A convergence of peer- reviewed research, press scrutiny, federal investigations, and serendipity may lead to a clearer understanding of the safety and efficacy of a spinal fusion product manufactured by Medtronic, Inc.
Data from industry-sponsored trials of a product containing a synthetic bone growth factor used in spine surgery will undergo independent scrutiny by academic researchers.
(Photo credit: Medtronic, Inc)
At issue is Infuse Bone Graft, a product containing recombinant bone morphogenetic protein-2 (BMP-2), a synthetic bone growth factor that was approved in 2002 by the US Food and Drug Administration (FDA) for single-level anterior interbody lumbar fusion. The product has become widely adopted in spine surgery; by 2007, it was used in more than 50% of primary anterior lumbar spine fusions. Since then, the product’s use has expanded further, as positive efficacy and safety results for other indications (posterolateral lumbar fusion, posterior lumbar interbody fusion, and anterior cervical discectomy and fusion) from industry-sponsored trials have been published in the peer-reviewed literature.
But as the product’s use increased, so did skepticism. A news story published in August 2010 by the Milwaukee Journal Sentinel (http://tinyurl.com/4yfcn24) raised conflict-of-interest issues surrounding the FDA’s approval of BMP-2. The Journal Sentinel reported that at the FDA’s advisory committee meeting, a committee member noted a number of physicians whose research had been submitted to the agency had a financial stake in BMP-2 and that their results were more positive with respect to the device than were findings submitted from researchers who did not have a financial interest in the product.
Some researchers were also becoming skeptical about the postmarketing studies that validated using BMP-2 off label. For example, researchers noted that industry-sponsored trials published after 2002 found no adverse events from BMP-2 (Carragee EJ et al. Spine J. 2011;11:471-491), but a retrospective study of 328 468 patients undergoing spinal fusion procedures from 2002 to 2006 found that the use of BMP-2 for anterior cervical fusion procedures was associated with a higher rate of complications (7.09% with BMP-2 compared with 4.68% without BMP-2), with the primary increases seen in wound-related complications and dysphagia (Cahill KS et al. JAMA. 2006;302:58-66).
The questioning came to a head in June, with the publication of an entire issue of The Spine Journal that contained articles and an editorial critical of the use of BMP-2. In part, the authors of articles in the issue noted that Medtronic had paid the clinical investigators of the industry-sponsored postmarketing studies between $560 000 and $23.5 million per study.
That same month, the Senate Committee on Finance launched an investigation of Medtronic by sending a letter to Omar Ishrak, PhD, the company’s chairman and CEO. The letter requested documents pertaining to adverse postoperative events and medical complications associated with BMP-2, communications with the clinical investigators, and discussions with the FDA regarding adverse events and medical complications. “We are extremely troubled by press reports suggesting that doctors conducting clinical trials examining the safety and effectiveness of Infuse on behalf of Medtronic were aware that Infuse, a treatment commonly used in spinal surgery, may cause medical complications, but failed to report this in the medical literature,” wrote Committee Chairman Sen Max Baucus (D, Mont) and Sen Chuck Grassley (R, Iowa).
Meanwhile, as controversy swirled around BMP-2, Harlan M. Krumholz, MD, professor of medicine at Yale University School of Medicine, was mulling how to get medical companies to release all relevant research data on medications and devices being developed to treat patients.
Krumholz’s interest in the topic was spurred by his involvement with Merck’s anti-inflammatory osteoarthritis drug rofecoxib (Vioxx), which received FDA approval in 1999 and was withdrawn from the market in 2004 because subsequent research showed that the drug was associated with an increased risk of cardiovascular disease. Krumholz served as an expert witness for plaintiffs suing Merck for damages because, they claimed, they had experienced cardiovascular events while taking rofecoxib. His involvement in the trial allowed him to view all of Merck’s data relating to rofecoxib, and he discovered the company had unpublished data going back to 2001 showing increased cardiovascular risk with the drug.
“I have been working on this problem of data shielded from public view for quite a while,” Krumholz said. “It became obvious to me when working on Vioxx that data could have shed light on the cardiovascular risk if it had been in the public domain. It was also clear that this was not a singular event; remember, about half of all clinical trials are not published. I have been critical of industry holding back data, but I am not interested in being a critic—I want to find a solution.”
