Protocol. This trial protocol has been provided by the authors to give readers additional information about their work.

Transcription

1 Protocol This trial protocol has been provided by the authors to give readers additional information about their work. Protocol for: Chimowitz MI, Lynn MJ, Derdeyn CP, et al. Stenting versus aggressive medical therapy for intracranial arterial stenosis. N Engl J Med 2011;365: DOI: /NEJMoa

2 Study Protocol Neurological Principal Investigator Marc I. Chimowitz MBChB Professor of Neurology Medical University of South Carolina Neurointerventional Principal Investigators Colin P. Derdeyn, MD David Fiorella, MD PhD Professor of Radiology Professor of Clinical Neurological Surgery Neurological Surgery & Neurology and Radiology Washington University Stony Brook University St. Louis, MO Stony Brook, NY Data Management and Statistical PI Michael J. Lynn M.S. Rollins School of Public Health, Emory University Department of Biostatistics Atlanta GA Version December 3, 2010 December 3, 2010 page 1 of 42

3 PROTOCOL SIGNATURE PAGE I have read this protocol and agree to adhere to the requirements. I will provide copies of this protocol and all pertinent information to the study personnel under my supervision. I will discuss this material with them and ensure they are fully informed regarding the investigational plan and the conduct of the study according to 21 CFR parts 50, 54, 56 and 812, ICH Good Clinical Practices Guidelines and Institutional Review Board (IRB) requirements. Clinical Site Site Principal Neurology Investigator Signature Date Site Principal Neurology Investigator Printed Name Site Principal Neurointerventionalist Investigator Signature Date Site Principal Neurointerventionalist Investigator Printed Name December 3, 2010 page 2 of 42

4 TABLE OF CONTENTS 1. Study Summary Study Rationale Primary Aim and Hypothesis Study Design Rationale for Antithrombotic Therapy Sample Size Estimates Primary Endpoints Secondary Endpoints Patient Selection Criteria Inclusion Criteria Exclusion Criteria Informed Consent Enrollment of Patients Randomization Angiographic Evaluation and Lesion Measurement Stenting Procedure Antithrombotic Protocol 19 Pre-Procedure 19 Intra-Procedure 20 Post Procedure 20 Anesthesia / Arterial Access 20 Procedure Steps Angiographic Assessment of Lesion and Stent Selection 20 Wingspan Stent System Preparation 21 Study PTA Balloon Catheter Preparation 22 Guidewire Positioning 22 Balloon Deployment 22 Stent Positioning and Deployment 22 Management of Blood Pressure During the Procedure 24 Imaging Immediately Post-Stenting 24 Post-Dilation 24 Rescue Procedures 25 Post-Procedure Care Schedule for Follow-Up Risk Factor Management. Primary Risk Factors Achieving Target LDL Blood Pressure Management Secondary Risk Factors Non-HDL Cholesterol Diabetes Management Smoking Cessation Targeted Weight Management Exercise December 3, 2010 page 3 of 42

5 Study Medications Biomarkers for Predicting Major Vascular Events Management of TIAs or Cerebral Infarct with Temporary Signs 30 During Follow-up 18. Adverse Events Evaluation of End Points and Adverse Events Statistical Analysis Quality Assurance Study Drug Supply and Accountability Study Device Labeling and Accountability Regulatory Requirements Institutional Review Board. 36 References.. 37 Appendices. Appendix 1 SAMMPRIS Study Protocol Flowchart Appendix 2 - Pathways to Randomization and Stenting Appendix 3 Schedule of Visits and Tests December 3, 2010 page 4 of 42

6 1. STUDY SUMMARY OBJECTIVE The primary aim of Phase 1 of this study of patients with 70%-99% stenosis of a major intracranial artery (MCA, carotid, vertebral, basilar) and TIA or non-disabling stroke within 30 days prior to enrollment is to determine whether intensive medical therapy plus intracranial angioplasty and stenting is superior to intensive medical therapy alone for preventing: any stroke or death within 30 days after study entry, or any stroke or death within 30 days of any revascularization procedure of the qualifying symptomatic intracranial artery done during follow-up, or ischemic stroke in the territory of the symptomatic intracranial artery from day 31 after study entry to completion of Phase I The aim of Phase 2 of the study, which will involve only patients in the stented arm, is to obtain longer term safety data in patients undergoing intracranial stenting STUDY DESIGN AND FUNDING SOURCE Phase 1 is an investigator-initiated, multi-center, randomized clinical trial funded by NIH / NINDS. Phase 2 is a longitudinal cohort study funded by Boston Scientific Corporation STUDY POPULATION CLINICAL SITES 764 patients with symptomatic intracranial stenosis Up to 60 actively enrolling sites in the USA PRIMARY ENDPOINTS Any stroke (ischemic, parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage) or death within 30 days after enrollment OR any stroke (ischemic, parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage) or death within 30 days of any revascularization procedure of the qualifying symptomatic intracranial artery done during followup, OR an ischemic stroke in the territory of the symptomatic intracranial artery from day 31 after study entry to completion of Phase 1. December 3, 2010 page 5 of 42

7 SECONDARY ENDPOINTS Clinical Secondary Endpoints 1. Ischemic stroke outside of the territory of the symptomatic artery beyond 30 days after enrollment 2. any parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage beyond 30 days after enrollment 3. myocardial infarction 4. major non-stroke hemorrhage (i.e., any subdural hemorrhage, any epidural hemorrhage, major systemic hemorrhage) 5. any death beyond 30 days after enrollment 6. disabling stroke 7. any stroke or death 8. any serious adverse event (see Section 18) 9. functional outcome at 1 year as measured by the Rankin Scale and Barthel Index 10. Cognitive outcome at 4 months (safety endpoint), 1 year, and end of Phase I as measured by Montreal Cognitive Assessment (MoCA) Technical Secondary Endpoints: 1. Balloon success 2. Stent success 3. Procedural success See section 8 for definitions of these technical endpoints. INVESTIGATIONAL DEVICE NAME Wingspan Stent System and Gateway PTA Balloon Catheter (manufactured by Boston Scientific Corporation) INTENDED USE The Wingspan Stent System is intended for the treatment of patients with stroke or TIA attributable to atherosclerotic disease in intracranial vessels with > 70% stenosis that are accessible to the system. The Gateway PTA Balloon Catheter is intended for balloon dilation of the stenotic portion of intracranial arteries prior to stenting for the purpose of improving intracranial perfusion. STUDY SUCCESS For the purpose of the IDE application associated with this device and this protocol, study success is December 3, 2010 page 6 of 42

