The Asymptomatic Carotid Surgery Trial-2 (ACST-2): an ongoing randomised controlled trial comparing carotid endarterectomy with carotid artery stenting to prevent stroke

Notes

Permissions

Chapter 1 Rationale and study design

The second Asymptomatic Carotid Surgery Trial (ACST-2) is a pragmatic international multicentre randomised controlled trial (RCT) that directly compares carotid surgery with carotid artery stenting (CAS). It includes patients thought to definitely need an intervention for asymptomatic carotid stenosis, but in whom there is substantial uncertainty as to whether or not to opt for carotid endarterectomy (CEA) or CAS. The ACST-2 seeks to randomise such individuals to either CEA or CAS to compare both the immediate hazards of the two procedures when carried out by experienced doctors with an approved track record and the long-term durability of protection against stroke conferred by both procedures. The trial will also collect the stroke rates in the patients over 5–10 years of follow-up, with follow-ups planned until 2025. The trial seeks to recruit any patient with a carotid lesion that has not recently (i.e. within 6 months) caused any symptoms (i.e. an asymptomatic lesion), who would be expected to benefit from a carotid procedure to reduce the risk of future stroke, and in whom there is clinical uncertainty as to which is the preferred treatment.

Potentially eligible trial participants should already be on suitable drug therapy for stroke prevention and be likely to live for at least 5 years, giving them long enough to benefit from a stroke prevention procedure (CEA or CAS). Prior to trial entry, non-invasive arch angiography [computed tomography (CT) or magnetic resonance imaging (MRI)] is also undertaken to ensure suitability for both procedures. These tests are routinely carried out before CAS and commonly, but not invariably, prior to CEA.

Randomisation of patients is by 1 : 1 allocation to CEA or CAS using a minimisation algorithm to ensure that both groups are well matched for key baseline prognostic factors that may determine early and long-term stroke risk. After intervention, patients are neurologically reassessed, including duplex scanning, and their drug treatment is adjusted if necessary before discharge. Early patient experience with both treatments is similar apart from the discomfort and wound care that is associated with open surgery, and the use of general versus local anaesthetic, which is determined by the centre’s standard of care (see Appendix 1).

Chapter 2 Recruitment

An international collaborative network has been established and comprises doctors across 27 countries. The network has been investigated and approved by our Technical Management Committee and by their local ethics and research committees. By March 2016, 2125 patients had been recruited from 110 centres worldwide. Recruitment pathways in the trial are similar to those used in routine clinical practice.

The ACST-2 has been designed to minimise the burden of research on the collaborators. The initial patient assessment determines that they are on suitable drug therapy for stroke prevention and that the patient is likely to live for at least 5 years, giving them long enough to benefit from a stroke prevention procedure (CEA or CAS). Prior to trial entry, non-invasive MRI or CT arch angiography is undertaken to ensure suitability for both procedures; this is usually done before CAS and commonly, but not invariably, done prior to CEA, and hence can be readily integrated into the participant’s care pathway.

The usual CAS procedure involves passing a guide wire from the femoral artery up to the aortic arch to gain access to the carotid artery. In some cases the anatomy of the aortic arch can be sufficiently tortuous to make CAS more difficult or hazardous; pre-randomisation aortic arch imaging with MRI or CT ensures that CAS is likely to be technically feasible. In routine practice, a duplex Doppler scan is often performed 1 month after the procedure to check the artery is open. Therefore, such a scan is used in this trial to check patency of the carotid artery 1 month after intervention, thereby confirming technical success. At least one clinical follow-up in outpatients is routine care for all carotid interventions and this is usually done 1 month after the procedure in this trial.

Within the UK we used a novel hub-and-spoke recruitment model, which allowed centres (i.e. the ‘spoke’) that offered one, but not both, treatments (usually CEA but not CAS) to identify patients who were eligible to enter the trial. The patients were then assessed (including use of MRI or CT) and then, after randomisation and depending on the treatment allocated, treated locally using CEA or sent for CAS to the ‘hub’ hospital (see Appendix 2).

