Hi again General,

Thanks for the update regarding 2nd generation scanner and revised strategy.

I hope you dont mind me adding some more cold water. Sorry!

Some years ago there was much excitement regarding thrombolysis (clot busting drugs) for ischaemic strokes. There was clear evidence of benefit for using thrombolytics in heart attacks: the thrombolytics were given in through a drip, dissolved the clot that was blocking the blood and oxygen flow getting to a part of the heart. The oxygen starved section of heart (and patient) got better. A small percentage of patients had life-threatening bleeds into their guts or brain for example. A number of them died, but mostly the risk of giving the drug and stopping the patient dying from a heart attack significantly outweighed the risk of dying from a heart attack. Studies were done proving this, with 10s of thousands of patients. 

NNT for STEMI

here is a link showing the numbers of pooled results from multiple studeis showing that the "number needed to treat" (NNT) to cause a good outcome, vastly outweighs the NNT to cause harm.

For years, in medicine we have argued about the risk:benefit ratio for thrombolysis in acute ischaemic CVA. The studies done were never of the same size (couple of hundred) which reduces the power and statistical validity. A bunch got stopped halfway through as they caused more harm than good - never a good look. Most of the push to use thrombolytics came from the AHA (American Heart Association) and people questioned what they were doing poking their oar into a neurolgocial condition. The spotlight beam sharpened when it was pointed out that their (the AHA) major financial sponsor was the manufacturer of tPA (which just happened to be the drug used for thrombolysis). And so it went on. A new study would come out and the talking heads would be dissecting the numbers to validate their underlying biases.

It is likely that, on balance, thrombolysis is helpful in a carefully selected group of patients who are younger, present earlier, have just the right size of stroke and no contra-indications:

NNT for ischaemic stroke

That is the premise for this product: identify these patients in the driveway, prove they have an ischaemic stroke and give them the brain saving thrombolysis.

Only problem is that we dont give thrombolysis anymore. We do clot retrieval. The patient is taken to a stroke hospital to have the catheter thingy I talked about earlier, fed up into the artery supplying the brain to suck out the clot. NO thrombolysis.

There is a case for something called "bridging thrombolysis": if a patient accidentally goes to a hospital without clot retrieval, gets a scan proving the "goldilocks size" and time of Stroke, have it discussed with a stroke team, get accepted for transfer for clot retrieval and have no contra-indications for thrombolysis; then thrombolysis is useful - probably. (I have done it once in the last 15 years of working as an Emergency Physician).

So, to apply these thoughts to pre-hospital thrombolysis:

patient gets loaded into back of ambulance with a possible stroke

establish that the stroke is within the time window for treatment (longer time = greater chance of bleeding and death or bad outcome if given thrombolysis)

establish the stroke is of goldilocks size (usually performed by a stroke physician or at a push an Emergency physician with an interest). But in this secenario, by an ambulance officer - I can't see them being overly keen to take that responsibility!

put the scanner on the patient and run the scan

get the result and have a scan with sufficient accuracy that it has a zero chance of misinterpreting the ischaemic stroke as a bleed (or you kill the patient)

Adminster the thrombolytic the drug, and drive off to the stroke unit.

All that has to happen faultlessly in one cohort of patients.

Then you have to compare that cohort of patients to the current standard of care and prove that patients will do better with our new fancy scanner + driveway thrombolysis option. You would need to run a trial with 100s and likely 1000s of patients to demonstrate an improvement in outcome. This will take many years to recruit into and run, and cost millions of dollars. There is a very real chance of it causing more harm than good (my guess is that this would indeed be the case).

And finally, as per previous, you would then need to demonstrate a cost:benefit case to convince health providers that this is worth the dollar spend, as opposed to every other good idea and interest group out there clamouring for funding.

Jeez that went for a bit longer than i intended, but I hope I have demonstrated just how difficult it is to make any money with a new "medical breakthrough". I lost money hand over fist when I started out investing, mostly on just such companies. I am not going back there again. High ROE, high margin, low cost of capital, no/low debt, structural tailwinds, predominantly FCF positive and asymmetric risk companies for me from now on.

All the best

C

It's great to see an evolving medtech company on the ASX. This is an amazing bit of technology, however.....

They are a long, long way away from demonstrating a use case, and even further away from demonstrating an adequate accuracy to guide management, and yet a few steps further away from demonstrating an improvement in clinical outcome and then a step or two to demonstrate a cost:benefit advantage over current managment.

They are targeting acute stroke as their clinical indication for use. There are two types of stroke, a bleed (haemorrhagic) and a blockage in an artery (ischaemic).

The treatments for each are quite different. Specifically, to treat ischaemic strokes drugs that thin the blood are given which will make a haemorrhagic stroke worse, potentially with a fatal outcome. In some instances a clot busting drug can be given (like for heart attacks). This is not a low risk proposition even if given for ischaemic strokes. If given for the wrong type of stroke ie a haemorrhage stroke, one can safely assume the patient will die. However, clot using drugs are nowadays used less and less as acute ischaemic strokes are currently managed with something clot retrieval. Here a catheter is inserted into the blood vessels in the brain and the offending blood sucked out, thus restoring blood and oxygen flow and limiting the damage to the brain.

Their most recent study on data sets at the PA hospital in Brizzy, shows the ability to differentiate between a haemorrhagic and ischaemic stroke of 93.3%. Which doesnt sound bad, but means that if we were to use this product in its current set up we could kill up to 6.5% of patients for, as yet, no demonstrated advantage (were clot busting drugs used, or huge resource overuse if clot retrieval used). There is a long way to go for this use case, just to prove equivalence in accuracy. You would then need to demonstrate  that the time saved by using a bedside scan as opposed to putting the patient through a conventional scanning technique which is all of 20m away in most hospitals, translates into a meaningful improvement in mortality and morbidity (death and deficit) - I have no confidence that this could be achieved.

That is the use case for this scanner within the emergency department.

It may have a better use case in the pre-hospital setting. The theory being that potential stroke patient can be identified at the scene and then transported to an acute stroke unit facility that can offer clot retrieval. Currently, patients that have symptoms of a an acute stroke are transferred to stroke units. It doesnt matter if they could be an ischaemic or haemorrhagic stroke. Given that 87% are ischaemic strokes, the number of haemorrhagic strokes transported unnecessarily to a clot retrieval facility is low. Applying the converse situation, what happens to the ischaemic stroke patient that is not transported to the correct facility ? They have a potentially far worse outcome that was entirely preventable under the old system. Medico-legal pay-outs related to this scenario would be punitive.

Like any other investment thesis, I would start with "what problem does this solve" and "how well does it do it".

Currently my feeling is "I dont know" and "not very well".

Now, it may be that with further training of the data sets this scanner becomes very accurate. Even then, the cost of trials to demonstrate its safety and efficacy will be prohibitive. The medical approval process is incredibly slow and expensive and has a very poor success rate.