I've stopped following Cogstate, but I thought the following opinion piece will be of interest to many investors here in light of Biogen pulling the plug on aducanumab.

https://www.statnews.com/2024/02/06/biogen-adulhelm-fda-eisai-lecanemab-lilly-donanemab-alzheimers-drugs/

TLDR: The monetary incentive for Eisai to profit from sales of Leqembi could have the effect of prompting mass withdrawal from a phase 3 trial designed to show it prevents cognitive decline, which will have the unintended effect of preventing it from reaching full approval for a healthy population as the scientific evidence of efficacy won't have been demonstrated. Lecanemab is already approved where cognitive decline is demonstrated.

The relevant point here is that if the same mistake happens again, the FDA accelerated approvals framework that Cogstate relies upon to chase new trials in the pipeline, becomes vulnerable to tightening, especially if the Democrats remain in the White House. Under the likely Republican candidate, it won't even register as an issue.

"The Company has recently been in discussions with a third party in relation to a potential control transaction with respect to the Company. These discussions substantially commenced in late December 2022, with access to due diligence granted in late January 2023. At all times, the Company considered that the discussions were and remained confidential, non-binding, incomplete and insufficiently definite to warrant disclosure. On 18 February 2023, it was determined that the potential transaction would not proceed, and accordingly, all such discussions have ceased."

So there was a non-binding, unsolicited takeover offer with due diligence occurring over a month ago which was not considered of material significance? And then the potential third party backs out without so much as a whimper?

This strikes me as very odd to stay quiet in the current M&A market where value is being sought out and then publicly announced at the earliest opportunity so as stoke the fires of a bidding war. Instead we have closed doors and an announcement of the suitor stepping away well after the fact.

There's more questions than answers in this response, and the 25% drop is well deserved.

Two data points for digestion.

https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(22)02480-1/fulltext

This passage is particularly salient:

"After such a long and fruitless wait for a successful therapy for Alzheimer's disease, a phase 3 trial showing efficacy on clinical outcomes is welcome news. However, a 0·45-point difference on the CDR-SB, an 18-point scale, might not be clinically meaningful. A 2019 study suggested that the minimal clinically important difference for the CDR-SB was 0·98 for people with mild cognitive impairment and presumed Alzheimer's aetiology, and 1·63 for those with mild Alzheimer's disease. Furthermore, development of ARIA—seen in one in five patients taking lecanemab—could potentially lead to unmasking, introducing bias."

Noting that these trials are showing improvement relative to placebo, for there to be any wide scale government funding should the drugs make it to market, monoclonal antibodies are going to need to show significant improvement relative to treatment as usual. 

A quick Medline scan finds no head to head studies of monoclonals vs cholinesterase inhibitors +/- NMDA antagonists (unsurprising given they are not yet in widespread use), but the following article gives some food for thought.

https://doi.org/10.7759/cureus.31065

Leaving aside the aducanumab and FDA accelerated approvals controversy (and I rate Lancet well above Cureus for independence), I suspect that unless monoclonals demonstrate a significant magnitude of improvement, irrespective of the assessment tool used, the risk of the drug class being a big flop, on efficacy, safety and cost-benefit measures, is a material risk to Cogstate.

Conclusion: For Cogstate's core business, I'd want to know that contracted forward revenue is watertight in the event big pharma pulls the pin and writes off the entire drug class on efficacy, safety or cost benefit grounds. You'd also want a good margin of safety with any forward projections in case contracted revenue suddenly drops off a cliff.

If anybody knows the finer details of Cogstate's contract terms and can disclose, I am sure that there would be plenty of members here who would love to know.

Disc: Not held

For those of you with university library or paid NEJM access, the full article is here:

https://www.nejm.org/doi/full/10.1056/NEJMoa2212948

Of relevance to the Cogstate thesis is the following screenshot:

What was the rating tool you ask? The CDR-SB.

What is the CDR-SB? A rating scale from 1993, that is well validated, has decades of validity data, and dare I say it... free to use.

https://doi.org/10.1212/WNL.43.11.2412-a

What is my point?

My previous argument that Cogstate is not the only tool in existence remains, and investors need to remember that new and digital does not mean it is the only, or necessarily the best way for a screening test to be administered. It also reinforces the idea that whilst Cogstate run trials and is primarily a trials business, it is not a monopoly, nor does this mean it will translate into widespread end user acceptance that is going to be profitable.

See my previous straw from April 2022 for further context.