One-third of European carotid procedures are performed in Italy, one-third are performed in Germany and the remaining one-third are performed in other European countries. Although Italy is the top recruiting country in the ACST-2, the UK is the second highest recruiter, accounting for about 20% of randomised patients (Figures 1 and 2).

FIGURE 1.

Recruitment per country (to the end of December 2015).

FIGURE 2.

Total recruitment by month (March 2010 to December 2015). Number of centres that ever recruited a patient and cumulative recruitment.

Chapter 3 Data collection

Randomisation is carried out by a telephone randomisation service (24-hour freephone number) or via a password-protected website via the internet. The collaborator is informed of the allocated treatment and the participant is ascribed a unique patient identifier number. The collaborator then either faxes or posts the randomisation form and the signed consent form to the ACST-2 trial office, which is based in the Nuffield Department of Surgical Science, University of Oxford, Oxford, UK. Procedural and post-procedure data are subsequently collected on a 1-month follow-up form completed by the collaborator and returned to the ACST-2 trial office. Data from these forms are entered on a trial database, which is held on secure servers on behalf of ACST-2 at the Clinical Trials Service Unit, University of Oxford, Oxford, UK [that have worked with us in designing and carrying out much of the work in the first Asymptomatic Carotid Surgery Trial (ACST-1) and ACST-2].

Annual follow-up of the patients in the trial is co-ordinated by the central ACST-2 office and annual questionnaires are sent either directly to the patient or to the collaborator, depending on local agreements.

Chapter 4 Patient and public involvement

The aims and design of ACST-2 have been discussed with the Oxford Clinical Trial Service Unit (CTSU) patient focus group, which was established in 2012 to allow members of the public who have, or are at risk of, vascular disease to criticise and help evaluate ongoing and planned studies conducted by the CTSU.

Mr David Simpson is the lay member on the ACST-2 Trial Steering Committee. He was closely involved in study design, drafting the trial protocol and original patient information leaflet, and eloquently represents the public interest, both formally at annual Trial Steering Committee meetings and informally throughout the years. His role is ongoing as the trial continues.

The ACST-2 trial was adopted by the Stroke Research Network as soon as funding was confirmed by Health Technology Assessment (HTA) programme. This helped us with local contacts, meetings, ethics approval applications, recruitment (of the planned 20% of UK patients) and follow-up in the UK. Annual Stroke Network meetings in Newcastle and London were particularly helpful by enabling us to give platform presentations of our work and discuss it with attendees from potential new centres in front of our posters. We also attended, presented and had stands at the Thames Cardiovascular Network Group meetings in 2013 and 2014, the Vascular Society (the UK National Society for Vascular Surgery) and the British Society of Interventional Radiology (for UK interventional radiologists). The annual UK Stroke Forum was also important; attendees came from every stroke care discipline and included stroke sufferers and representatives from patient groups. We have had a stand at most of these meetings since the trial’s inception, winning a prize for our novel UK hub-and-spoke recruitment model at the UK Stroke Forum.

Chapter 5 Economic evaluation and quality of life

The design of ACST-2 includes a health economic component with evaluation of resource use during treatment and follow-up. The use of UK data will be particularly relevant to the NHS because much of the evidence on costs of CEA and, particularly, CAS has been based on evidence from studies outside the UK. The main components for current and future study will be (1) initial procedural costs, (2) short-term retreatment costs (repeat or further procedures within 1 month), (3) costs of any MIs and strokes within the first month and (4) the costs of any strokes after the first month, almost all of which will not be procedural. Duration of hospital stay is also recorded. At annual follow-ups with the patients, we seek information about strokes or carotid procedures beyond the first month and standard costs will be assumed for these procedures. Economic analyses will evaluate stroke-related quality of life at 1 month after the trial procedures as well as short- and longer-term stroke outcomes and costs.

Annual follow-up questionnaires sent by the trial office to the patient are used to collect data on whether or not the patient has had a subsequent stroke or further treatment on their carotid arteries. In addition, we have extended collection of the standard EuroQol-5 Dimensions (EQ-5D^®) from the UK alone to five more countries (Belgium, Germany, Italy, Serbia and Sweden), accounting for 85% of the patients recruited so far. All stroke patients will be and have been asked annually how their stroke still affects them. Their current medication, including names and dosage of all blood pressure, antithrombotic and lipid-lowering drugs prescribed, has been recorded for analysis.

