Problems with Drug-Eluting Coronary Stents — The FDA Perspective
http://www.100md.com
《新英格兰医药杂志》
The introduction of drug-eluting coronary stents in the United States has rapidly and profoundly affected the field of interventional cardiology. Met with widespread anticipation by the clinical community, drug-eluting stents are now used in a majority of intracoronary stenting procedures, justifying their characterization as a "transforming technology." However, both such stents that are currently approved by the Food and Drug Administration (FDA), the Cordis Cypher sirolimus-eluting stent and the Boston Scientific Taxus Express2 paclitaxel-eluting stent, have also been associated with highly publicized adverse events after they were approved for marketing, leading some to question their safety. As a by-product of this concern, there has been considerable interest in how these stents are regulated by the FDA. Questions have been raised about the agency's evaluation of their safety and effectiveness before marketing and about its actions after the adverse events became known.
(Figure)
Cypher, the Sirolimus-Eluting Coronary Stent Made by Cordis, and Taxus, the Paclitaxel-Eluting Coronary Stent Made by Boston Scientific.
The difficulties in interpreting data on reports of adverse events during the postmarketing period were recently highlighted by our decision to issue public health notifications regarding the Cypher stent. The first drug-eluting stent, Cypher was approved by the FDA in April 2003.1 Soon afterward, we saw an influx and clustering of reports of subacute thrombosis associated with the stent, and Cordis, in conjunction with the FDA, issued a letter in July 2003 notifying physicians of these initial reports. By late October 2003, we had received reports of approximately 300 cases of subacute thrombosis through our Medical Device Reporting system, including more than 60 associated patient deaths.
Because of the voluntary nature of the reporting system, and given the absence of reliable information on the denominator (the number of patients receiving coronary stents), it was impossible to determine precisely whether subacute thrombosis was occurring more frequently in patients who received drug-eluting stents than in those who received standard bare-metal coronary stents. In addition, because the reporting system typically captures limited clinical information, determining the cause of the thromboses was problematic. Despite these difficulties, the continued appearance and early clustering of reports of subacute thrombosis and continued concern about inappropriate use led us to issue a public health notification in October 2003,2 which reminded clinicians to follow the product's labeling, including patient selection, sizing of the stent, and concomitant antiplatelet therapy. Although there was no definitive evidence at the time to directly link variations in implantation technique or patient selection to subacute thrombosis, we decided to issue the notification because of concern about the potential risks posed by untested, off-label use of this device.
Within about a month, additional information from Cordis's own clinical trial data, along with our review of comparable data on bare-metal stents, led us to conclude that there was no evidence suggesting that use of the Cypher stent, when implanted in accordance with the approved indications, was associated with an excess of subacute thromboses as compared with bare-metal stents. An updated notification was posted on the FDA Web site to apprise the public of the agency's current understanding of the situation.3
Another example of balancing risk and benefit to the public health was illustrated in the case of the Boston Scientific Taxus Express2 coronary stent system, which was approved for marketing by the FDA in March 2004 on the basis of results from a clinical trial involving 1314 patients.4 The delivery catheters used with the stent, the Express2 Over-the-Wire and Express2 Monorail delivery systems, were originally approved for use with the uncoated Express stent in September 2002. Approval of the Taxus Express2 system was granted after laboratory testing and the analysis of data on short-term safety and performance from more than 1100 patients in an additional clinical trial demonstrated the safety of the Taxus stent mounted on an Express2 catheter.
After the product was approved, Boston Scientific and the FDA received more than 40 reports of failures of the stent-delivery balloon to deflate after the stent was deployed; these events were associated with 18 serious injuries and 1 death. Through engineering analyses, Boston Scientific determined that the failures of balloon deflation were related to a condition of the manufacturing process for the Express2 stent-delivery systems that could lead to focal weakening of the outer lumen of the delivery catheter. When a pulling force was exerted on the catheter shaft, the weakened area could restrict the flow of contrast medium out of the balloon and prevent the balloon from deflating.
The company implemented a modification to the manufacturing process for the catheter that was approved by the FDA on May 4, 2004. As additional reports of adverse events were received, Boston Scientific initiated a recall on July 2 of 200 Taxus Express2 stents that had been made before the manufacturing changes occurred. Ongoing investigations by the company identified additional production lots that might have similar problems. As a result, the recall was expanded on July 16 to include 85,000 Taxus stent systems, and on August 5, an additional 3000 were recalled. The FDA continues to work with the company in an ongoing investigation of the issue.