I know there is widespread love for this stock and for the story amongst the Strawman community. I too hope that there is a viable treatment. But as an investor, i cannot be blind to the facts that Cogstate is not the be and end all in this clinical trials space.

Fire away Strawpeople.

Disc: not held.

The Eisai contract value paid to Cogstate, of not less than USD 30m over 10 years, is a essentially a rounding error for Eisai. They can afford to bet on it and lose it all if it falls over.

Given the astronomical costs in drug development (multiple billions by the time it makes it to market), Eisai will only give a test away if they can steer a consumer towards their product. 

That's fine if the Eisai product is efficacious and superior to others (it may be or it may not), and there will be some benefit in markets where direct marketing of pharmaceuticals to individuals is permitted (e.g. USA), i.e. take this test, then go and see your doctor to ask about the Eisai product. 

Given the multiple drug candidates in the pipeline, why would Cogstate enter an exclusive agreement with Eisai for end user testing when they could quite easily play the field and market to all the players?

My guess is that Cogstate realise the margins in end user testing are likely to be thin and Eisai giving them something over 10 years is better than nothing once the clinical trials are done and dusted.

Investors need to be cautious about the '21st century digital tool' vs 'old school pen and paper' narrative that seems to have crept into the investment thesis, particularly if widespread adoption in clinical practice is part of your justification.

Irrespective of delivery method a tool is only useful for what it is designed to measure.

The Cogstate Brief Battery, is designed to measure short term memory and attention. That is what it is good at doing, as per article posted by @Marsdrix.

Nobody doubts the Cogstate battery test retest reliability, and absolutely I agree that it removes some of the problems of interrater reliability of an assessment that requires a human to deliver it.

None is this tells us whether a person has a dementia, or indeed whether they have any other condition affecting their cognition.

Memory and attention are only two of many aspects of forming an opinion about whether a person is affected by a cognitive condition.

Memory and attention are also affected in many other conditions, as correctly identified by @mbry9625, and any 'caseness' as identified by a Cogstate battery (or any other tool for that matter) tells you little about the diagnosis. 

As per @Marsdrix, it tells you that is something that needs to be examined further, and requires a skilled practitioner to piece together the information to make a diagnosis if one is there.

It's useful in clinical trials for monoclonals for dementia because the diagnosis isn't what's being tested. You only enter a stage 2 or stage 3 trial once you have a dementia diagnosis, or no diagnosis as is the case if you are a control in a stage 1 trial.

Memory and attention, which are two of the many endpoints being measured in any clinical trial, are what the Cogstate Brief Battery is good at. That's why a trial provider is happy to pay for it, because it gives them that information reliably and cost effectively compared to a manually delivered test. First mover advantage gives Cogstate the upper hand in the clinical trials space, and there is a very wide moat here.

It's not so clear cut in clinical practice, which is where I think many are seeing a large untapped potential TAM.

There is absolutely a role for Cogstate's Brief Battery or similar tool in the screening or monitoring of symptoms, especially in a telehealth context. 

There being a role does not mean it is the only way if doing it, nor does it mean it will be the 'gold standard.'

I remain sceptical that there is much upside for Cogstate in screening or monitoring, because this needs to be available at low or no cost to be taken up widely.

If it is low cost then a huge TAM can still be profitable.

If it's no cost (as many screening tools in clinical practice are), an infinite TAM still has zero profit.

The CEO is on record as saying that Cogstate have no interest in giving away their test for no cost, and frankly nor should they as a privately owned company, unless you can make a case for it being a loss leader.

As the need for computerised delivery in a screening/monitoring context increases with the tidal wave of disease burden, it would be foolish to think that you won't see an equivalent computerised test developed by a publicly funded university that is intended for widespread use at no cost.

Pick any other area of neuropsychological assessment the tools that are most used in clinical practice are the ones that are available at no cost, as long as it is a 'good enough.' There are many tools that are 'good enough', and there is a very thin moat when it comes to use in clinical practice. 

What does this mean for Cogstate? There might be some upside in testing/screening as the tidal wave of dementia hits, but it won't, in my opinion, be the goldmine some are arguing it will be, because demand will bring in competition. The competition here comes from public universities who are motivated by the greater good and will give away a comparable test for free.

I have previously written multiple bear case straws for Cogstate. I took another look at it when the price went to $1.85 on the back of new contracts recently, and still couldn't convince myself that there was sufficient upside with a margin of safety to take a bite. I was almost there, but not quite.