The BMP-2 issue enabled Krumholz to put his ideas into practice. “I approached Medtronic, and they were already thinking that they wanted to do something with the data,” Krumholz said. “I proposed a model to serve everyone’s purposes, allowing researchers access to data and putting the company in the strong position of promoting the best science.”
Previous SectionNext Section
Richard E. Kuntz, MD, MSc, Medtronic’s senior vice president and chief scientific, clinical, and regulatory officer, liked Krumholz’s pitch. BMP-2 “was challenged by a series of articles, and ultimately we wanted to address some of the issues raised,” Kuntz said. “We thought the easiest way to address this issue, and provide a reasonable action for the public, was to quickly get an external review of the data so there will be no question of what it says and does not say.”
Medtronic is providing Krumholz and his colleagues at Yale a $2.5 million, 18-month grant. The company said it will provide Yale with all available patient-level data on BMP-2 from company-sponsored clinical trials, both published and unpublished, as well as all FDA-filed adverse event reports. Yale, in turn, will assemble a panel of experts and commission 2 academically recognized clinical research organizations specializing in systematic reviews to conduct the analyses. Yale will then release its findings. “For credibility’s sake, we have complete freedom; Medtronic will have no say on who we select or what the rules will be,” said Krumholz.
In addition, Medtronic said it will make all of its trial data on BMP-2 publicly available on ClinicalTrials.gov, the clinical trial registry run by the National Library of Medicine. The company will also work with Yale to develop a program to provide other researchers access to all data on BMP-2 through a defined registration process and Web site. “Medtronic must release all the data and make it available to investigators. Everyone who applies will get it, but their funding and purpose will be made public, and they have to agree to post results,” said Krumholz.
Kuntz said Medtronic is taking a risk in releasing all of its data, but it is a needed risk. “A lot of us in the industry worry about malicious reviews, and making our data publicly available makes ourselves vulnerable to someone using inappropriate statistical measures,” Kuntz said. “But we need to gain public trust; we are not interested in manipulating data.”
Eugene J. Carragee, MD, a professor of orthopaedic surgery at Stanford University School of Medicine and editor of The Spine Journal, hopes the Krumholz model works. “This is, in my opinion, one of the ways that industry-sponsored trials can be evaluated by high-quality researchers—with a firewall between the sponsor and the people doing the analysis,” said Carragee, who has been a vocal critic of BMP-2 research and researchers. “Hopefully this can serve as a model for future research.”
But Carragee added caveats, saying his research into BMP-2 revealed flaws in company-sponsored study designs. “Based on the information we had, we found the trials to be seriously biased and designed to show superiority for the BMP group,” Carragee said. “You can handle all the numbers you want, but if the outcomes were systematically biased in favor of the BMP group, you can, in good faith, turn over all the numbers, and you can say it looks equivalent. But if you do not close the loop with the structural problems, you are then just analyzing the organic bias.”
Carragee said he hopes the Yale researchers dig deep. “If Yale takes Medtronic’s study protocols at face value, they will get one answer, but if they can get the protocols showing how the studies were really done, they will get a much different answer,” he said.
Also hopeful, but with reservations, of the Yale effort is Richard A. Deyo, MD, MPH, professor of medicine in the Department of Family Medicine at Oregon Health and Science University in Portland. “The idea of having an independent academic investigational review of a controversial situation like this seems a potentially valuable approach,” said Deyo, who has published a number of papers on spine therapies. “The reservations I have are I do not know what kind of access [Yale] will have to all the Medtronic records.”
A NEW PARADIGM?
Kuntz said that if the Krumholz model is shown to work, he envisioned that Medtronic would use it to assess its other products and that ultimately the approach might be adopted by the device industry as a whole. “Our real goal is to develop a new paradigm,” Kuntz said.
Krumholz said that while shining a light on the safety and efficacy of BMP-2 is important, his ultimate interest in the review is to determine whether the process works and can be applied to other research surrounding medical products. “This is a test to see if this model will work. It is voluntary, but companies can see it is in their best interest,” he said. “Their image needs some rehabilitation, and this could be a way for them to show that they are truly committed to improving health care.”
Find Full Text at :