In addition to the reports of balloon-deflation failure, Boston Scientific and the FDA have received reports about resistance of the balloon to withdrawal attempts after the stent is deployed — a problem sometimes described as "stickiness" of the balloon during withdrawal. This problem does not appear to be related to the mechanical defect observed in the investigation of the balloon-deflation failures; rather, it seems to result from a combination of patient-related and device-related factors. These factors are still under review but are currently presumed to include friction between the stent-delivery balloon and the drug-polymer coating on the stent, as well as certain characteristics of the coronary vessel. The association, if there is any, between these reports of resistance to withdrawal and adverse clinical events is currently being assessed.
The Boston Scientific case illustrates the dilemma we face in determining the appropriate threshold for product recall and notification when the risk is of a severe but rare adverse event. In these cases, the potential benefit to be accrued from recalling the product must be balanced against the possible negative effects this action might have on public health. Denying a potentially beneficial therapy to all patients, as in a complete recall of a product, might have greater adverse consequences than allowing the device to remain in use.
These cases exemplify the challenges that we face in regulating breakthrough technologies. In almost all instances, as in the Cypher case, we do not have a complete accounting of all cases or information on the number of devices actually in use, making it impossible to compare similar products in terms of the rate of adverse events. Our responsibility to protect the public health, however, necessitates that we consider early notification about any legitimate risk, weighing the risk posed by neglecting to notify practitioners about a clinically significant problem against the consequences of "crying wolf" on the basis of inconclusive information. In attempting to strike the right balance, we may engage clinical and other stakeholders for review and comment on our draft notifications before we release them. Such a dialogue is particularly important in the case of breakthrough technologies, in which our decisions can have an immediate and profound effect on people's lives.
Source Information
From the Center for Devices and Radiological Health, Food and Drug Administration, Rockville, Md.
References
CYPHER sirolimus-eluting coronary stent on RAPTOR over-the-wire delivery system or RAPTORRAIL rapid exchange delivery system — P020026. Rockville, Md.: Food and Drug Administration, April 24, 2003. (Accessed September 29, 2004, at http://www.fda.gov/cdrh/pdf2/p020026.html.)
FDA public health Web notification: information for physicians on sub-acute thromboses (SAT) and hypersensitivity reactions with use of the Cordis CYPHER coronary stent. October 29, 2003. (Accessed September 24, 2004, at http://www.fda.gov/cdrh/safety/cypher.pdf.)
FDA public health Web notification: updated information for physicians on sub-acute thromboses (SAT) and hypersensitivity reactions with use of the Cordis CYPHER sirolimus-eluting coronary stent. November 25, 2003. (Accessed September 28, 2004, at http://www.fda.gov/cdrh/safety/cypher2.pdf.)
TAXUS Express 2 paclitaxel-eluting coronary stent system (monorail and over-the-wire) — P030025. Rockville, Md.: Food and Drug Administration, March 4, 2004. (Accessed September 28, 2004, at http://www.fda.gov/cdrh/pdf3/p030025.html.)(Neal I. Muni, M.D., M.S.P)
(Figure)
Cypher, the Sirolimus-Eluting Coronary Stent Made by Cordis, and Taxus, the Paclitaxel-Eluting Coronary Stent Made by Boston Scientific.
The difficulties in interpreting data on reports of adverse events during the postmarketing period were recently highlighted by our decision to issue public health notifications regarding the Cypher stent. The first drug-eluting stent, Cypher was approved by the FDA in April 2003.1 Soon afterward, we saw an influx and clustering of reports of subacute thrombosis associated with the stent, and Cordis, in conjunction with the FDA, issued a letter in July 2003 notifying physicians of these initial reports. By late October 2003, we had received reports of approximately 300 cases of subacute thrombosis through our Medical Device Reporting system, including more than 60 associated patient deaths.
Because of the voluntary nature of the reporting system, and given the absence of reliable information on the denominator (the number of patients receiving coronary stents), it was impossible to determine precisely whether subacute thrombosis was occurring more frequently in patients who received drug-eluting stents than in those who received standard bare-metal coronary stents. In addition, because the reporting system typically captures limited clinical information, determining the cause of the thromboses was problematic. Despite these difficulties, the continued appearance and early clustering of reports of subacute thrombosis and continued concern about inappropriate use led us to issue a public health notification in October 2003,2 which reminded clinicians to follow the product's labeling, including patient selection, sizing of the stent, and concomitant antiplatelet therapy. Although there was no definitive evidence at the time to directly link variations in implantation technique or patient selection to subacute thrombosis, we decided to issue the notification because of concern about the potential risks posed by untested, off-label use of this device.
