WSC 2026 | Approaches to enhancing reperfusion therapies for AIS: AI, lightweight imaging devices, & more

Geoffrey Donnan

My name is Geoffrey Donnan, Professor of Neurology at the University of Melbourne and co-chair of what is known as the Australian Stroke Alliance, which is an alliance of interested parties to further the pre-hospital treatment of stroke in Australia.

Stephen Davis

I’m Stephen Davis, also Professor of Neurology at the University of Melbourne and Director of the Melbourne Brain Centre at the Royal Melbourne Hospital.

Geoffrey Donnan

So Steve, you and I have worked together for years and we’ve got a common interest in enhancing reperfusion therapies basically, particularly in the pre-hospital space through the Australian Stroke Alliance, which you and I co-chair. And there are lots of ways we can improve reperfusion. We’ll go through a few of them, will we? We’ll start with some of the newer thrombolytic agents, and I know you’re giving a talk on tenecteplase. Will we chat about that first?

Stephen Davis

Let’s do that first. I think it’s interesting because the American guideline says that alteplase and tenecteplase can be equally used under 4.5 hours. But when you look at all of the data and meta-analyses, clearly tenecteplase is superior. It’s easier to give. There’s similar safety profile. Meta-analysis shows slight superiority. It’s unequivocally non-inferior and much easier to give.

Geoffrey Donnan

Yes. In my mind, what stands out so much above everything else is ease of use. If there’s something equivalent or even a bit better, especially as you and I work in the mobile stroke unit, there is nothing simpler than giving tenecteplase and nothing more difficult than giving TPA. Absolutely. It’s difficult than giving TPA. It’s a long half-life and the fact that it’s probably more efficacious whether that’s dose related or mode of administration related is a bit unclear but whatever is the end of the game is it’s a bit more effective and certainly easier to use.

Stephen Davis

And the biology is that it does seem to enhance reperfusion more than alteplase. At any time point, which Vignan Yogendrakumar showed.

Geoffrey Donnan

Yes, yes, yes. Well, we’ll touch on that because we will also talk about how to improve reperfusion therapies by improving workflows. And workflows are critical, as we know, in acute stroke management. And if you reduce the metrics or improve the metrics is a better way of putting it, you do improve reperfusion and thereby we understand quite clearly that improves stroke outcomes. And we just did a very nice study led by Vignan Yogendrakumar, with us from Canada. And he showed very nicely using the statistical approach of DOORS, win-loss ratios, hierarchical ratios, that it was much more efficient for a telemedicine model with a mobile stroke unit than the method we’ve been using in the past of having a neurologist on the stroke unit. So I think that is the first time that that evidence has been provided. So this, I think, should give comfort to those around the world that are using mobile stroke units and now so many countries around the world are doing so that with comfort they’ll be able to use the telemedicine model rather than the onboard neurologist model. Would you agree?

Stephen Davis

I totally agree and I think that more and more we’re going to see advances based on modern technology, digital communication. I think AI is going to play an increasing role. And I think the pre-hospital component is very, very important and involvement of the paramedics. I mean, that’s a very important aspect, isn’t it? Communication between the paramedics and a stroke specialist. Is this likely to be a stroke? Where should the patient go? And then that’s going to lead eventually to lightweight imaging being installed in these stroke smart ambulances.

Geoffrey Donnan

Yes, I think you and I are at one on this that over the last decade, we’ve worked with paramedics for the first time in our careers. Nothing has been more rewarding because they are so responsive in terms of being involved in the research process, which enhances their performance. It enhances the outcomes of the patients, I think. And as you say, particularly in this field, we’re working with them with the introduction of ultra lightweight imaging, which we’re doing in Australia. Will we talk about the lightweight in more detail? Do you want to talk about the cone beam CT? And I’ll talk about the Envision product.

Stephen Davis

So a CT scanner in the hospital weighs two or three tons. The CT scanners we use in mobile stroke units is 700, 800 kilograms, still very, very heavy units and impractical for routine ambulances and certainly for aircraft. So the cone beam CT is aiming for a weight of about 70 kilograms. The current prototype weighs a bit more than that and it looks very effective in phantom imaging and shows quite good resolution rather similar to early CT scans.