8 defined as showing that compared with intensive medical therapy alone, intracranial angioplasty and stenting combined with intensive medical therapy decreases the risk of the primary endpoint by 35% over a mean follow-up of two years in high-risk patients with symptomatic stenosis of a major intracranial artery PATIENT SUCCESS Patient success occurs when a patient reaches completion of follow-up in the trial without having a primary endpoint PRINCIPAL INVESTIGATORS Neurology P.I.: Marc I. Chimowitz MBChB Sponsor-Investigator Professor of Neurology Medical University of South Carolina Charleston, SC Neurointerventional PIs: Colin Derdeyn MD Professor of Radiology, Neurological Surgery and Neurology Washington University, St Louis, MO David Fiorella, MD PhD Professor of Clinical Neurological Surgery and Radiology Stony Brook University Stony Brook, NY Data Management and Statistical PI: Michael J. Lynn M.S. Rollins School of Public Health, Emory University Department of Biostatistics Atlanta GA December 3, 2010 page 7 of 42

9 2. STUDY RATIONALE Atherosclerotic stenosis of the major intracranial arteries (carotid siphon, middle cerebral artery, vertebral artery, basilar artery) is an important cause of ischemic stroke, especially in Blacks, Asians, and Hispanics 1-8. In the USA, intracranial stenosis causes 8% 10% of ischemic strokes (approximately 50,000 strokes per year) at a cost of $750,000,000 in year 1 and $4.5 billion over the life time of these patients. Based on the ethnic and racial makeup of the world s population, some investigators have suggested that intracranial atherosclerosis is the most common cause of stroke world-wide. Given the socio-economic and personal burden of stroke from intracranial stenosis, our study group has been committed to evaluating therapies for this disease for the past 15 years. Our most important advancement to date was completing the NINDS funded Warfarin Aspirin Symptomatic Intracranial Disease (WASID) trial. In WASID, the safety and efficacy of warfarin (target INR 2-3) vs. aspirin (1300 mg per day) were compared in patients with symptomatic intracranial stenosis. The major findings in WASID were 1. aspirin was as effective and safer than warfarin for preventing stroke and vascular death in patients with symptomatic 50%-99% stenosis of a major intracranial artery patients with severe stenosis (70%-99%), recent symptoms, female gender, diabetes, and baseline NIH stroke scale > 1 were at highest risk of stroke in the territory of the stenotic artery 10. In particular, patients with 70-99% stenosis and TIA or stroke within 30 days prior to enrollment had the highest rate of ischemic stroke in the territory of the symptomatic artery: 22.9 % at 1 year (95% CI 15.4% %); 25.0% at 2 years (95% CI 17.2% %). 3. patients with poorly controlled blood pressure or elevated low denisty lipoprotein during follow-up had a significantly higher rate of stroke, vascular death, or MI compared with patients with good control of these risk factors. These results indicate that aspirin should be used rather than warfarin for intracranial stenosis and that certain subgroups of patients are at particularly high risk of stroke on antithrombotic therapy and usual management of risk factors, suggesting that alternative therapies are needed for these patients. Possible therapeutic options include intensive risk factor management strategies (yet to be tested in patients with large artery extracranial or intracranial atherostenosis) and intracranial stenting. Stenting has emerged as a promising treatment for intracranial stenosis that is increasingly being used in clinical practice in the USA and other developed countries. Even before the results of WASID were known, some investigators were already recommending that stenting should be first line treatment for intracranial stenosis. However, the use of stenting without evidence that it provides additional benefit over intensive medical therapy alone would bypass the scientific process necessary to develop an evidencebased approach for the treatment of this disease. Such an approach will require a randomized trial comparing stenting and intensive medical therapy vs. intensive medical therapy alone in patients with symptomatic intracranial arterial stenosis, which is described in this protocol. 3. PRIMARY AIM AND HYPOTHESIS OF THE RANDOMIZED TRIAL Primary Aim: To determine whether intracranial angioplasty (using the Gateway PTA balloon) followed by stenting (using the Wingspan self-expanding nitinol stent) combined with intensive medical therapy is superior to intensive medical therapy alone for preventing the primary endpoint (any stroke or death within 30 days after enrollment OR any stroke or death within 30 days of any revascularization procedure of the qualifying symptomatic intracranial artery done during follow-up, OR stroke in the territory of the symptomatic intracranial artery from day 31 after enrollment to completion of Phase 1) in high-risk patients * with intracranial stenosis. * patients with TIA or stroke within 30 days prior to enrollment that is attributed to 70% - 99% stenosis of a major intracranial artery (MCA, carotid, vertebral, basilar) December 3, 2010 page 8 of 42

10 Primary Hypothesis: Compared with intensive medical therapy alone, intracranial angioplasty and stenting combined with intensive medical therapy will decrease the risk of the primary endpoint by 35% over a mean follow-up of two years in high-risk patients with symptomatic stenosis of a major intracranial artery. 4. STUDY DESIGN This will be an investigator initiated and designed Phase III multicenter trial in which patients with TIA or non-disabling stroke within 30 days prior to enrollment that is caused by 70% - 99% stenosis of a major intracranial artery (MCA, carotid, vertebral, or basilar) will be randomized (1:1) at approximately 60 sites to: intensive medical therapy alone (aspirin 325 mg / day for entire follow-up, clopidogrel 75mg per day for 90 days after enrollment unless cardiologist recommends continuing clopidogrel beyond 90 days for a cardiac indication, and aggressive risk factor management primarily targeting blood pressure < 140 / 90 mm Hg (< 130 / 80 if diabetic) and LDL < 70 mg / dl) OR intracranial angioplasty and stenting using the Gateway balloon and Wingspan self-expanding nitinol stent (or any future FDA approved iterations of the balloon, stent, or the delivery systems) plus intensive medical therapy (aspirin 325 mg / day for entire follow-up, clopidogrel 75mg per day for 90 days after enrollment unless cardiologist recommends continuing clopidogrel beyond 90 days for a cardiac indication, and aggressive risk factor management primarily targeting blood pressure < 140 / 90 mm Hg (< 130 / 80 if diabetic) and LDL < 70 mg / dl). Risk factor management will be performed by the study neurologist at each site who will be assisted by an innovative, evidence-based, educational, lifestyle modification program (INTERxVENT) that will be administered at regularly scheduled times to all patients throughout the study. There are two follow-up phases to this study: Phase 1 includes all patients enrolled in the study and is complete after all active patients (i.e., those that have not already reached a primary endpoint, died, or withdrawn consent to participate in the trial) have been followed for at least 1 year after enrollment. The mean length of follow-up in Phase I is expected to be two years. The purpose of Phase 1 follow-up is to compare intensive medical therapy plus angioplasty and stenting with intensive medical therapy alone to determine which therapy is more effective at preventing the primary endpoint. Phase 2 begins at the conclusion of Phase 1 and includes only active patients in the stenting arm who have not yet completed three years of follow-up. When all active patients in the stenting arm have completed 3 years of follow-up, they will have completed their participation in the study. The purpose of Phase 2 is to determine the outcome of stented patients over a longer follow-up period. We anticipate Phase 1 will be completed in year 5 of the trial and Phase 2 will be completed 2 years later. All patients randomized to the medical arm who do not have a primary endpoint or die during the trial will be followed for a minimum of 1 year and a maximum of 3 years, whereas all patients December 3, 2010 page 9 of 42