I will reiterate, my view is that Cogstate remains a bet on clinical trials alone, and any valuation needs to factor in negligible upside for use in clinical practice.

https://www.medscape.com/viewarticle/963464

Requires a Medscape account, but free to join as a non-professional. There's been a flurry of articles on monoclonal antibodies for Alzheimer's this week, and this one was noteworthy because of the high cost of these treatments.

This isn't good news if part of your investment thesis is that Cogstate has the potential to be used as an end user screener with a huge potential TAM. Ultimately when big money is involved, governments and health bureaucrats tend to make screening harder and the criteria to access an expensive medication stricter, particularly if the evidence suggests any benefit is marginal, as is the case with these medications. Tighter screening and diagnostic criteria means a bias to more established and readily accessible tests.

Cogstate's battery thus remains limited to utility in the clinical trials space, rather than of any meaningful impact to the end user. See my previous Straws for context.

Disc: not held.

As promised, I am giving the Strawman community some feedback from Biogen's seminar this evening.

It was a 3h long seminar (quite an effort given most of these are typically 90m or less), predominantly targeted at GPs, who are the biggest prescribers of most medications in this country.

Roughly an hour was spent on case identification. Lots of screening and diagnostic tools were spoken about, and Cogstate's tools managed to get a grand total of zero mentions. Zip. None.

The preferred tools of the expert enlisted by Biogen was a screen using the MoCA, the Addenbrooke's, MMSE, QDRS (a newish tool I hadn't heard of until today), or full cognitive assessment (taking a few hours) +/- biomarkers.

As I have previously mentioned, all of the above tools are free or low cost, equivalent in terms of sensitivity and specificity, and of most importance, familiar to clinicians and regulators alike.

My view that there is a negligible end user market for Cogstate's tools remains unchanged.

An investment in Cogstate is thus purely a bet on clinical trials.

If recent neuroscience history over the last 20 years is anything to go by, the likely outcome is a few years of growth from all the 'me too' drug candidates, and then stasis as the demand for the clinical trials product tapers off.

Neuroscience and psychopharmacology are fundamentally loss making enterprises, and whilst I can see some growth (making it potentially a short to medium term trading opportunity) I don't see this being a long term bottom drawer growth stock.

The marketing campaign for doctors has begun. From Biogen ANZ:

I'll sign up and see what, if anything is made of Cogstate in this presentation in terms of end user reach.

My previous argument that investing in Cogstate is not a direct proxy for investing in the theme of growing need for treatment in dementia is unchanged, as is my view on the stock.

I'm not going to go into an exhaustive blow by blow rebuttal to @mkkle's post, and suggest we agree to disagree.

The straws are there for all to see, particularly with where we disagree, where we do agree (for instance that there is some growth clinical trials, which was mentioned in my very first straw), and where what I have posted as straws can be understood in context rather than selectively quoted.

In summary, the areas of disagreement are whether investment in Cogstate as a company is a valid proxy for the theme of growing demand/need for dementia treatment, and whether the valuation is reasonable.

I am on record with where I stand with both.

That disagreement is what makes a market.

I am very much enjoying this robust discussion, which is what Strawman is all about.

The more I read, the more I think you need to conceptualise Cogstate as akin to a mining services company. Yes, this is a deliberately provocative statement.

I was not aware of VeraSci, but looking at their history, VeraSci came about at a very different stage in the modern history of neuroscience research, and have a very diversified suite of assessments which allows them to tap into many areas of neuroscience. This has been a huge handbrake for VeraSci because the funding tap for neuroscience research hit a wall about 15 years ago because big pharma pulled out because they decided the prospects of a return on investment were poor.  Neuroscience research since then has largely survived on government grants, which whilst hugely important, are of limited impact financially.

The big money in neuroscience research comes from pharma, and there has been huge stasis in this area the last 20 or so years. It's only been with the relatively recent legs that monoclonal antibodies have gotten that has generated this interest.

The recent growth in Cogstate and acquisition of VeraSci then look to be plays on the prospective tailwinds of pharmacotherapy for dementia. So from an investment standpoint, Cogstate is a bit like a mining services company that provides the tools to big pharma to test multiple drug candidates in the hope that one or more gets legs.

Let's make a few assumptions.

