FDA Insights: Top 10 Issues and Insights for Endovascular Grafts
Perspectives on the current state of endovascular grafting technology.
The views and opinions in this article are those of the authors and do not necessarily reflect those of the Food and Drug Administration, the US Department of Health and Human Services, or the Public Health Service.
With all the recent press regarding endovascular grafts, it seems an appropriate time to consider the issues and insights regarding these products from a device-evaluation point of view. The following perspectives include a reflection on what has been learned, where additional knowledge is needed, and how the field has changed over time from my vantage point. Some points seem quite basic, yet worth stating, given the misconceptions regarding endovascular grafts. My focus is on the clinical performance of these devices; the limitations and challenges of the preclinical evaluation of these devices has been covered in the July/August 2003 issue of Endovascular Today.
#1. Successful treatment is both physician- and patient-dependent. Successful surgical repair and successful endovascular repair are both very much physician- and patient-dependent. The patient’s anatomy, comorbidities, compliance to medication, lifestyle, and follow-up protocols are all significant factors in either type of vascular repair. Similarly, the physician’s experience and familiarity with the limitations of the device and/or treatment are critical factors contributing to the probability of success.
#2. All medical devices have inherent risks. The devices used for surgical and endovascular aneurysm repair have inherent risks. Unfortunately, prosthetic medical devices in general are imperfect; technology has yet to provide the ability to design replacement parts that function as well as healthy human tissue. There are also inevitable interactions between the natural tissue and the device, some of which are not favorable. For example, hip replacements may break, heart valves can fail, vascular anastomoses can become aneurysmal, and endovascular grafts may not offer complete protection from aneurysm rupture. Additionally, there are the potential problems with placing the prostheses, whether they are in the delivery and deployment of an endovascular graft or the challenges of major surgery.
#3. Each endovascular graft design has a unique risk/benefit profile. For endovascular grafts, the particular issues depend on device design. One device may be more difficult to place but has excellent long-term performance. Another may be associated with a greater number of growing aneurysms, and yet another with wire-form breaks, neither of which have been correlated to clinical sequelae. Additional long-term clinical data are needed to fully define the risk/benefit profile for most devices. Consideration of the limitations of each design and what is known about the design is needed in determining the best treatment option for a patient.
#4. Clinical updates are needed to ensure availability of current data for each marketed device. From a regulatory standpoint, this translates into an inability to dictate the optimal treatment for general patient populations. The significant number of confounding factors makes a direct comparison between surgical and endovascular repair, or even between endovascular devices, virtually impossible. As a regulator, the FDA tries to ensure that the clinical community and public have access to accurate data for individual devices that allows for a treatment strategy to be formulated between the physician and patient, with an emphasis on informed consent. There is a postapproval requirement for every manufacturer of a marketed endovascular graft to provide a clinical update to physician users on at least a yearly basis to inform them of the latest data available for the device.
#5. Five-year clinical follow-up is required. There is a need to continue to collect data on endovascular grafts to optimize the amount of information available for making treatment decisions. The FDA still requires that data continue to be reported after market, with at least 5-year follow-up. Another source of data is from devices known to have problems that would preclude them from obtaining marketing approval. There is an inclination for investigators and sponsors to terminate studies after the identification of failure modes or device limitations. Although it makes sense not to enroll new patients, patients who have already been treated with the device should continue to be closely followed, not only for their own safety, but also in the interest of obtaining new information. These de facto “positive controls” can be very useful in investigating whether an identified problem, such as wire fractures or other material defects, results in any difference in clinical performance over time as compared to devices without the problems.
#6. Additional data are needed to optimize study designs. As we learn more about endovascular grafts, we realize how much has been misunderstood. Additional data will not only help with treatment selection, but also with optimally designing clinical studies for these devices. For example, the optimal surrogate endpoint for devices intended to prevent aneurysm rupture has yet to be identified. Initially, there was much emphasis on endoleak. Once it became clear that although endoleaks may not be desirable, they are not always correlated to aneurysm size changes, the focus was then changed to aneurysm size. However, this surrogate endpoint is also under scrutiny because there is not a correlation with aneurysm growth and adverse events for some patients.
#7. The data must be complete and accurate. Complete and accurate data are needed in order to advance endovascular graft technology, whether it be in treatment selection or in designing studies. There has recently been much focus on follow-up compliance for endovascular graft studies. All reports to the FDA regarding clinical studies of endovascular grafts need to include tables describing not only how many of the eligible patients were evaluated at each scheduled follow-up interval, but also how many patients were assessed for each specific parameter, such as migration, endoleak, device integrity, and aneurysm size changes.
#8. Regulatory responsibilities must be taken seriously. Clearly, adherence to the specified protocol is needed to obtain useful information, but it is also necessary to ensure patient safety and to fulfill regulatory requirements. Recent events, such as the Guidant Ancure case, have demonstrated that there is little tolerance for not adhering strictly to regulatory responsibilities, as described previously in the September 2003 issue of Endovascular Today.
#9. FDA advisory panel review may not be needed for a new endovascular graft for the treatment of abdominal aortic aneurysms (AAA). Assuming the protocol and all the applicable regulatory rules were appropriately followed and no new concerns were identified during the review of a pre-market approval application for an endovascular graft intended to treat AAA, it may be possible for a device to be reviewed by the Agency without the need for an FDA advisory panel review. The panel has already discussed issues such as aneurysm rupture and wire fractures, so if these occur with similar frequency as for previously reviewed devices, a new device could potentially have had these problems identified during the clinical study but theoretically would not need to undergo panel review. The need for panel review is determined on a case-by-case basis. New device applications also benefit from having the labels and summaries of safety and effectiveness data (SSED) from recently approved devices to serve more or less as templates for new labels and SSEDs.
#10. Thoracic endovascular grafts present unique issues. All of the previously mentioned points were written with treatment of AAA in mind. The evaluation of endovascular grafts for the treatment of thoracic lesions presents enough challenges to warrant an entire separate list. Given the relatively small amount of information available for these devices, however, such a reflection is premature. What is known is that it is even more difficult to compare thoracic endovascular repair to surgical repair; there are significant challenges in clinical study design related to the different etiologies and implant locations; the thoracic environment is even less well understood than that of the AAA; and patients with the need for thoracic repair often do not have good alternative treatment options.
In conclusion, the field of endovascular grafting, although maturing, is still in the “toddler stage” at best. Much additional care and nurturing are needed to allow for this technology to achieve its full potential. Appropriate evaluation, communication, and cooperation are needed to allow for continued progress and to avoid setbacks that can be brought on by even unintentional breeches in proper conduct of studies and reporting.
Dorothy B. Abel, is the Convener of the ISO Vascular Prostheses Working Group; a Regulatory Review Scientist with the US FDA Center for Devices and Radiologic Health in Rockville, Maryland; she is also a regular columnist for Endovascular Today. Ms. Abel may be reached at (301) 443-8262; ext. 165; firstname.lastname@example.org.