Evelyn Pringle November 2006
In January, 2006, the FDA issued a Public Health Advisory for Trasylol, based on the serious health risks found in two studies, “A Propensity Score Case-Control Comparison of Aprotinin and Tranexamic Acid in High Transfusion Risk Cardiac Surgery,” by Karkouti et al, in the Journal of Transfusion, and “The Risk Associated with Aprotinin in Cardiac Surgery,” by Mangano et al, in the New England Journal of Medicine.
After the studies were published, the FDA scheduled a meeting on September 21, 2006, of the FDA’s Cardiovascular and Renal Advisory Committee to discuss the findings and other relevant data.
At the meeting, FDA briefing officials advised the panel that in 2005, “about 250,000 patients received Trasylol in the U.S.”
The NEJM study compared outcomes the groups of patients receiving the drugs aprotinin, aminocaproic acid and tranexamic acid, and one group of patients receiving no drug.
In the study, Trasylol was found to be associated with a 109% increase in heart failure, a 48% increase in myocardial infarction, a 181% increase in strokes, and a risk of kidney failure 259% greater than patients who received no drugs.
The study was conducted at the Ischemia Research and Education Foundation in San Bruno, California. Study leader, Dr Dennis Mangano, founder of the Foundation, told Consumer Affairs.com, that the results raised even more concerns than Vioxx because: (1) aprotinin has been on the market three times as long as Vioxx, yet few studies have been conducted; (2) the life-threatening complications occurred far more frequently; and (3) equally effective less expensive alternatives are available, but they are underused.
At the Advisory Committee meeting, Dr Mangano informed the panel that both aminocaproic and tranexamic showed blood-sparing capabilities equal to aprotinin. “Yet,” he stated, “head-to-head comparisons were few and far between despite the enormous difference in cost between these agents with aprotinin between $500 and $1,200 a patient and the others, less than $50 a patient.”
The NEJM study estimates that ending the use of Trasylol could prevent kidney failure in as many as 11,000 patients a year, which in turn would save more than $1 billion annually in dialysis costs.
Dr Keyvan Karkouti, presented the findings of the Transfusion study and told the panel, “Our primary outcome was efficacy transfusions and our hypothesis was that we would see a marked superiority with aprotinin mainly because that was the clinical impression and it still is at probably most places across Canada.”
“But what we found,” he reported, “was that there was really no difference in the
He said going in, their hypothesis was that if aprotinin is better, it should show a reduction in the risk of MI, stroke, renal failure, infection and death. “We didn’t find any reduction of these adverse events with aprotinin use,” he told the Committee.
“We did find a renal dysfunction increase with aprotinin,” he said, “and a trend toward increased renal failure.”
“So, in conclusion,” he advised, “there is as yet no conclusive evidence that aprotinin is better than tranexamic acid in our opinion.”
For Bayer’s presentation to the FDA Committee, Mike Rozycki, Bayer’s Director of US Regulatory Affairs, said that when the company learned of the new studies, “We immediately began a comprehensive review of all the data that we had in CABG surgery for aprotinin.”
“This was conducted,” he stated, “in very close association and under the guidance of the FDA.”
“All that information has been submitted and is under review by the FDA,” he added.
Mr Rozycki specified that the company’s presentation that day was based on the data available to Bayer.
On Bayer’s behalf, Dr Robert Makuch, a Professor of Biostatistics at the Yale School of Public Health, discussed the methodological considerations of randomized, controlled clinical trials and dismissed the NEJM study as unreliable. “In summary,” he said, “the Mangano study results should not be considered reliable at this time.”
While not revealing Bayer’s hidden study, he noted the discussions about the need for further analysis and independent access to Dr Mangano’s study data and said, “I certainly would endorse that recommendation, as well.”
Dr Pamela Cyrus appeared on behalf of Bayer and thanked the committee for having Bayer “here today to review our clinical trial data.”
As to safety information on Trasylol provided to the FDA, she specifically said, that Bayer had “submitted data on an ongoing basis to our NDA, as well as with our ongoing pharmacovigilance additional data.”
“With the two recent observational studies,” she told the panel, “we have also conducted a very thorough analysis of our global CABG data base.”
“We have submitted that analysis,” she noted, “as well as our data sets, to the FDA for their review.”
Dr Cyrus denied that Trasylol was associated with the same adverse effect revealed in the hidden study. “Bayer’s summary on cardiac safety is very simple,” she said, “Aprotinin was not associated with an increased incidence of myocardial infarction looking across all CABG patients.”
“Bayer’s conclusions on cerebrovascular safety,” she told the Committee, “is that aprotinin was not associated with an increased incidence of either stroke or encephalopathy with encephalopathy also including coma.”
Dr Jerrold Levy, Professor of Anesthesiology, Director of Cardiothoracic Anesthesiology and Deputy Chair of Research at Emory University, also appeared on behalf of Bayer to discuss the risk-benefit assessment and also flat-out denied the increased risk of adverse effects associated with Trasylol found in the hidden study, going so far as to say it decreased the risk of stroke.
Dr Levy told the panel, “if you look at the clinical studies of the randomized clinical trials in CABG surgery, there is no greater risk of mortality, MI, or renal failure and, at least from this data, there was a reduction in stroke in these patients.”
