Updates From the Investigators
Principal investigators from four major US and international carotid artery therapy clinical trials discuss their latest developments, progress to date, and long-term goals.
THE CREST TRIAL
Thomas G. Brott, MD, presents the status of the landmark CREST trial, which anticipates completion of enrollment in the coming months and data presentation by early 2009.
Endovascular Today: In September 2007, the CREST investigators estimated that the trial might complete patient enrollment in mid-2008. How close is the trial to achieving this milestone?
Dr. Brott: CREST is on target. Our centers maintained an impressive rate of enrollment from September 2007 to the present. As a result, CREST will complete enrollment by mid-2008.
Endovascular Today: What were some of the early barriers to enrollment, and how have they been overcome? Which factors might still be hindrances?
Dr. Brott: CREST faced three major challenges early on that impacted the rate of patient enrollment.
First, CREST leadership placed very high priority upon including only highly qualified surgeons and interventionists. The Surgical Management Committee undertook the careful (and time-consuming) process of surgeon credentialing at the multiple CREST centers. The multidisciplinary Interventional Management Committee (IMC) performed careful initial selection of CREST stenters. These interventionists were required to complete animal training, submit procedural and follow-up reports for peer review, attend a carotid artery stenting (CAS) operator's course, and submit to a first-case observation by the device manufacturer's field representative—all just to receive credentialing into the lead-in phase. After that, each interventionist was required to treat up to 20 lead-in patients using the Acculink and Accunet stent system (Abbott Vascular, Santa Clara, CA; formerly of Guidant Corporation, Indianapolis, IN), demonstrating excellent technique and results, before receiving approval to enroll patients into the randomization phase. The CREST lead-in credentialing study, completed as part of this process, enrolled 1,561 patients.
Second, the Centers for Medicare & Medicaid Services (CMS) did not initially reimburse CAS within CREST. After a Presidential order was made in 2001, CAS became reimbursable for Medicare recipients. However, the order was not administratively accomplished at the state level until the end until the end of 2002. Only then were CREST enrollees covered by CMS. Most but not all private insurers followed, however, but even today, some carriers will not reimburse CAS within CREST.
Third, the introduction of embolic protection devices to CAS slowed CREST enrollment. At the time of the original CREST protocol, use of embolic protection during CAS was being developed but was not yet generally accepted by the endovascular interventional community. As a result, the use of embolic protection was not part of the stent procedure specified in the original CREST protocol. As the data supporting use of an embolic protection device rapidly evolved, CREST leadership evaluated it as well as reports from leaders within the CAS community. A consensus was reached to require use of an embolic protection device when technically feasible. As a result, an enrollment suspension was necessary; the CREST protocol was amended, endorsements were obtained at the FDA and NIH, and IRB approvals were accomplished at the multiple CREST centers.
Endovascular Today: How many centers are currently enrolling patients, and how has this number increased since the trial began in 2000?
Dr. Brott: CREST has 115 centers approved for participation in its randomization phase. This is a considerable increase from the 40-center maximum that the FDA had originally approved for trial participation. Despite slower than anticipated recruitment early on, we were never short on applications from centers interested in starting up. CREST was fortunate to have support from the NIH and FDA to increase its upper limit of sites on multiple occasions.
Endovascular Today: Being among the largest and farthest along of the current CAS trials, is there added external pressure on top of the already difficult process of conducting a major randomized controlled trial?
Dr. Brott: Actually, I think the external pressures may have lessened. Physicians and other health professionals have been eager for the results of our trial, perhaps even more so after the conflicting results of EVA-3S and SPACE. The CREST investigators and CREST centers have received impressive support from the research community.
Endovascular Today: Has the approval of new CAS platforms affected device selection in the trial at all?
Dr. Brott: As new devices became available and utilized by our interventionists outside of CREST, we did receive requests to consider their inclusion in the protocol. CREST leadership considered these requests and actually met with several device manufacturers to inquire about their possible participation in the trial. It was ultimately decided that it would be too time-consuming and cost-prohibitive to recredential every interventionist for each additional device approved.
Endovascular Today: What are the most significant findings in the interim data, from both the lead-in phase and the primary enrollment?