11 randomized to the stenting arm who do not have a primary endpoint or die during the trial will be followed for 3 years. Once a patient in either arm has a primary endpoint, their study participation is considered complete and no more follow-up is required except for a follow-up visit at 90 days after a non-fatal primary endpoint to assess disability. 5. RATIONALE FOR ANTITHROMBOTIC THERAPY IN SAMMPRIS Stenting Arm: The standard of care for antithrombotic therapy following coronary or intracranial stenting is to use a combination of clopidogrel and aspirin for at least 28 days after the procedure. Accumulating evidence from coronary percutaneous intervention (PCI) trials in which bare metal stents were predominantly used suggests that the duration of combination therapy should be extended beyond 28 days 11,12. In the PCI-CURE study, treatment with either ticlopidine or clopidogrel in addition to aspirin beyond 28 days of PCI for a mean duration of 8 months was associated with a lower rate of cardiovascular death, MI, or any revascularization (p=0.03) and of cardiovascular death or MI (p=0.047) than treatment with aspirin alone beyond 28 days 11. In the CREDO trial, treatment with clopidogrel (75 mg per day) and aspirin beyond 28 days of PCI for 1 year was associated with a 26.9% relative reduction in the combined risk of death, MI, or stroke (95% CI 3.9% %; p = 0.02, absolute reduction, 3%) compared with treatment with aspirin alone beyond 28 days 12. Additionally, patients who received a 300 mg loading dose of clopidogrel at least 6 hours before PCI had a 38.6% relative reduction in the combined risk of death, MI, or urgent target vessel revascularization at 28 days (95% CI -1.6% %; p = 0.051) compared with patients who did not receive the loading dose. The results of these trials, coupled with results of MATCH and CHARISMA (two secondary prevention trials) that showed a divergence in the Kaplan-Meier major hemorrhage curves beyond 90 days in patients treated with aspirin and clopidogrel vs. either agent alone (higher with combination) 13-14, provides strong scientific evidence for using combination clopidogrel and aspirin for 90 days following stenting in SAMMPRIS. Of note, on polling potential interventionalists in SAMMPRIS, the most commonly used antithrombotic regimen after intracranial stenting is combination aspirin and clopidogrel for 90 days followed by aspirin alone. Medical Arm: Our choice of antithrombotic therapy for both arms were influenced by the following factors: 1. the importance of having the same medical therapy in both arms so that any difference in outcome between the two arms of the study could be ascribed to angioplasty plus stenting alone; 2. the available data on the use of various antithrombotic therapies in patients with intracranial stenosis, other stroke subtypes, and CAD which was present in 27% of patients in WASID; 3. the impact of the choice of antithrombotic therapy on withdrawal of therapy rates, recruitment, and cost of the trial. Based on these factors, our choice of antithrombotic therapy for the medical arm is aspirin 325 mg per day for the entire duration of the trial and clopidogrel for the first 90 days after enrollment. Besides balancing the antithrombotic therapy in both arms of the trial, the combination of aspirin and clopidogrel is supported by the results of MATCH and CHARISMA, both of which strongly suggested that combination therapy was more effective than monotherapy in patients enrolled within 30 days of their qualifying event (a requirement for enrollment in SAMMPRIS). Additionally, it is likely that the pathophysiology of recurrent stroke related to recently symptomatic intracranial large artery atherosclerosis is similar to that of MI related to recently symptomatic coronary atheroclerosis, which the CURE trial showed was more effectively treated with aspirin and clopidogrel than aspirin alone 15. Limiting the use of clopidogrel to 90 days is warranted because the highest risk period for recurrent stroke in WASID was the first three months after enrollment, and there was a significant increase in major hemorrhage from combination therapy beyond 90 days in MATCH and CHARISMA. Our choice of antithrombotic therapy for both arms of the trial satisfies the other selection criteria: aspirin is December 3, 2010 page 10 of 42