Regarding stroke, if you look at four different meta-analysis of studies,” he said, “there was a significant reduction in stroke.”
Committee member, Dr Valluvan Jeevanandam, asked Bayer representatives whether there were other studies. “Could you comment on other trials other than CABG trials where there might have been an effect on renal function,” he asked, “because a lot of the questions we had in our first presentation by Dr. Mangano was perhaps concomitant procedures with higher incidence of renal dysfunction, so do you have other trials other than just CABG trials looking at renal function?”
Dr Cyrus described one study as “the most recent” stating, “Probably the most recent study where you could just remove the effect of bypass totally is a study that was just conducted by Bayer in hip surgery.”
“So, if you forget the bypass effect on the kidneys and let’s look at a patient population that may not be at increased risk,” she stated, “there were no differences between the groups in that patient population.”
Dr Susan Heckbert also asked whether there were other studies stating, “I have a quick question about your global database, clinical trial database.”
“It looks to me like the U.S. studies in that database,” she said, “most of them are from the early nineties and they would reflect the kind of patient that would present and might be considered for a clinical trial in those days.”
Dr Heckbert specifically asked, “You don’t have anything beyond the early nineties in the United States in that database.”
“What about from other countries,” she inquired, “are we seeing the kind of patients that now go for CABG?”
“Do we have data in your global database from more contemporary types of patients?” she asked.
Dr Cyrus said, “The bulk of our clinical trial experience that I shared with you in the safety database was between the late eighties and late nineties.”
“There were a few studies that went to 2001,” she added.
“We do not have randomized clinical trials beyond 2001,” she responded, “in the database that I shared with you today.”
In the end, the Committee decided that there was no need for additional warnings on Trasylol labeling and Bayer quickly put out a press release stating, “the committee overwhelmingly affirmed (18 yes votes and one abstention) that the totality of clinical data presented in today’s meeting supports acceptable safety and efficacy for Trasylol among coronary artery bypass graft (CABG) surgery patients.”
The Committee agreed that the data are consistent with an association with aprotinin use and renal impairment, however, most members said they were not convinced that there was a definite increased risk of renal failure requiring dialysis.
Without the hidden study that proved otherwise, the Committee agreed that overall, there was no association between aprotinin and an increased risk of myocardial infarction, heart failure, stroke or encephalopathy.
Additionally, the Committee commented that these are short-term outcomes and that there is little or no data on the long-term cardiovascular outcomes in these patients.
Recommendations were made that, because there is limited current and long-term data, Bayer should be encouraged to sponsor such prospective studies to gather this information.
The Committee vote was taken to answer the specific question: “Based upon the presentations today, do you regard the totality of clinical data as supporting acceptable safety and efficacy for Trasylol usage among certain CABG/CPB patients?”
After being notified of the hidden study, the FDA issued another Advisory warning that the results of the study demonstrate “that use of Trasylol may increase the chance for death, serious kidney damage, congestive heart failure and strokes.”
Committee members were rightly outraged when they learned of the hidden study. Dr Robert Harrington, from Duke University, told Heartwire on October 6, 2006, “The process of evaluation, comment, and advice on important drug efficacy and safety data, only works when all of the involved parties are open and honest about their data.”
He noted that he found Bayer’s failure to disclose the study especially troubling, “when several panel members, including me, commented that more data and more analyses were needed to fully understand the risks and benefits of the drug.”
He called it “ironic” that the Committee spent part of the day criticizing Dr Mangano for failing to allow the FDA to analyze his database, “yet Bayer failed to even acknowledge the existence of these data.”
Committee member, Dr Michael Lincoff, from the Cleveland Clinic, made similar comments to Heartwire. “It was astounding to me,” he said, “that they did not disclose the information that the study had been conducted even if the findings were considered preliminary.”
“They were in the midst of an entire day’s discussions at the FDA on that precise topic,” he noted, “where there was substantial comment regarding the desirability of more contemporary data than their older trials.”
Dr Lincoff said it was “inconceivable,” that Bayer representatives did not know of the study or its potential relevance to the Committee.
In the November 23, 2006, New England Journal of Medicine, Dr Jerry Avorn, published an opinion paper and stated, “The analysis completed in time for the FDA meeting, reached conclusions similar to those of Mangano et al.”
“But neither Bayer nor its contractor,” he wrote, “had provided the report to the FDA or even acknowledged its existence before the meeting.”
Dr Avorn points out that a confirmatory observational study would have lent key support to the conclusions of Mangano et al — “if its findings had been aired,” he added.
Bayer has admitted that suppressing the study was “a mistake,” but Dr Avorn explains that this is not the first time the drug maker has been caught hiding research. When Bayer was accused of hiding data on its cholesterol-drug Baycol before it was taken off the market, litigation later unearthed a memo from a Bayer executive arguing against performing a study that would reveal its risk.
“If the FDA asks for bad news, we have to give,” the memo states, “but if we don’t have it, we can’t give it to them
If patients or family members suspect that injuries or death may have been caused by Trasylol, they need to ask the surgeon whether the drug was administered during surgery.