Chapter 6 Interim blinded results

Unblinded results for the trial will be reported after patient recruitment is complete (and, if recruitment continues at the present rate, the target of 3600 participants will be reached by December 2019) and this report is planned for 2021. Long-term follow-up (which is less onerous and usually by direct patient letter with confirmation of any strokes through the participating physician) will continue until December 2025, after which a final publication will report long-term results in late 2026.

Data on the baseline characteristics of participants recruited to date are available (Table 1). Around one-quarter of participants are > 75 years old and around one-third are female. Diabetes mellitus is more common in ACST-2 than in ACST-1 (one-third of ACST-2 participants have diabetes mellitus), and almost half of the participants had prior evidence of stroke (clinically evident stroke or a silent infarction detected on pre-procedural cross-sectional brain imaging), leaving them at higher stroke risk without appropriate carotid intervention. Owing (in part) to the minimisation algorithm used during randomisation, these characteristics are similar in patients randomised to stenting and surgery.

The use of triple medical therapy is excellent at baseline and maintained or improved during long-term follow-up (Tables 2 and 3). Predictably, CAS is associated with a much higher rate of dual antiplatelet therapy at 1 month post procedure than with CEA, but this difference largely disappears with longer-term follow-up. Over 50% of patients are receiving either rosuvastatin or atorvastatin, with one-quarter taking simvastatin.

Compliance: because the trial is ongoing, final compliance is not available. As of March 2016, the overall crossover rate was 3.6% and a further 5.3% of patients were still awaiting intervention, but this number will likely reduce with time (Table 4). The mean time from randomisation to treatment was similar for both CEA (23 days) and CAS (26 days).

Techniques: similar numbers of CEA were performed under general anaesthesia (56%) and local/regional anaesthesia (44%), but the majority of carotid stents were performed under local anaesthetic (94%). Carotid patching was used in 44% of patients undergoing CEA and 22% of CEA patients were shunted. For CAS, eight types of stent were used (46% tapered) and WALLSTENT™ (Boston Scientific Corporation, Marlborough, MA, USA) was the most commonly used device. Cerebral protection devices were used for 86% of CAS (Table 5) and eight types of cerebral protection device were employed (79% filters, 10% proximal systems), including flow reversal and flow arrest systems (see Table 5).

The ACST-2 provides yearly reports to the independent Data Monitoring Committee of confirmed strokes, MI and deaths within 1 month of treatment and of long-term stroke rates. With their consent, in our first 1000 patients we have reported an overall (i.e. blinded) 30-day rate of disabling stroke or death of around 1%, which compares favourably with that seen for CEA alone in ACST-1 (1.7%), confirming that ACST-2 collaborators are performing trial procedures to a high standard.

Chapter 7 Statistical analysis

No material difference in fatal or disabling procedural events is expected (≈1% in each group), so power calculations for this outcome are not given. The primary outcome of particular interest is the stroke risk in the period > 30 days post procedure, but the rates > 5 years post procedure will provide an important subgroup analysis. Accordingly, person-years accrued are more relevant than numbers of patients randomised. For the main outcome of the annual stroke rate after day 30 (i.e. after the end of the perioperative period), ACST-2 will have 18,000 person-years in its first report (2021) and 36,000 by December 2025.

If mature results from ACST-2 show little difference in long-term stroke rates, this key result will be established reliably in the first report and even more reliably by the time of the final report (2026). However, there could be important absolute differences in long-term stroke rates. Suppose, for example, that the stroke rate per decade is 6% after CEA and 9% after CAS, a clinically meaningful difference (with stroke rate ratio = 0.67, easily compatible with the results from the previous trials). Then, with 18,000 person-years by mid-2020, the stroke rate would have about a 70% chance of getting a p-value < 0.05 and a 50% chance of getting a p-value < 0.01. However, with 36,000 person-years (as of December 2025), it would have a 93% chance of getting a p-value < 0.05 and an 82% chance of getting a p-value < 0.01. Moreover, its expected result of 120 versus 180 strokes would ensure high significance (p = 0.002) in a meta-analysis combining it with the apparently null final results from the previous (small) trials.