12 the only antiplatelet agent that has been evaluated for long term therapy in patients with intracranial stenosis, aspirin and clopidogrel have been shown to be effective for patients with CAD and in patients undergoing stenting, withdrawal of therapy rates are low on these agents, short term use of clopidogrel and long term use of aspirin are relatively inexpensive, and recruitment should be enhanced as patients who have their qualifying event on antithrombotic therapy (particularly aspirin) will be treated with a more powerful antithrombotic regimen during their highest risk period for recurrent stroke. Several factors were considered in choosing 325 mg per day of aspirin for both arms in SAMMPRIS rather than the 1300 mg per day dose chosen in WASID. These factors include: 1. the use of 1300 mg per day could have increased the risk of major hemorrhage in patients in both arms in the first 90 days when clopidogrel 75 mg per day is also required (few of the participating physicians in this trial would have agreed to such a protocol), 2. meta-analyses performed after WASID was initiated strongly suggest that high-dose aspirin offers no advantage over lower doses of aspirin for stroke prevention but does increase the risk of major gastrointestinal hemorrhage 16. These analyses have led the FDA and the American Stroke Association to recommend a maximum dose of 325 mg per day of aspirin for stroke prevention, 3. even though 1300 mg per day was the dose studied in WASID, a very small minority of physicians, including WASID investigators, are using that dose in practice. This suggests that very few study investigators would be comfortable using 1300 mg per day in SAMMPRIS and, regardless of the results of the trial, would choose 325 mg per day or less in practice (as they have after WASID). We also considered using combination aspirin and dipyridamole (rather than aspirin and clopidogrel) as the antiplatelet regimen in both arms of the trial but this was rejected because there are no data on the efficacy of this antiplatelet regimen in patients with intracranial stenosis, patients with CAD, or patients undergoing stenting. Moreover, withdrawal of therapy rates in patients treated with combination aspirin and dipyridamole in secondary prevention stroke trials is high (29% - 34%) 17,18. Consideration was also given to other antithrombotic protocols that allowed participating physicians to choose antiplatelet therapy either immediately after randomization or after a required 90 day period of aspirin and clopidogrel in both arms. In those scenarios, patients would be prescribed these antiplatelet therapies as part of routine care rather than the trial supplying the medications. These strategies were also rejected because it is likely that many study investigators would recommend continuing combination aspirin and clopidogrel or either agent alone beyond 90 days in patients randomized to stenting, whereas combined aspirin and dipyridamole would more likely be prescribed beyond 90 days in the medical arm. If this occurred, any difference in outcomes between the two arms could not be ascribed to stenting alone. Additionally, given that patients or their private payers would be responsible for paying for the antiplatelet therapy beyond 90 days, it is likely that patients in poorer socioeconomic circumstances would be prescribed aspirin whereas other patients in the trial may be more likely to be prescribed clopidogrel or the combination of aspirin and dipyridamole. 6. SAMPLE SIZE ESTIMATES The primary hypothesis for SAMMPRIS is that the addition of angioplasty plus stenting will reduce the occurrence of the primary endpoint over intensive medical management alone. The rate for the medical arm was derived from WASID patients that met the inclusion criteria for SAMMPRIS (different from WASID in the following: upper age restriction of 80 years, qualifying event within 30 days of enrollment, 70% - 99% stenosis, no tandem stenoses in 70%-99% December 3, 2010 page 11 of 42

13 range in SAMMPRIS). Among these 122 WASID patients, the rate for the primary endpoint was 29% at two years. Assuming that aggressive risk factor management in SAMMPRIS decreases the primary event rate by 15% over 2 years compared with usual risk factor management (used in WASID), the projected rate of the primary endpoint in the medical arm in SAMMPRIS is 24.7% at two years. In estimating a 15% reduction in the rate of the primary endpoint from aggressive risk factor management in SAMMPRIS, we were influenced by the risk factor data in WASID patients as well as the results of SPARCL and PROGRESS 19,20. The WASID risk factor data suggest that achieving both targets for blood pressure and LDL in all SAMMPRIS patients may lower the primary endpoint by 14% - 22% but these rates may be over-estimates because only 60% -75% of our patients are likely to achieve our blood pressure and LDL targets, the major impact of risk factor management is likely to occur well beyond the highest risk period in the trial (the first 3 months after enrollment), and post-hoc analyses typically overestimate treatment effects. We also considered the results of SPARCL and PROGRESS in estimating the benefit from aggressive risk factor management in SAMMPRIS. In SPARCL, which compared high-dose atorvastatin vs. placebo, there was a 16% reduction in stroke during a mean follow-up of 4.9 years from atorvastatin but review of the Kaplan-Meier curves indicates that the benefit at 2 years was substantially lower 19. In PROGRESS, which compared an ACE inhibitor with or without a diuretic vs. placebo, there was a 28% reduction in stroke over 4 years from antihypertensive therapy but the benefit at 2 years was only in the 15% range 20. Moreover, if these treatments had been compared to usual medical management rather than placebo, the benefits from these therapies may have been even smaller. For the reasons discussed above, our estimate of a 15% decrease in the primary endpoint rate at 2 years in SAMMPRIS from aggressive vs. usual risk factor management seems reasonable. The sample size calculation for SAMMPRIS was also based on detecting a 35% relative risk reduction in the primary endpoint with stenting, i.e., a reduction in the rate from 24.7% to 16.1% at 2 years. In deciding on a 35% relative risk reduction from stenting we were influenced by: i. the results of a survey we sent to 246 physicians (stroke neurologists and members of the American Society of Interventional and Therapeutic Neuroradiology) to determine the minimal benefit from stenting that would make it the treatment of choice of a majority of physicians participating in the survey. ii. a comparison of the numbers needed to treat (NNT) in SAMMPRIS vs. other completed stroke trials that compared surgical procedures with medical therapy (e.g. NASCET, ACAS). The survey results indicated that, for a primary endpoint rate of 25% over 2 years in the medical arm (the rate proposed in SAMMPRIS), 33% of respondents would require at least a 25% reduction from stenting, 17% would require at least a 33% reduction, 21% would require at least a 40% reduction, 14% would require at least a 50% reduction, and 10% would require at least a 60% reduction to make stenting their treatment of choice (the other 5% would continue medical therapy regardless of the benefit from stenting). Therefore, 33% was the minimum relative risk reduction required from stenting to make it the treatment of choice for 50% of physicians who participated in the survey. Thus, we would expect that a higher relative risk reduction (i.e., 35%, which is the hypothesized effect in this trial) will result in a majority of physicians choosing stenting as the treatment of choice for these patients. In choosing the 35% relative risk reduction required from stenting in this trial, we were also influenced by the absolute risk reduction and number needed to treat (NNT) with stenting to prevent one patient from having a primary endpoint over 2 years. For a reduction in the primary December 3, 2010 page 12 of 42