Within about a month, additional information from Cordis's own clinical trial data, along with our review of comparable data on bare-metal stents, led us to conclude that there was no evidence suggesting that use of the Cypher stent, when implanted in accordance with the approved indications, was associated with an excess of subacute thromboses as compared with bare-metal stents. An updated notification was posted on the FDA Web site to apprise the public of the agency's current understanding of the situation.3
Another example of balancing risk and benefit to the public health was illustrated in the case of the Boston Scientific Taxus Express2 coronary stent system, which was approved for marketing by the FDA in March 2004 on the basis of results from a clinical trial involving 1314 patients.4 The delivery catheters used with the stent, the Express2 Over-the-Wire and Express2 Monorail delivery systems, were originally approved for use with the uncoated Express stent in September 2002. Approval of the Taxus Express2 system was granted after laboratory testing and the analysis of data on short-term safety and performance from more than 1100 patients in an additional clinical trial demonstrated the safety of the Taxus stent mounted on an Express2 catheter.
After the product was approved, Boston Scientific and the FDA received more than 40 reports of failures of the stent-delivery balloon to deflate after the stent was deployed; these events were associated with 18 serious injuries and 1 death. Through engineering analyses, Boston Scientific determined that the failures of balloon deflation were related to a condition of the manufacturing process for the Express2 stent-delivery systems that could lead to focal weakening of the outer lumen of the delivery catheter. When a pulling force was exerted on the catheter shaft, the weakened area could restrict the flow of contrast medium out of the balloon and prevent the balloon from deflating.
The company implemented a modification to the manufacturing process for the catheter that was approved by the FDA on May 4, 2004. As additional reports of adverse events were received, Boston Scientific initiated a recall on July 2 of 200 Taxus Express2 stents that had been made before the manufacturing changes occurred. Ongoing investigations by the company identified additional production lots that might have similar problems. As a result, the recall was expanded on July 16 to include 85,000 Taxus stent systems, and on August 5, an additional 3000 were recalled. The FDA continues to work with the company in an ongoing investigation of the issue.
In addition to the reports of balloon-deflation failure, Boston Scientific and the FDA have received reports about resistance of the balloon to withdrawal attempts after the stent is deployed — a problem sometimes described as "stickiness" of the balloon during withdrawal. This problem does not appear to be related to the mechanical defect observed in the investigation of the balloon-deflation failures; rather, it seems to result from a combination of patient-related and device-related factors. These factors are still under review but are currently presumed to include friction between the stent-delivery balloon and the drug-polymer coating on the stent, as well as certain characteristics of the coronary vessel. The association, if there is any, between these reports of resistance to withdrawal and adverse clinical events is currently being assessed.
The Boston Scientific case illustrates the dilemma we face in determining the appropriate threshold for product recall and notification when the risk is of a severe but rare adverse event. In these cases, the potential benefit to be accrued from recalling the product must be balanced against the possible negative effects this action might have on public health. Denying a potentially beneficial therapy to all patients, as in a complete recall of a product, might have greater adverse consequences than allowing the device to remain in use.
These cases exemplify the challenges that we face in regulating breakthrough technologies. In almost all instances, as in the Cypher case, we do not have a complete accounting of all cases or information on the number of devices actually in use, making it impossible to compare similar products in terms of the rate of adverse events. Our responsibility to protect the public health, however, necessitates that we consider early notification about any legitimate risk, weighing the risk posed by neglecting to notify practitioners about a clinically significant problem against the consequences of "crying wolf" on the basis of inconclusive information. In attempting to strike the right balance, we may engage clinical and other stakeholders for review and comment on our draft notifications before we release them. Such a dialogue is particularly important in the case of breakthrough technologies, in which our decisions can have an immediate and profound effect on people's lives.
Source Information
From the Center for Devices and Radiological Health, Food and Drug Administration, Rockville, Md.
References
CYPHER sirolimus-eluting coronary stent on RAPTOR over-the-wire delivery system or RAPTORRAIL rapid exchange delivery system — P020026. Rockville, Md.: Food and Drug Administration, April 24, 2003. (Accessed September 29, 2004, at http://www.fda.gov/cdrh/pdf2/p020026.html.)
FDA public health Web notification: information for physicians on sub-acute thromboses (SAT) and hypersensitivity reactions with use of the Cordis CYPHER coronary stent. October 29, 2003. (Accessed September 24, 2004, at http://www.fda.gov/cdrh/safety/cypher.pdf.)
FDA public health Web notification: updated information for physicians on sub-acute thromboses (SAT) and hypersensitivity reactions with use of the Cordis CYPHER sirolimus-eluting coronary stent. November 25, 2003. (Accessed September 28, 2004, at http://www.fda.gov/cdrh/safety/cypher2.pdf.)
TAXUS Express 2 paclitaxel-eluting coronary stent system (monorail and over-the-wire) — P030025. Rockville, Md.: Food and Drug Administration, March 4, 2004. (Accessed September 28, 2004, at http://www.fda.gov/cdrh/pdf3/p030025.html.)(Neal I. Muni, M.D., M.S.P)