Chapter 8 Discussion

Almost 200,000 carotid procedures (surgery or stenting) are performed annually, commonly on asymptomatic patients with carotid stenosis, although numbers in the UK are presently lower than in some other European countries. Regardless of how many should be performed here or elsewhere, as long as such procedures continue to be performed widely, large-scale randomised evidence directly comparing surgery with stenting is needed.

Among 1990s-era asymptomatic patients who were on triple-drug therapy (blood pressure-lowering, lipid-lowering and antithrombotic treatment) in the ACST-1 trial of carotid surgery compared with those who were not, there was net benefit from surgery despite the protective effects of the triple therapy. 1 Two new trials, ECST-2 (European Carotid Surgery Trial) and CREST-2 (Carotid Revascularization Endarterectomy versus Stenting Trial) (surgery vs. no surgery in Europe, surgery vs. no surgery in the USA, stenting vs. no stenting in Europe and stenting vs. no stenting in the USA) are currently recruiting to determine whether or not, for the 2010s era triple therapy asymptomatic patients, carotid procedures are still of net benefit. 2,3 If these trials confirm by the early 2020s that such procedures are of net benefit, then this will greatly increase use in the UK as well as elsewhere (especially as carotid screening is increasing), strengthening the need, both in the UK and elsewhere, for directly randomised evidence to be available from ACST-2 during the 2020s as to which procedure is better.

The procedural hazards are substantially lower for asymptomatic (1.0% disabling stroke or death in the first 1000 patients in ACST-2) than for symptomatic patients. Even 1.0% is a serious risk, but so too is the risk (over the next 5 or more years) of entirely trusting drug therapy and not doing any protective procedure when severe carotid disease is found in a currently asymptomatic patient. Moreover, recent claims that asymptomatic patients with serious carotid disease are at negligibly low risk on triple-drug therapy are methodologically unsound.

The ACST-2 is already the world’s largest trial of CEA versus CAS in asymptomatic patients. It currently randomises 350–400 patients per year, which is the highest recruitment rate of any large trial of carotid interventions.

Many (43%) of the > 2000 patients recruited to date have had previous ipsilateral stroke symptoms or symptoms in another cerebral territory, or have evidence of brain infarction at the time that they enter ACST-2. A recent analysis of our previous ACST-1 trial data has shown that these patients have a 50% higher risk of future stroke than those who have never had neurological symptoms. 4

The medical treatments that ACST-2 trial patients take will be analysed in much greater detail than in any previous trial of carotid intervention. The findings from ACST-1, SPARCL (Stroke Prevention by Aggressive Reduction in Cholesterol Levels)5 and individual patient data meta-analyses from the Cholesterol Treatment Triallists’ Collaborative Group6 suggest that addition of statins to antiplatelet and antihypertensive medications will lower overall stroke risk by about one-third (in ACST-1 from about 20% to about 14% over 10 years), but that the addition of CEA will reduce stroke risk still further (halving it in the first 5 years and reducing risk from 15% to 8–9% by 10 years). 1

If CAS, a newer and less invasive treatment, offers stroke protection which is as good as, or better than, CEA, then it is likely to replace invasive surgery for suitable patients in the future. This would lead to a significant change in practice in the UK as, in marked contrast to practice seen in continental Europe and the USA, very few routine CAS procedures are performed in ‘normal risk for surgery’ patients outside trials. With increasing use, costs of stents, filters and wires for CAS are decreasing and, in some higher-volume centres, costs of CEA and CAS are broadly similar. The shorter hospital stay associated with CAS would also be an advantage, saving hospital in-patient costs. In addition, most CAS is carried out under local anaesthetic, in contrast to CEA for which general anaesthesia is currently used for 50% of patients.