14 endpoint rate from 24.7% to 16.1% at 2 years, the absolute risk reduction is 8.6%, giving a NNT of 11.6 patients. This NNT is almost twice as high as the NNT for carotid endarterectomy in symptomatic patients with 70%-99% stenosis (NNT = 6 to prevent 1 stroke at 2 years) for which there is general agreement on the need for endarterectomy. Any higher NNT for stenting (or lower relative risk reduction than 35%) would raise questions about the widespread acceptance of stenting as the treatment of choice, as is the case for carotid endarterectomy in asymptomatic patients with 60%-99% stenosis, for which there is still debate about the need for surgery. Our sample size calculation included an adjustment for potential crossover of patients from medical therapy alone to stenting. We assumed that 5% of patients randomized to the medical arm would undergo stenting after 6 months of follow-up, on average. The adjustment for crossover was made by computing weighted averages of the hazard rates in the two arms as follows: The hazard rates for the two treatment groups were calculated from the event rates at two years assuming an exponential distribution for the time to event; Medical: λ M = -log e ( )/2 = , Stenting: λ S = -log e ( )/2 = The hazard rate for patients who crossover from the medical to stenting arm was determined as: λ MCS = 0.25λ M λ S = These weights were based on our assumption that of the two years of follow-up, 6 months would be spent under the medical arm hazard rate and 18 months would be spent under the stenting arm hazard rate. The adjusted hazard rate for the medical group was determined as: λ M* = 0.05λ MCS λ M = based on our assumption that 5% of the patients in the medical arm would undergo stenting. Thus, the hazard rates used for the sample size calculation for the two treatment groups were: Medical: , Stenting: giving a hazard ratio (medical / stenting) of This strategy for adjusting the hazard rates is similar to that proposed by Lachin and Foulkes 21 for adjusting for noncompliance to study medications. The test of the primary hypothesis will be done using a two-sided log-rank test. Formulas relating the hazard ratio, sample size, and probabilities of Type I and Type II errors for the logrank test can be found in Machin et al 22. These formulas are implemented in the software system Power Analysis and Sample Size (PASS) 23. With probability of Type I Error = 0.05 and power = 0.80, the required sample size was a total of 734 patients in both groups. Also using PASS, we incorporated an adjustment to the sample size for interim analyses for efficacy with an O Brien-Flemming 24 alpha spending function approach. Although the DSMB will ultimately determine the frequency of interim analyses for efficacy, for the purposes of sample size planning, we assumed 2 interim reviews before the end (when 33% and 66% of the anticipated events have occurred). This raised the required sample size to 748 patients. We also allowed for a 2% loss to follow-up, as observed in WASID, resulting in a sample size of 764 patients (382 per treatment group). 7. PRIMARY ENDPOINTS Any stroke (ischemic, parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage) or death within 30 days after enrollment, OR Any stroke (ischemic, parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage) or death within 30 days of any revascularization procedure of the qualifying symptomatic intracranial artery, OR an ischemic stroke in the territory of the symptomatic intracranial artery from day 31 after enrollment to completion of Phase 1. December 3, 2010 page 13 of 42

15 Stroke is diagnosed if the patient develops either: 1. a neurological deficit that is thought to have an ischemic cause and is detectable on bedside examination at least 24 hours after onset 2. a symptomatic intracerebral, subarachnoid, or primary intraventricular hemorrhage Brain imaging is required to distinguish between ischemic and hemorrhagic causes of stroke. For detailed definitions of ischemic stroke (in and out of the territory), parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage, see the Manual of Operations. 8. SECONDARY ENDPOINTS Clinical Secondary Endpoints Ischemic stroke outside of the territory of the symptomatic artery beyond 30 days after enrollment any parenchymal brain hemorrhage, subarachnoid or intraventricular hemorrhage beyond 30 days after enrollment myocardial infarction major non-stroke hemorrhage (i.e., any subdural hemorrhage, any epidural hemorrhage, major systemic hemorrhage) any death beyond 30 days after enrollment disabling stroke any stroke or death any serious adverse event (see Section 17) functional outcome at 1 year as measured by the Rankin Scale and Barthel Index cognitive outcome at 4 months (safety endpoint), 1 year, and end of Phase I as measured by Montreal Cognitive Assessment (MoCA) The definitions for these endpoints are provided in the Manual of Operations. Technical Secondary Endpoints Balloon success is defined as accessing the lesion and inflating the balloon within the target lesion. Stent success is defined by deployment of the stent within the target lesion with less than 50% residual stenosis December 3, 2010 page 14 of 42

16 Procedural success is defined by achieving balloon and stent success without stroke or death at discharge or within 72 hours post-procedure, whichever is sooner 9. PATIENT SELECTION CRITERIA INCLUSION CRITERIA 1. TIA or non-severe stroke within 30 days of enrollment attributed to 70-99% stenosis of a major intracranial artery (carotid artery, MCA stem (M1), vertebral artery, or basilar artery) may be diagnosed by TCD, MRA, or CTA to qualify for angiogram performed as part of the study protocol but must be confirmed by catheter angiography for enrollment in the trial 2. Modified Rankin score of 3 3. Target area of stenosis in an intracranial artery that has a normal diameter of 2.00 mm to 4.50 mm 4. Target area of stenosis is less than or equal to 14 mm in length 5. Age 30 years and 80 years. Patients years are required to meet at least one additional criteria (i-vi) provided in the table below to qualify for the study. This additional requirement is to increase the likelihood that the symptomatic intracranial stenosis in patients years is atherosclerotic. i. insulin dependent diabetes for at least 15 years ii. at least 2 of the following atherosclerotic risk factors: hypertension (BP > 140/90 or on antihypertensive therapy); dyslipidemia (LDL > 130 mg /dl or HDL < 40 mg/dl or fasting triglycerides > 150 mg/dl or on lipid lowering therapy); smoking; non-insulin dependent diabetes or insulin dependent diabetes of less than 15 years duration; family history of any of the following: myocardial infarction, coronary artery bypass, coronary angioplasty or stenting, stroke, carotid endarterectomy or stenting, peripheral vascular surgery in parent or sibling who was < 55 years of age for men or < 65 for women at the time of the event ii. history of any of the following: myocardial infarction, coronary artery bypass, coronary angioplasty or stenting, carotid endarterectomy or stenting, or peripheral vascular surgery for atherosclerotic disease iv. any stenosis of an extracranial carotid or vertebral artery, another intracranial artery, subclavian artery, coronary artery, iliac or femoral artery, other lower or upper extremity artery, mesenteric artery, or renal artery that was documented by non-invasive vascular imaging or catheter angiography and is considered atherosclerotic v. aortic arch atheroma documented by non-invasive vascular imaging or catheter angiography vi. any aortic aneurysm documented by non-invasive vascular imaging or catheter angiography that is considered atherosclerotic 6. Negative pregnancy test in a female who has had any menses in the last 18 months 7. Patient is willing and able to return for all follow-up visits required by the protocol December 3, 2010 page 15 of 42

