MR CLEAN: expert opinion


Since the first presentation of data at the World Stroke Congress, the MR CLEAN trial has sent a ripple across the entire neurointerventional arena, raising questions about the future of stroke treatment and the fate of similar trials, some of which have already been halted.

At the Society of Vascular and Interventional Neurology meeting (SVIN) the as yet unpublished MR CLEAN (Multicentre randomised clinical trial of endovascular treatment for acute ischaemic stroke in the Netherlands) results were discussed by an expert panel. The panel, including Jeffrey Saver, Tudor Jovin, Osama Zaidat, Italo Linfante, Raul Nogueira, and Dileep Yavagal, focused on the effect of the trial results on other trials and where that leaves clinical practice going forward. 

MR CLEAN is a pragmatic phase III multicentre randomised clinical trial with blinded outcome assessment. Endovascular treatment (intra-arterial thrombolysis, mechanical treatment or both) was compared with no treatment, against a background of optimal medical management, with or without intravenous alteplase. With the enrolment of 500 patients complete, the results presented at World Stroke Congress were in favour of interventional treatment.

Jeffrey Saver, who was present at the World Stroke Congress, gave a brief summary of MR CLEAN: “MR CLEAN had a very broad design. They could enrol patients who were getting t-PA or not getting t-PA as the background therapy up to six hours after onset. They had broad clinical criteria—NIHSS score of two or higher. Patients had to have an occlusion in the terminal ICA, in the M1 M2 or in the A1 A2. Even though they were broadly inclusive and started in the pre-stent retriever era, most of the patients (97%) were treated with retrievable stents, so MR CLEAN is basically a stent retriever trial. The average time from onset to t-PA was under 90 minutes, but the average time from onset to randomisation was 200 minutes. Because there was 110 minutes between t-PA start and randomisation, they ended up with t-PA failure patients. Onset to groin puncture was 60 minutes faster than some of the prior studies. Serious adverse events were not greatly different between the two groups. In terms of results, in the onset to randomisation, whether patients were randomised early or late, within the first two hours or beyond the first two hours, they benefited, so two hours is not a key cut point.”

Way forward

The question was then raised: with MR CLEAN and similar trials such as ESCAPE and EXTEND IA declared positive, what is the next step? “Are we done with our trials; can we just go into clinical practice and not have to worry about anything yet?”

Saver responded that just looking at the results of MR CLEAN, the answer is no. He pointed to the fully able to return to work/non-disabled outcome which reportedly showed that 88% of patients in the intervention arm are still unable to return to work and 97% of patients still have symptoms from their stroke. “So there is still a huge need for additional therapies. We have had a major step forward but we need to keep doing better, and so we need to move on to the next generation of trials.”

As for the process of making decisions for other trials based on data that are yet to be published, the expert panel maintained that care must be taken.

Tudor Jovin cautioned, “We must remember that these results are in non-published form; we do not know how these results are going to look in their definitive form. The same applies to ESCAPE. Before ESCAPE we could have said that this could be an outlier, but to me, the fact that ESCAPE was stopped due to crossing pre-defined efficacy boundaries tells me that this probably is the kind of treatment effect that we are looking at. So I suspect the data are real and because of that I have no reason to believe that other ongoing trials are not going to go the same direction. The question to me is less ‘should the trials be stopped’, and more, ‘should consenting be changed’, because for all of these reasons you can make a good argument that trials should not be stopped. The question is ‘how do we incorporate this kind of information into the consenting process’.”

Following on from Jovin, Raul Nogueira stated, “Equipoise is not what we [personally] think, it is about what the whole medical community thinks, and we do not know if things are going to change, and stopping before things change would be a mistake. We have duties to the patient in front of us and to the whole community of patients we will see for the rest of our lives. It is a complicated ethical dilemma, I think we need to measure the degree of equipoise in the whole community, and I think the trials should probably be put on hold as opposed to stopped.”

Up to time of publication, the ESCAPE trial has been permanently halted, the SWIFT-PRIME trial has been temporarily halted, and the REVASCAT trial continues to enrol, with a scheduled pre-planned analysis due in the near future.