If CAS is as effective as CEA, practice could also change for high stroke risk patients prior to coronary bypass grafting. In the past, prophylactic CEA to reduce stroke risk from bypass has been hazardous, particularly when patients have had recent stroke symptoms, or have ongoing coronary symptoms; with CAS, future hazards might be significantly less.

Chapter 9 Conclusion

Carotid artery stenosis causes around 20,000 strokes in the UK each year. Many patients have no prior symptoms or have failed to recognise warning signs, and about half die or are severely disabled by their first stroke. Our ACST-1 trial showed that, even on good triple therapy (including statins), stroke risk for the next 10 years could be halved by preventative surgery (CEA). Two recent carotid trials are now including asymptomatic patients (ECST-2, current recruitment 180/2000 planned, and CREST-2, current recruitment 300/2840 planned) comparing a stenting or surgery procedure with no procedure. 2,3 Should these confirm the ACST-1 finding of additional benefit from a procedure in certain patients then, throughout the 2020s and beyond, the key question will be which procedure to recommend. With new technology and increasing experience, CAS can now rival or prove superior to CEA. The ACST-2 is the only trial now recruiting that directly compares CEA with stenting and will provide uniquely reliable evidence about the short-term safety and, perhaps more importantly, the long-term durability of surgery compared with stenting in patients with asymptomatic carotid stenosis. There will be an initial report in 2020, describing 4-year follow-up for 3600 patients randomised to CEA versus CAS, and a subsequent report in 2025 with around 10-year follow-up. Until then, how to intervene on asymptomatic patients with a carotid stenosis will be based on patient and professional preferences alone.

Acknowledgements

First and foremost, the principal acknowledgement is to the patients taking part in the trial, now and in the future. Second, to the funders to date (HTA programme and BUPA Foundation), to the UK Stroke Association, to all the UK and international collaborators, our office staff, the Steering and Data Monitoring Committees as well as to the trial Endpoint Review and Technical Management Committees. We also acknowledge the support of the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre Programme. Finally, to the CTSU and Epidemiological Studies Unit, the Nuffield Department of Population Health, University of Oxford, Oxford, UK, and the Nuffield Department of Surgery, University of Oxford, Oxford, UK, for their invaluable support.

Alison Halliday and Richard Bulbulia, together with Richard Peto, Hongchao Pan (Statistical Co-principal Investigators) and Leo Bonati (Neurological Co-principal Investigator) are responsible for the design, conduct and analysis of ACST-2.

