a licensing agreement with Bayer for generic Yaz and Yasmin. Dr Burke said that she fully disclosed this to the agency.
According to Dr Kessler’s report, an internal Bayer document indicated that a fifth advisor received payment from Bayer, but that advisor told the Washington Monthly and BMJ that although she initially agreed to serve as a consultant for Bayer, the company never pursued the agreement. She said FDA records indicated that she had ties to Bayer due to her disclosure about her agreement, but after she was nominated to be on the drospirenone advisory committee, she corrected the error.
A sixth advisor, meanwhile, confirmed that he received consulting fees from a law firm representing Bayer in 2006. He told BMJ and the Washington Monthly that he did not disclose the information to the FDA because the FDA “did not require information for that time frame for consulting unrelated to the meeting topic.”
Bayer spokesperson, Rosemarie Yancosek, said in an e-mailed statement: “Bayer had no input on who serves on the U.S. FDA Advisory Committee panel as the FDA has its own process for selecting panel members. Furthermore, it is Bayer’s understanding that the FDA has a procedure for determining conflicts of interest for potential panel members.”
The FDA does indeed have such a procedure, but critics argue that its guidelines define conflicts of interest too narrowly and provide too much flexibility in how they are applied. The guidelines are technically “suggested or recommended, but not required” provisions (http://www.fda.gov/downloads/RegulatoryInformation/Guidances/UCM125646.pdf). Whether an advisor can participate depends on “whether the discussion at the meeting or outcomes of the meeting will have a direct and predictable effect on the individual’s interest.” For instance, someone who was previously involved in another role for a manufacturer, or whose university received money from a manufacturer, may be allowed to participate. Even having a contract for $100,000 over a five-year period would not necessarily exclude an advisor, according to the guidelines.
Despite the FDA’s frequent claims that there aren’t enough experts independent of industry, a list of over 100 industry-independent experts has been presented to the agency – yet none of those experts has been tapped as an advisor.
The Project on Government Oversight, a Washington-based nonpartisan watchdog group, plans to raise concerns about the FDA’s conflict of interest policies in a letter to the agency’s top official, Margaret Hamburg.
“It shouldn’t require a lawsuit and investigative journalism to learn about these kinds of conflicts,” said the group’s executive director, Danielle Brian. “The Yaz case puts FDA’s feckless ethics policy into stark relief. It also reveals the human impact of conflicts of interest and why we need an FDA that shuns even the appearance of such conflicts.”
The organization intends to urge the FDA to routinely disclose industry ties and more aggressively determine whether an advisor has a conflict. “Right now,” said Brian, “the bar for considering whether an advisor has a conflict is too high.”