Geoffrey Donnan

Very similar to our earlier days of doing CT with the early generation CT scans, which is adequate. Which is adequate. Very, very adequate.

Stephen Davis

What we really need to do is exclude blood so that we can use thrombolysis in the field. And if we see blood, of course, to reduce the blood pressure. Yes, yes. And send them to an appropriate centre.

Geoffrey Donnan

Yes, I think the really important piece about this cone beam CT is what the sensitivity and specificity for blood is going to be. And if it is high, I think that is going to be extremely useful. I was chatting today with another colleague about the GFAP blood story with hemorrhage. I think I see, and I’d be interested in your view, I see later an algorithm between the lightweight imaging and a biomarker in the blood to give us getting near to 99.9%.

Stephen Davis

And clinical information. So if you have atrial fibrillation, sudden onset stroke, it’s highly likely to be embolic and a stenic stroke. So all of this information will be put together with AI.

Geoffrey Donnan

Yes, yes. AI is the way to go. And the other, of course, is the electromagnetic approach that’s being used by the Envision company in Brisbane and Sydney, which is a novel approach. The wavelength is about the same as our mobile phone, so it’s totally harmless. And it uses an incident wave, reflected wave technique to identify pathologies within the brain. And they’ve demonstrated pretty clearly that for obvious hemorrhages and infarcts, they’re going to be able to pick it up. We’re not quite sure what the sensitivity and specificity is going to be as yet. Preliminary information suggests it’s probably well over 80%, probably over 90%. But we’ve got our pivotal trial being run with centres in the US, in Australia. We’re probably going to spread into Asia as well, which will give us that detail. And I think with that ultra lightweight device, which is a bit like a bike helmet, isn’t it? Only 13 kilograms. You can have it in your backpack. I think that is going to be incredibly useful around the world. And getting back to our point about the combination with other factors, like clinical, as you say, blood biomarkers, with that device also, I think we should be able to get up to sensitivity for blood and perhaps for ischemia as well, up to as high as 99%. And that’s what you need.

Stephen Davis

And the other area we’re both interested in as a platform is aircraft and helicopters. Yes, yes, yes. And there’s a lot of needs, particularly in large countries with remote populations, Australia, Canada, for example. Norway, I know, very interesting. Yes, we collaborate closely with our Norwegian friends. We collaborate closely with our Norwegian friends. So it’s a very exciting area. So if you can put in a 13 kilogram backpack, as you say, and get information about the likelihood of stroke in a remote region of Norway or Australia, treat the patient, thrombolyse them on the spot or lower their blood pressure, work out where they’re going to go, what a breakthrough.

Geoffrey Donnan

Metrics will improve, therefore outcomes will improve. So I think it’s a time of revolution. And to go back to the mobile stroke unit concept, which you and I have been very familiar with having two mobile stroke units in Melbourne, is that they are large trucks and they’re not cheap. But I suppose not cheap is always relative. It’s about one and a half million Australian dollars to set up and to run annually about the same amount. What we see in the future, if you would agree, is that we would have standard ambulances, which we call stroke smart ambulances with an elite trained paramedic force, we have two paramedics in our trucks they’ll have special stroke training about stroke scores, and the addition of these ultra lightweight devices whether it be the cone beam CT or the Envision device. Doesn’t matter, but I think that will bring the cost down so that the ability to have these in most cities, most parts of the world and in rural and regional areas, I think would be very important.

Stephen Davis

I think globally this could have an enormous impact. The other thing we should talk about briefly is the new adjuvant experimental therapy to increase brain perfusion, which you talked about here at the meeting. Two very interesting approaches. Do you want to touch on those?