Dr. Brott: The important data from the randomization phase are blinded and will remain so until mid-2009. The lead-in results, at 30 days and 1 year, were presented in February 2008 at the AHA International Stroke Conference and should be published before the end of the year. We can share that the 30-day complication rates of CAS in the CREST lead-in were lower or comparable to results from the major randomized controlled trials of carotid endarterectomy and lower or comparable to the complication rates from the randomized controlled trials of CAS. Stroke and death at 30 days were more common for octogenarians, consistent with the findings from the SPACE trial.
Our long-term data regarding risk-factor management reveal improvements from baseline in systolic blood pressure and cigarette smoking. However, medical management of these risk factors could be improved upon. We are working with the CREST investigators, who in turn are working with the primary physicians of the CREST enrollees, striving for improvement.
Endovascular Today: What advice would the principal investigators of CREST offer physicians who are looking to be involved in the next generation of randomized CAS trials?
Dr. Brott: Come up with good questions that need answering, and then find great sites with great research teams to answer them.
Endovascular Today: Please describe the goals and study design of the TACIT trial.
Dr. Rundback: The TACIT trial is a three-arm trial comparing best medical therapy alone, best medical therapy with CAS, and best medical therapy with carotid endarterectomy in patients with duplex evidence of more than 60% stenosis and in asymptomatic patients with duplex evidence of stenosis of more than 60%. This is the largest population of patients with carotid disease, and the goal is to define the best form of therapy.
For the asymptomatic population, TACIT looks to resolve the prevailing issue, namely whether revascularization by any modality is better than contemporary medical treatment using extraordinary potent statin medications, antihypertensive drugs, and other therapies. The interesting thing is that asymptomatic patients are both the largest treated and the most untreated population of individuals at risk for stroke—most patients who undergo revascularization are asymptomatic, although most patients who have asymptomatic carotid arterial stenosis are never referred for revascularization! TACIT provides a unique window of opportunity not only to determine if revascularization is indeed better, but potentially, using measures such as neurocognitive function and plaque characteristics, to define populations in whom each of these therapies may actually be most beneficial. Even if you end up with a limited utility for revascularization within the wide asymptomatic population, it could potentially increase overall utilization of revascularization using stent technology.
Endovascular Today: What elements distinguish TACIT from other previous and current CAS trials?
Dr. Rundback: In terms of study design, TACIT is distinguished from all other trials by the inclusion of a medical treatment arm. The medical therapy alone arm is critical—many primary care doctors and neurologists believe that these patients are often best treated with contemporary medical care alone. In TACIT, we will have established risk factor reduction targets for LDL, blood pressure, smoking cessation, and diabetes. Also unique to TACIT is the primary outcome measure, which is a composite of 30-day mortality and 5-year all strokes and neurocognitive decline. TACIT is the only trial looking at neurocognitive decline as an endpoint measure, which many people believe is an important determinant of carotid disease evolution. Patients with "asymptomatic" carotid disease may not have overt symptoms of stroke, but they have clinically important neuropsychologic deterioration, and this will be measured in TACIT. A final distinguishing component of TACIT when compared to other studies is that within the secondary outcomes, we have two other very important features: (1) plaque characteristics and (2) the economic cost-effective analysis.
The characteristics of plaque are worthy of some mention. What is so critical about this is that clearly all plaque is not the same. There are numerous studies demonstrating that plaque characteristics by any measure, particularly duplex imaging, are a critical determinant of de novo events. What has not been well defined is how plaque characteristics affect interventional outcomes. Plaque characteristic assessment in TACIT may allow us to define this field so we can basically determine, on the foundation of the duplex imaging, which patients might be treated best medically, or with interventional or surgical strategies.
Endovascular Today: How is TACIT looking at neurocognitive decline?