17 8. Patient is available by phone 9. Patient understands the purpose and requirements of the study, can make him/herself understood, and has provided informed consent EXCLUSION CRITERIA 1. Tandem extracranial or intracranial stenosis (70%-99%) or occlusion that is proximal or distal to the target intracranial lesion (NOTE: an exception is allowed if the occlusion involves a single vertebral artery proximal to a symptomatic basilar artery stenosis and the contralateral vertebral artery is supplying the basilar artery) 2. Bilateral intracranial vertebral artery stenosis of 70%-99% and uncertainty about which artery is symptomatic (e.g. if patient has pontine, midbrain, or temporal occipital symptoms) 3. Stenting, angioplasty, or endarterectomy of an extracranial (carotid or vertebral artery) or intracranial artery within 30 days prior to expected enrollment date 4. Previous treatment of target lesion with a stent, angioplasty, or other mechanical device, or plan to perform staged angioplasty followed by stenting of target lesion 5. Plan to perform concomitant angioplasty or stenting of an extracranial vessel tandem to an intracranial stenosis 6. Presence of intraluminal thrombus proximal to or at the target lesion 7. Any aneurysm proximal to or distal to stenotic intracranial artery 8. Intracranial tumor (except meningioma) or any intracranial vascular malformation 9. CT or angiographic evidence of severe calcification at target lesion 10. Thrombolytic therapy within 24 hours prior to enrollment 11. Progressive neurological signs within 24 hours prior to enrollment 12. Brain infarct within previous 30 days of enrollment that is of sufficient size (> 5 cms) to be at risk of hemorrhagic conversion during or after stenting 13. Any hemorrhagic infarct within 14 days prior to enrollment 14. Any hemorrhagic infarct within days that is associated with mass effect 15. Any history of a primary intracerebral (parenchymal) hemorrhage (ICH) 16. Any other intracranial hemorrhage (subarachnoid, subdural, epidural) within 30 days 17. Any untreated chronic subdural hematoma of greater than 5 mm in thickness December 3, 2010 page 16 of 42

18 18. Intracranial arterial stenosis due to arterial dissection, Moya Moya disease; any known vasculitic disease; herpes zoster, varicella zoster or other viral vasculopathy; neurosyphilis; any other intracranial infection; any intracranial stenosis associated with CSF pleocytosis; radiation induced vasculopathy; fibromuscular dysplasia; sickle cell disease; neurofibromatosis; benign angiopathy of central nervous system; post-partum angiopathy; suspected vasospastic process, suspected recanalized embolus 19. Presence of any of the following unequivocal cardiac sources of embolism: chronic or paroxysmal atrial fibrillation, mitral stenosis, mechanical valve, endocarditis, intracardiac clot or vegetation, myocardial infarction within three months, dilated cardiomyopathy, left atrial spontaneous echo contrast, ejection fraction less than 30% 20. Known allergy or contraindication to aspirin, clopidogrel, heparin, nitinol, local or general anesthesia 21. History of life-threatening allergy to contrast dye. If not life threatening and can be effectively pretreated, patient can be enrolled at physician s discretion 22. Active peptic ulcer disease, major systemic hemorrhage within 30 days, active bleeding diathesis, platelets < 100,000, hematocrit < 30, INR > 1.5, clotting factor abnormality that increases the risk of bleeding, current alcohol or substance abuse, uncontrolled severe hypertension (systolic pressure > 180 mm Hg or diastolic pressure > 115 mm Hg), severe liver impairment (AST or ALT > 3 x normal, cirrhosis), creatinine > 3.0 (unless on dialysis) 23. Major surgery (including open femoral, aortic, or carotid surgery) within previous 30 days or planned in the next 90 days after enrollment 24. Indication for warfarin or heparin beyond enrollment (NOTE: exceptions allowed for use of systemic heparin during stenting procedure or subcutaneous heparin for deep vein thrombosis (DVT) prophylaxis while hospitalized) 25. Severe neurological deficit that renders the patient incapable of living independently 26. Dementia or psychiatric problem that prevents the patient from following an outpatient program reliably 27. Co-morbid conditions that may limit survival to less than 3 years 28. Pregnancy or of childbearing potential and unwilling to use contraception for the duration of this study 29. Enrollment in another study that would conflict with the current study 10. INFORMED CONSENT The principles of Informed Consent, according to FDA Regulations and ICH guidelines on GCP, will be followed. Each Investigator will submit a copy of the proposed consent form, together with the study protocol, to the appropriate IRB for approval. All patients must provide informed consent to participate and only the participant can provide informed consent, which must be cosigned by a study investigator. December 3, 2010 page 17 of 42

19 11. ENROLLMENT OF PATIENTS Patients will be enrolled in one of two ways: 1. Patients who meet the entrance criteria and have already undergone catheter angiography showing 70% - 99% intracranial stenosis will be asked to participate and sign consent. Once this is completed, the patient will be enrolled and, if randomized to stenting, will undergo stenting as soon as possible after enrollment (within 3 business days). 2. Patients who meet the entrance criteria, have only had a TCD, MRA, CTA suggesting 70% - 99% intracranial stenosis, and whose treating physicians would not ordinarily recommend diagnostic angiography as part of their usual practice will sign consent to i. undergo a single vessel diagnostic angiogram as part of the study protocol (paid for by the grant), and ii. agree to be randomized if the angiogram shows 70% - 99% stenosis. Once the consent form is signed, the site coordinator will call the project manager at the Clinical Coordinating Center to get permission to schedule a diagnostic angiogram paid for by the grant. The reason for requiring patients to agree to randomization before this diagnostic angiogram is to limit selection bias (e.g., if patients only consented to randomization after the angiogram, patients perceived as having a particularly high-risk lesion that should be stented in the opinion of the treating physicians may be excluded). We used exactly the same strategy for study paid angiograms in WASID to eliminate selection bias and to help with recruitment. If the study angiogram shows 70%-99% stenosis and the patient is randomized to the stenting arm, the stenting procedure can either be done immediately after the angiogram (will require anesthesia team to be on standby) or scheduled within 1-3 business days (at those sites where anesthesia needs to be pre-scheduled). Patients who undergo an angiogram as part of the study protocol and who have < 70% stenosis will not be enrolled in the trial but will be followed for 24 hours to document any complications of angiography. Patients are considered enrolled in the trial once they are randomized. For patients randomized to stenting, if the angiogram that is done as part of the interventional procedure (e.g., to size the stent) is measured at 50-69% stenosis (rather than the 70-99% measurement obtained on the qualifying angiogram), the patient will still undergo stenting. However, if the stenosis is < 50% or the artery is occluded, stenting will not be performed. This strategy, which complies with the HDE angiographic criteria for the use of Wingspan, will ensure that the small number of patients in both arms with 50-69% stenosis because the qualifying angiogram overestimated the degree of stenosis will still get the treatments mandated by the study. For a flow chart of the screening and enrollment process see Appendix 1. For a flowchart of the pathways to randomization and stenting see appendix 2. Refer to the Manual of Operations for more detailed information on the enrollment process. 12. RANDOMIZATION In any of the two scenarios described in 11 above that patients could be enrolled, a call will be made by the site coordinator to the Project Manager at the CCC to randomize the patient. In Scenario 1 the call will take place once the site coordinator has determined that the patient meets all eligibility criteria and the patient has signed informed consent. In Scenario 2 the call will take place once the diagnostic angiogram has been read and the stenosis is found to be 70-99%. The Project Manager will verify that the patient meets all eligibility criteria. If the patient is eligible, the Project Manager completes the on-line enrollment form and the site coordinator is automatically ed the treatment assignment. If any aspect of the on-line process does not December 3, 2010 page 18 of 42