References

1. Halliday A, Harrison M, Hayter E, Kong X, Mansfield A, Marro J, et al. 10-year stroke prevention after successful carotid endarterectomy for asymptomatic stenosis (ACST-1): a multicentre randomised trial. Lancet 2010;376:1074-84. https://doi.org/10.1016/S0140-6736(10)61197-X.
2. UCL Institute of Neurology . ECST-2 Protocol Summary: Version 3.10 n.d. http://s489637516.websitehome.co.uk/ECST2/protocolsummary.htm (accessed 3 April 2017).
3. CREST-2 Trial Website . The Carotid Revascularisation and Medical Management for Asymptomatic Carotid Stenosis Study n.d. www.crest2trial.org/ (accessed 1 August 2016).
4. Streifler JY, den Hartog AG, Pan S, Pan H, Bulbulia R, Thomas DJ, et al. Ten-year risk of stroke in patients with previous cerebral infarction and the impact of carotid surgery in the Asymptomatic Carotid Surgery Trial. Int J Stroke 2016;11:1020-7. https://doi.org/10.1177/1747493016660319.
5. Sillesen H, Amarenco P, Hennerici MG, Callahan A, Goldstein LB, Zivin J. Stroke Prevention by Aggressive Reduction in Cholesterol Levels Investigators . Atorvastatin reduces the risk of cardiovascular events in patients with carotid atherosclerosis: a secondary analysis of the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial. Stroke 2008;39:3297-302. https://doi.org/10.1161/STROKEAHA.108.516450.
6. Cholesterol Treatment Triallists’ (CTT) Collaboration . Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010;376:1670-81. https://doi.org/10.1016/S0140-6736(10)61350-5.
7. Baigent C, Peto R, Gray R, Parish S, Collins R, Warrell DA, et al. Oxford Textbook of Medicine. Oxford: Oxford University Press; 2010.
8. Gray WA, Chaturvedi S, Verta P. Investigators and the Executive Committees . Thirty-day outcomes for carotid artery stenting in 6320 patients from 2 prospective, multicenter, high-surgical-risk registries. Circ Cardiovasc Interv 2009;2:159-66. https://doi.org/10.1161/CIRCINTERVENTIONS.108.823013.
9. Smout J, Macdonald S, Weir G, Stansby G. Carotid artery stenting: relationship between experience and complication rate. Int J Stroke 2010;5:477-82. https://doi.org/10.1111/j.1747-4949.2010.00486.x.
10. Kallmayer MA, Tsantilas P, Knappich C, Haller B, Storck M, Stadlbauer T. Patient characteristics and outcomes of carotid endarterectomy and carotid artery stenting: analysis of the German mandatory national quality assurance registry – 2003 to 2014. J Cardiovasc Surg 2015;56:827-36.
11. Bonati LH, Dobson J, Featherstone RL, Ederle J, van der Worp HB, de Borst GJ, et al. Long-term outcomes after stenting versus endarterectomy for treatment of symptomatic carotid stenosis: the International Carotid Stenting Study (ICSS) randomised trial. Lancet 2015;385:529-38. https://doi.org/10.1016/S0140-6736(14)61184-3.
12. Heart Protection Study Collaborative Group . MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20 536 high-risk individuals: a randomised placebo controlled trial. Lancet 2002;360:7-22. https://doi.org/10.1016/S0140-6736(02)09327-3.
13. International Carotid Stenting Study Investigators . Carotid artery stenting compared with endarterectomy in patients with symptomatic carotid stenosis (International Carotid Stenting Study): an interim analysis of a randomised controlled trial. Lancet 2010;375:985-97.
14. Brott TG, Howard G, Roubin GS, Meschia JF, Mackey A, Brooks W, et al. Long-term results of stenting versus endarterectomy for carotid-artery stenosis. N Engl J Med 2016;374:1021-31. https://doi.org/10.1056/NEJMoa1505215.
15. Eckstein HH, Reiff T, Ringleb P, Jansen O, Mansmann U, Hacke W, et al. SPACE-2: A missed opportunity to compare carotid endarterectomy, carotid stenting, and best medical treatment in patients with asymptomatic carotid stenoses. Eur J Vasc Endovasc Surg 2016;51:761-5.
16. Rosenfield K, Matsumura JS, Chaturvedi S, Riles T, Ansel GM, Metzger DC, et al. Randomized trial of stent versus surgery for asymptomatic carotid stenosis. N Engl J Med 2016;374:1011-20. https://doi.org/10.1056/NEJMoa1515706.
17. Nuffield Department of Population Health . Data Access and Sharing Policy for the Nuffield Department of Population Health 2016. www.ndph.ox.ac.uk/about/rdca_dataaccessandsharingpolicy_2016-08-26.pdf (accessed 10 April 2017).

Appendix 1 Trial flow diagram

Appendix 2 Hub-and-spoke model for ACST-2 in North East England

(PDF download)

Appendix 3 Italian collaborators’ newsletter

(PDF download)

Appendix 4 Randomisation

(PDF download)

Appendix 5 One-month form

(PDF download)

Appendix 6 One-year form

(PDF download)

List of abbreviations

ACST-1

The first Asymptomatic Carotid Surgery Trial

ACST-2

The second Asymptomatic Carotid Surgery Trial

CABG

coronary artery bypass grafting

CAS

carotid artery stenting

CEA

carotid endarterectomy

CSTC

Carotid Stenosis Triallists’ Collaboration

CT

computed tomography

CTSU

Clinical Trial Service Unit

EQ-5D

EuroQol-5 Dimensions

HTA

Health Technology Assessment

ICSS

International Carotid Stenting Study

MI

myocardial infarction

MRI

magnetic resonance imaging

NIHR

National Institute for Health Research

RCT

randomised controlled trial