Dr. Rundback: The TACIT trial includes, in the primary endpoint, well-defined and validated measures of neurocognitive decline, including trail-making tests A and B. The reason this is so important is that our definition of events may be outdated, and it may well be that asymptomatic carotid artery stenosis is associated with cognitive and functional declines outside of what is typically deemed to be the hard endpoints of stroke and death. It may be that these cognitive and functional declines are quite clinically relevant and important from a health care prospective and societal view. As such, cognitive and functional declines are going to be measured both within the primary endpoint using broad measures of cognitive function and as part of a substudy of approximately 400 patients using a battery of seven tests and a detailed neurocognitive assessment. Although we have neurocognitive decline as part of the primary endpoint and we have accounted for this in our estimation of event rates, we will continue to follow patients who reach the neurocognitive decline endpoint for the emergence of other hard endpoints, mainly stroke and death.
Endovascular Today: What are the criteria for the centers and the specialists or surgeons participating in the TACIT trial?
Dr. Rundback: The participating centers must have a team consisting of a surgeon and an interventionist (who can be the surgeon if he or she is qualified), as well as a neurologist. The sites must have demonstrated proficiency, as determined by the site selection committee (Kenneth Rosenfield, MD, in the US and Mark Sopoval, MD, in Europe), in the utilization of specific stent systems. We do not yet know which stent and distal protection systems will be used in the trial because it is subject to funding support by industry, but there will be more than one device utilized. Obviously, the site will have to provide logs and, although the exact level of proficiency has not been established, there will be a well-defined metric for the number of cases performed, documented success rates, and documented stroke rates.
The criteria for surgeons and interventionists will be that they have a documented level of proficiency and acceptable stroke rates within the previous year in certain volumes. The numbers that have been discussed are in the range of 25 to 50 endarterectomies in the previous year, with a combined stroke and death rate of less than 3%, which is consistent with American Heart Association guidelines. In the endovascular arm, trialists must have performed at least 25 endovascular cases, with documented stroke and death rates of less than 5%. However, these are conditional benchmarks, which the site selection committee will finalize once funding is complete.
So far, we have 130 sites in Europe and the US that have enlisted as TACIT sites, roughly two-thirds in the US and one-third in Europe. Approximately 80% of the interventionists are traditional endovascular practitioners (ie, radiologists, cardiologists) and roughly 20% are surgeons, with a recent surge in interest from surgeons. At the VEITH meeting, the surgeons expressed tremendous interest and support of the trial, and we are going to try to achieve balanced participation.
Endovascular Today: Do you anticipate that the results of TACIT will be able to show quality of outcomes by specialty?
Dr. Rundback: Unless there are striking differences in outcomes by specialty, TACIT is not powered for that. TACIT is a secondary analysis only. Although it is an intriguing concept, we are only qualifying sites that have demonstrated proficiency. We do not anticipate that there should be dramatic differences going forward.
Endovascular Today: You have mentioned that specific devices have not been determined. Can you provide any other insight as to what devices might be studied?
Dr. Rundback: We have worked quite extensively with the FDA Center for Devices and Radiologic Health (CDRH), to try to determine how many devices can be used within TACIT, whereby we would still obtain meaningful data for those companies to apply for a PMA. The feedback we get is that up to three devices can be utilized in TACIT and still get the necessary data based upon the number of implants used for approval of other similar devices. They will have to be devices that are available in Europe and the US (preapproval and via IDE), which is a bit of a limiting factor, and we will try to block randomize or control utilization of the devices so there is equal distribution of the devices within the trial.
Endovascular Today: What involvement has industry had in funding or conducting the trial?
Dr. Rundback: We have worked rather closely with industry and all the major stent manufacturers going back a number of years, which led to some initial funding for trial development, creation of the executive committee, and original trial design. That money has since been expended and we are now working with them to obtain funding to set up the TACIT home office, to file for the IDE, and to do the subcontracting with core labs and sites. The greatest hindrance we have had with the device companies at this particular time is that they are burdened by a very rigorous and costly postapproval study requirement. Unfortunately, we worked extensively but without success with CDRH to try to create a regulatory pathway whereby at least one of the companies would be able to enroll patients in TACIT in lieu of other postapproval study participants.
The other big stumbling block we have had is that, pending the results of the CREST trial, companies are a little hesitant to make a major investment in an asymptomatic carotid stent trial until at least some evidence of noninferiority of stenting compared to surgery is demonstrated. If early CREST reporting shows that stenting is inferior, it is going to be difficult to enroll in TACIT. We want to set up the trial so we are ready to go pending the results of those data.