20 function, Statistical Coordinating Center staff are available by cell phone as back-up. Please refer to the Manual of Operations for more detailed information on the randomization process. 13. ANGIOGRAPHIC EVALUATION AND LESION MEASUREMENT Efforts to ensure the quality of the angiographic images in the trial include choosing sites that have access to state-of-the-art cerebral angiography facilities, and providing training on image acquisition and measurement techniques at an investigators meeting to be held during the startup phase of the trial. The images showing the most severe diameter stenosis and longest length of stenosis will be chosen for measuring the stenotic diameter (mms) and length of lesion (mms). To measure the percent diameter stenosis, the study interventionalist will measure the diameters of the stenosis (Ds) of the target lesion and the reference normal vessel (Dn) according to the WASID rules for establishing where to measure Dn 25 (see Manual of Operations for more details on measurement technique). The actual measurements by the interventionalists will be made by using 2-dimensional quantitative angiography software available on all modern angiographic equipment. By placing an electronic cursor on the margins of the vessel, the program provides the measurements required. By using the formula (1 [D s / Dn]) x 100, the percent diameter stenosis of the target lesion will be calculated. Length of stenosis will be measured in millimeters from the proximal to distal shoulder of the lesion (more precisely, from the most distal part of the vessel with parallel margins just proximal to the lesion to the most proximal part of the vessel with parallel margins just distal to the lesion). If a patient undergoes stenting, two orthogonal images of the artery post angioplasty and post stent placement must be obtained for central review. Technical success of the procedure is defined by deployment of the stent at the target lesion with full coverage of the lesion and less than 50% residual stenosis. When a patient is enrolled in the trial, CD copies of angiographic images showing the most severe diameter stenosis and the longest length of stenosis and the post-angioplasty and post-stent images for the interventional arm will be sent for central reading (see manual of operations for more details of central reading of angiograms). The percent diameter stenosis and the length of stenosis will be measured using the WASID technique 25. These readings, which will be the final readings in the trial, will be done blinded to the site readings. Based on the intent- to-treat principle, patients whose central qualifying angiogram readings indicate < 70% intracranial arterial stenosis will still be included in the intention-to-treat analysis. 14. STENTING PROCEDURE Antithrombotic Protocol Pre-procedure: All patients who are scheduled to undergo stenting as part of the study protocol must have been pretreated with: 1. aspirin 325 mg per day (enteric coated) x 5 days if possible, but minimally 325 mg within 24 hrs before the procedure, and 2. clopidogrel 75 mg per day x 5 days or loading dose of 600 mg at least 6 hrs before procedure (note: if the patient has been on clopidogrel 75 mg per day for less than 5 days, then a loading dose of 600 mg must be used 6 24 hrs before the procedure). December 3, 2010 page 19 of 42

21 Since stenting must be performed within 3 business days of randomization in patients in the stenting arm, there are various possible scenarios for the use of clopidogrel pre-procedure: Example 1. Patient has been on clopidogrel 75 mg per day for > 3 days prior to randomization and stenting is scheduled on day 2 after randomization. Clopidogrel 75 mg per day would be given daily after randomization, i.e. on day 0 (day of randomization), day 1, day 2 (day of stenting) and no loading dose should be used. Example 2. Patient has never been on clopidogrel prior to randomization and stenting is scheduled on day 3 after randomization. Clopidogrel 75 mg per day would be given daily on day 0 (day of randomization), day 1, and day 2, and a loading dose of clopidogrel 600 mg must be given > 6 hrs before the procedure on day 3. Intra-procedure Heparin: A bolus of intravenous heparin that is weight adjusted must be given to maintain the activated clotting time (ACT) between seconds throughout the procedure. It is recommended that the ACT be checked during the procedure at intervals of no longer than 60 minutes, as well as at the end of the procedure to verify conformity with this requirement. IIb / IIIa agents can only be used if intraluminal thrombosis develops during the procedure. If a GP IIb-IIIa Inhibitor is given during the procedure, then the ACT should be maintained at less than 220 seconds to minimize the risk of bleeding. Post procedure 1. Aspirin 325 mg per day (enteric coated) must be used for the entire duration of follow-up (includes Phase I and Phase 2, if applicable), and 2. Clopidogrel 75 mg per day must be used for 90 days after randomization in both treatment arms. Only patients with coronary disease or peripheral vascular disease who have been prescribed long term use of clopidogrel by non-study physicians will be permitted to take clopidogrel beyond 90 days of enrollment. Both aspirin and clopidogrel (for 90 days after enrollment) will be supplied to the patients to help maximize compliance. Anesthesia / Arterial Access The stenting procedure must be performed under general anesthesia to prevent the patient from moving during the procedure. The brachial artery, radial artery or femoral artery approach may be used, according to the interventionalist s preferences. Procedure Steps Angiographic Assessment of Lesion and Stent Selection 1. Using angiography, determine the location and size of the lesion and vessel diameter. Careful stent sizing is important for successful stenting. In general, the stent size should be chosen to match the normal vessel diameter adjacent to the lesion. 2. Select a stent length that is at least 6 mm longer than the lesion to extend a minimum of 3 mm on both sides of the lesion. Stent sizing guidelines for each stent diameter are given in Table Select a balloon size to match the lesion length. Select a balloon diameter that at nominal inflation will not exceed, and should be less than, the reference vessel diameter (proximal or December 3, 2010 page 20 of 42