Geoffrey Donnan

Yes. Adjuvant therapies are, I suppose, the other cutting edge way of improving re-perfusion. But let me preface by saying it’s a tough area. It’s a tough area. And the reason I say that is with thrombolysis back in 1995 and now in 2015, thrombectomy, you’ve got two of the most powerful, particularly thrombectomy, probably one of the most powerful interventions in modern medicine. To improve on that is no mean task. But it doesn’t mean it can’t be done because not all patients recover and there are other means of improving it. And one of these adjuvant therapies, and we’ve got two that we’re working with in Australia, one is TB309, which is an antiplatelet agent, which works via modifying integrin. And the advantage of this is it’s been shown to be very effective in numerous animal models of ischemic stroke. When you add it to TPA, it improves the recanalization rates, the reperfusion rates. But even more importantly, unlike adding a GP2B3A inhibitor, it doesn’t increase the bleeding. And that is absolutely critical.

Stephen Davis

I mean, the other thing that’s relevant to this discussion is use of dual antiplatelet therapy in certainly a minor stroke, high risk minor stroke, but we’re using it more extensively for several weeks after onset. But it seems that for mild stroke, that antiplatelets must have some therapeutic effect beyond secondary prevention.

Geoffrey Donnan

Yes, yes. And you do need to give, we’re discussing this just today, you do need to give them early. And there is undoubtedly a hyperacute effect, isn’t there? And that might lean into what you’re saying. There might be other benefits that we don’t quite understand.

Stephen Davis

And the other thing I’d just ask you about is after thrombectomy, particularly with large cores, we know thrombectomy works in large core ischemic stroke, but still a lot of patients have a bad outcome. In fact, with large cores, most of them, despite the benefits of thrombectomy, have bad outcomes. So what about following thrombectomy? What’s happening in the microcirculation, this no-reflow phenomenon? That’s an interesting area.

Geoffrey Donnan

Yes, yes. No-reflow is you and I have known about for years because we both spent time at the Mass General Hospital where William Ames was the first to describe no reflow in a stroke model years ago, this was back in the 70s. And it is one of the major reasons why you don’t get full recovery post-thrombolysis or thrombectomy. The difficulty, probably about 30% have no reflow after intravenous therapy or thrombectomy. So how do you get to that group, which is probably a microcirculatory phenomenon, and what treatments do you give? And this is, I think, just the beginning. I mean, intra-arterial follow-up thrombolysis is one way, intra-arterial other agents of various sorts to modify pericytes or anti-inflammatory agents, etc. What do you think?

Stephen Davis

I think that’s a very fertile area. But the other thing we went to the meeting, one of the workshops was the question of how do you design a trial with sequential interventions? That’s an interesting challenge, isn’t it? If you’re going to do sequential randomizations, thrombectomy and then followed by reperfusing if there’s no reflow phenomenon. Success or failure and then perhaps adding on another arm of an antiplatelet or even a neuroprotective strategy is quite complex, isn’t it?

Geoffrey Donnan

Yes. And it is a good way to push the envelope, isn’t it? For, let’s say, there’s a failure of thrombectomy. Then you re-randomise to say, well, let’s look at the ones who have failed and randomise them into something else, other two possibilities, whereas the successes, of course, you wouldn’t randomise. So I think it’s a new and novel design feature of clinical trials. I must say one of the other great advantages, and perhaps I might finish on this, that has occurred during our career is the change in clinical trial design from those first trials, Australian Streptokinase trial that we ran back in the 90s, how these modern approaches to trials, clinical trials, allows greater flexibility, and I think that you can run them at lower cost with, I think, more precise outcomes.

Stephen Davis

It’s a pretty exciting area, isn’t it? It’s a very exciting area. I mean, the whole reperfusion area, it’s a very, very exciting area. Encourage younger people, get involved. When we first went to stroke meetings, there’d be negative trial after negative trial, and now you sort of expect a lot of positive results at stroke meetings.

Geoffrey Donnan

And to finish off, I remember at the meeting earlier yesterday, one of the young neurologists said, well, I’ve got about 40 years ahead of me. And I can see that during my lifetime, I’m going to be able to solve this. But it’s not going to be yours and mine. But the young people, I think, are going to have a lot of work ahead of them.

Stephen Davis
Exactly.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.