Endovascular Today: What is the medical treatment regimen that is recommended in TACIT?
Dr. Rundback: We are going to determine a medical treatment table and provide therapeutic guidelines. However, what is unique is not just that there is a medical arm, but that TACIT will provide treatment targets for risk factor reduction. This is vastly different from other trials such as ACST, in which antilipidemic statin medications were given but outcomes were not necessarily monitored. A report card system is being implemented to ensure that appropriate levels of lipid lowering, smoking cessation, blood pressure control, and other risk factor modification is achieved.
Endovascular Today: Has the decision been made as to whether octogenarians and nonagenarians will be included in TACIT?
Dr. Rundback: At this point, we are intent on including octogenarians and nonagenarians who do not have other significant comorbid risk factors that would prevent follow-up to the completion of the trial.
Endovascular Today: What are some of the other obstacles that can hinder enrollment in a large-scale trial such as TACIT?
Dr. Rundback: The biggest hindrance to enrollment into TACIT is randomization within three arms, particularly because two of the arms are interventional, and one is medical therapy. They are such disparate forms of therapy. It is often hard to enroll patients in two-arm trials; it's even harder in three-arm trials. You have to explain at length to patients the very divergent alternatives and that they will be randomized. Our strategy is to explain that what we are offering the patients is two therapeutic alternatives; one is medical therapy and the other is revascularization. Once the patients are assigned to revascularization, it will be determined whether revascularization will be percutaneous or surgical, which I think eases the burden on the investigators having to explain three widely different alternatives. The other hindrance we have had is actually considered by many to be one of the advantages of the trial; that is, we are trying to compile extensive data on most of the meaningful and relevant areas of controversy within asymptomatic carotid populations, namely neurocognitive function, plaque characteristics, economics, and quality of life. However, in the long run, I think most people believe that the advantages of gathering these data are worth it; this may be the last chance we have to do that.
Endovascular Today: What is the study design of the ACST-2 trial?
Dr. Halliday: The ACST-2 trial (www.acst.org.uk) is a multicenter, international trial randomizing patients in whom intervention is definitely planned between carotid endarterectomy (CEA) and carotid artery stenting (CAS). Patients should not have had symptoms for at least 6 months before being entered, and both the clinician and the patient should be happy that there is no reason not to enter the study.
ACST-2 intends to randomize 5,000 patients equally between CEA and CAS. Patients will be randomized by telephone. The trial is a simple design and involves very little work for the collaborating physicians. A one-page randomization form is completed and, after CAS or CEA, a one-page follow-up form is returned to ACST. Subsequent follow-up, once a year, is obtained by sending a letter to the patient asking a few simple questions about how they are.
Endovascular Today: What kinds of subset analyses will be conducted?
Dr. Halliday: Subset analyses are planned for appropriately powered groups comparing gender, age, stenosis, and plaque composition. It is a trial powered to determine differences between treatments; for example, CAS might be associated with fewer myocardial infarctions, but there may be a greater stroke risk than CEA. This is why large numbers are needed, because otherwise a clear answer will not emerge. Our previous trial (ACST-1) gave clear answers on surgery for asymptomatic stenosis, showing that CEA reduced the 5-year risk of fatal and disabling stroke, as well as overall stroke. ACST-1 also showed that this benefit was clear for patients up to 75 years, and that those with stenoses of 70%, 80%, or 90% would benefit from surgery. These clear answers were only possible because ACST-1 was large enough (3,120 patients). An underpowered trial would not be helpful, especially when both of the two treatments being compared may be beneficial, and clarity on subgroups is desirable.
Endovascular Today: What characteristics most distinguish ACST-2 from past, ongoing, and planned large-scale trials?
Dr. Halliday: ACST-2 will be larger than any other single trial comparing surgery with stenting. It is simpler for clinicians to participate in ACST-2 than any other trial; ACST-2 is designed by the Oxford Clinical Trial Service Unit, which has expertise in the design of large-scale simple trials that give clear answers. The Heart Protection Study, which demonstrated the benefits of statin therapy, was designed and conducted by this group and changed worldwide practice in cardiovascular medicine.