22 distal to the lesion, whichever is smaller), (See study PTA Balloon Catheter Directions for Use.) Labeled Stent Diameter 2.5 mm 3.0 mm 3.5 mm 4.0 mm 4.5 mm Labeled Stent Length 1 (mm) 9 mm 15 mm 20 mm 9 mm 15 mm 20 mm 9 mm 15 mm 20 mm 9 mm 15 mm 20 mm 9 mm 15 mm 20 mm TABLE 1 Wingspan Stent System Recommended Sizing Guidelines Self - Expanded Stent Diameter 2 Recommended Vessel Diameter 3 (mm) 2.8 mm >2.0 to mm >2.5 to mm >3.0 to mm >3.5 to mm >4.0 to 4.5 Delivery System Useable Length Maximum Guidewire Diameter Minimum Guide Catheter ID 135 cm in in 1 Select a Stent length that is at least 6mm longer than the lesion to extend a minimum of 3mm on both sides of the lesion. 2 Stent will not expand beyond the self-expanding diameter. 3 Select a Stent diameter based both on the sizing recommendations in this table and on the larger vessel diameter (proximal or distal reference vessel diameter). Wingspan Stent System Preparation 1. Open the pouch to remove the packaging tray, and inspect for compromised packaging. 2. Flush the dispenser hoop with sterile heparinized saline, carefully pull out the proximal hub assemblies from tray, tighten the rotating hemostasis valve onto the Inner Body, and remove the delivery system. Inspect delivery system for damage, such as kinks. The stent should be preloaded into the distal tip of the delivery system. 3. Connect a rotating hemostasis valve to the hub of the Delivery System Inner Body, and flush the lumen of the Delivery System Inner Body with sterile, heparinized saline. 4. Loosen the Delivery System Outer Body rotating hemostasis valve, flush the Delivery System Outer Body with heparinized saline, and tighten the hemostasis valve onto the Delivery System Inner Body. 5. Continue to flush the Delivery System Outer Body to purge air from the system. 6. Connect the hemostasis valve side port of the Delivery System Outer Body and Delivery System Inner Body to a pressurized sterile heparinized saline flush. December 3, 2010 page 21 of 42

23 7. Loosen the hemostasis valve on the Delivery System Outer Body that is locked onto the Delivery System Inner Body, and gently retract the Delivery System Inner Body so that there is a 1-2 mm gap between the proximal end of the dual tapered tip and the distal end of the Outer Body. This should result in a rapid saline drip from the Outer Body tip. Note: Do not use excessive force or lodge the Inner Body tip inside the Delivery System. 8. Tighten the Delivery System Outer Body hemostasis valve around the Delivery System Inner Body to hold the Delivery System Inner Body in place during advancement of the Wingspan Stent System. Study PTA Balloon Catheter Preparation 1. Prepare the study PTA balloon catheter as outlined in its Directions for Use. Guidewire Positioning 1. Position an access guidewire across the lesion using standard microcatheter and guidewire techniques. Recommended guide catheter specifications include a minimum 90cm length and 1.63 mm (0.064in) ID. 2. Replace the access guidewire with an exchange length 0.36 mm (0.014in) guidewire, and remove the microcatheter. Leave the exchange length guidewire across the lesion. Soft guidewires are recommended rather than support guidewires. Balloon Deployment 1. Insert the study PTA balloon catheter over the guidewire and pre-dilate the lesion with the balloon as described in the PTA balloon catheter Directions for Use. Ensure that Balloon inflation does not exceed, and should be less than, the reference vessel diameter (proximal or distal to the lesion, whichever is smaller). Once pre-dilation has been completed, two orthogonal images of the lesion must be obtained for central review. Stent Positioning and Deployment 1. Carefully backload the Wingspan Stent System onto the 0.36 mm (0.014in) guidewire through the delivery system. 2. Carefully advance the Wingspan Stent System into the guide catheter. 3. Open the guide catheter hemostasis valve. Under fluoroscopic guidance, advance the Wingspan Stent System over the guidewire until the stent is slightly distal to the target lesion site (use the four distal radiopaque markerbands to identify the stent position). See Figure 1, Delivery Catheter Advancement. December 3, 2010 page 22 of 42

24 Guidewire Artery Wall Plaque Stent Distal Markerbands Stent Guidewire Stent Distal Markerbands Stent Fig 1 Delivery Catheter Advancement Fig 2 Stent Pre-Deployment Alignment 4. Loosen the Delivery System Outer Body rotating hemostasis valve, and advance the Delivery System Inner Body until the proximal radiopaque markerband bumper is just proximal to the stent. Tighten the Outer Body rotating hemostasis valve. 5. Slightly withdraw the hub of the Delivery System Outer Body until the stent is directly aligned with the target lesion site. Pull back on the Delivery System to make the final adjustment for stent positioning. This will ensure that slack has been removed from the Delivery System just prior to deployment. See Figure 2, Stent Pre-Deployment Alignment. 6. The stent is now ready to be deployed. NOTE: The best fluoroscopic view for positioning the stent for deployment is the view that shows the vessel distal to the lesion. This view may not be the same view as that used as the working position for stent deployment. Artery Wall Plaque Guidewire Stent Artery Wall Plaque Stent Delivery Catheter Fig. 3 Stent Positioning Fig. 4 Deployed Stent 7. Loosen the rotating hemostasis valve on the Delivery System Outer Body. Deploy the stent by holding the Delivery System Inner Body Hub stationary with one hand while continuing to carefully withdraw the hub of the Delivery System Outer Body hub with the other hand. This will deploy the Stent. See Figure 3, Stent Positioning. 8. As the stent deploys, you will see the markerbands on the distal end of the stent spread out from one another. This is the stent opening. Continue deploying the stent in a continuously smooth motion. Do not attempt to move the stent after deployment has begun. Be careful to not advance the Delivery System Outer Body as the stent is deploying. December 3, 2010 page 23 of 42