Endovascular Today: Aside from the difference in size, what are some of the ways a trial such as ACST-2 differs from the recent SPACE and EVA-3S trials?
Dr. Halliday: ACST-2 differs from these other trials in that it is for asymptomatic patients only. CREST will have asymptomatic patients but not in numbers sufficient to determine important differences in treatment outcomes between CEA and CAS, according to the late Dr. Hobson.
SPACE 2, starting soon, is a three-armed trial in German-speaking countries only, comparing CEA, CAS, and medical treatment alone. ACST-2 and SPACE 2 will collaborate sharing data to inform meta-analysis at the end of both trials.
Endovascular Today: What is the level of experience required to participate in ACST-2?
Dr. Halliday: ACST-2 collaborators must complete a simple track record, showing overall experience, as well as experience in the last 2 years. For the last 2 years, the collaborator must have undertaken at least 25 procedures. Their track record must be countersigned by their neurologist/stroke physician, and their results must be in line with previous trials (ie, <8% stroke/death for symptomatic patients and <3% for asymptomatic patients). The record is blinded and reviewed by an independent committee, and the collaborator is able to join ACST-2 once approval has been gained.
Endovascular Today: What is the current enrollment status of ACST-2?
Dr. Halliday: The trial opened for randomization in 2008. Currently, we have 11 centers in six countries performing randomization. Larger numbers of centers are joining, and over 400 clinicians in more than 40 countries have expressed an interest, with more than 100 track records approved.
Endovascular Today: At what point do you think interim data will be presented? What is the target for full data presentation?
Dr. Halliday: It will be several years before we have interim data and results. We are actively seeking many more centers and plan to be recruiting for at least 5 years, with continuing follow-up, so patient follow-up is at least 5 years.
Endovascular Today: What is the ACT 1 trial designed to determine? What distinguishes it from other past and present CAS trials?
Dr. Matsumura: ACT 1 will compare carotid endarterectomy to carotid stenting with routine embolic protection in asymptomatic nonoctogenarian patients at standard risk for endarterectomy. No other multicenter randomized trial has focused on this comparison.
Endovascular Today: How is ACT 1 funded?
Dr. Matsumura: The trial is sponsored by Abbott Vascular (Santa Clara, CA).
Endovascular Today: What is the experience level of the physicians participating in the study, and how and why were the criteria determined?
Dr. Matsumura: The qualifications of both surgeons and interventionists are carefully evaluated by independent physician committees based on extensive reviews, including records of training, certification, prior experience, personal recommendations, and independent audits. There is a broad range of interventional specialists, including neuroradiologists, neurosurgeons, cardiologists, and vascular surgeons.
Endovascular Today: Which devices are being used?
Dr. Matsumura: The trial tests the Xact carotid stent and Emboshield filter (Abbott Vascular) in the CAS arm.
Endovascular Today: What measures have been taken to ensure scientific integrity in an industry-sponsored trial, and how are the outcomes and data being reviewed and verified?
Dr. Matsumura: There are several internal quality processes and external checks. The external checks include the multispecialty physician leadership (eg, national principal investigators, executive committee, surgical management committee/interventional management committee, operations committee, etc.), oversight by the FDA, independent core labs, each local institutional review board, and the independent Clinical Events Committee and Data Safety Monitoring Board (DSMB) through Harvard Clinical Research Institute.
Endovascular Today: What milestones have been met in the trial so far with regard to patient enrollment and any interim data? Are there any surprising observations or trends to date?
Dr. Matsumura: The enrollment of randomized patients has been above goal, and there are 545 randomized patients as of July 4, 2008. I have no insight into the randomized results other than the DSMB has not stopped the trial. The lead-in phase has a combined death/stroke/myocardial infarction rate of 1.7% (two minor strokes) in the first 118 patients.
Endovascular Today: When will the trial be complete?
Dr. Matsumura: This randomized trial is planned for 5-year follow-up, and as such, it will continue well into the next decade. It is a testament to the commitment of the research subjects, the many skilled investigators, and the sponsor.