1. There is negligible revenue growth in direct to consumer or health care usage

2. The only revenue source is from clinical trials.

3. Valuation is cheap or fair based on projected earnings for the next two years.

I still think the long term growth assumptions need to be challenged, especially if the company relies on at least one $10m+ trial per year to grow revenue. This is a business strategy that is incredibly concentrated. The investment case comes down to valuation and time frame, and whether you are prepared to accept the risk of the theme (pharmacotherapy for cognitive decline) may not get the legs that are promised.

Clinical trials are risky, because if the trial falls over, the next stages don't proceed. These will more than likely be the much lower value contracts. The big, lumpy, revenue streams from $10m + contracts will be from large late stage trials. The risk of these falling over is low, but so is the prospect of unlimited growth. It comes back to a limited TAM.

Assuming the contracts are tight, the downside of a trial falling over for Cogstate will be low.

The biggest risk for Cogstate, in my mind, is that pharmacotherapy for cognitive decline dementia turns out to be a non-event. If the early monoclonal antibody trials prove successful, there will be a flood of me-too candidates, and there will be a runway of growth for say 10 years, and then investment will hit a wall as patents run out. 

Recent neuroscience history is full of similar examples - SSRIs from the 1980s, second generation antipsychotics from the late 1990's, and cholinesterase inhibitors and NMDA receptor antagonists from the early 2000s. Massive growth for 5-10 years, then pharma trying to string out and milk patents for as long as possible before market saturation.

Cogstate is not a play on dementia. It's a bet on the success of monoclonal antibodies (or some other future drug candidate) for the treatment of dementia. Valuation needs to consider these risks, and whilst the downside of a failed trial is limited for Cogstate, so is the upside.

Hence the comparison to a mining services company.

I've never said Cogstate's tests have no merit. I'm arguing that a huge TAM for a drug that treats Alzheimer's does not translate into a huge market opportunity for Cogstate, which is the basis for the phenomenal valuation that had been attributed to it.

I won't turn this into neuroscience 101 and will bring this back to the investment thesis. Any market participant is welcome to disagree, and it is those disagreements that make a market.

The bear case is thus:

1. Dementia is a huge problem for the developed world as we age not so gracefully. There is a huge market for a pharmacological intervention that reduces the burden of disease.

2. Diagnosis of dementia has always been, and will continue to be a clinical diagnosis, not made on the basis of screening tests. Screening tests may assist with identifying people in the early stages of dementia or monitor the progress of a person who has been diagnosed with dementia.

3. Cogstate sells computerised neuropsychological tests, which have been benchmarked against well established tests, both those that have no cost, and those that come with a cost. The computerised tests that Cogstate sells are at least equivalent to established tests for what they are intended to measure.

4. Their primary market for Cogstate has, and continues to be clinical trials.

5. Dementia screening in primary care will more than likely use no cost equivalents, unless there is substantial evidence of greater sensitivity or specificity from Cogstate, which I think is unlikely, based on my knowledge of how these types of screeners work.

6. Screeners to determine qualification thresholds for any new intervention for dementia are more than likely to be the same standardised no cost tools that have been and continue to be used for cholinesterase inhibitors or NMDA receptor antagonists (a MMSE score out of 30), as required by the PBS in Australia, or a similar yardstick as required by HMOs in the USA, NHS in the UK and the EU equivalents.

7. Consumers in the developed world outside of the USA cannot have drugs directly marketed to them. If there is a direct to consumer angle, it will be a pharma company paying Cogstate to make screeners available to people at the same time as a massive marketing drive directed at GPs to get them to prescribe drug X. Anybody with a less than perfect score will be encouraged to discuss the results with their GP. More than likely, this will be a one off cost incurred by the pharma company, and they will no doubt negotiate a nice big fat discount from Cogstate.

8. So the primary market was, and remains big pharma for clinical trials. That's not an unlimited TAM. In fact, it's a very limited TAM. A modest sized Randomised Controlled Trial is n=100, a large one is n=1000, and a very large, multi site RCT would be of the magnitude of n>5000. The aducanumab RCTs were n=3500 over three studies, so let's call it n=1200 each for these drug X vs placebo RCTs.

9. Any future RCT with n>5000 will more than likely be an effectiveness trial, which will measure drug X vs treatment as usual (rather than a proof of concept RCT, which is drug X vs placebo). Effectiveness trials will probably need to use the same screeners/measures that will be used in primary care or current treatments. This means large scale effectiveness trials are unlikely to be Cogstate's batteries, unless they give them to the studies for no cost, effectively as a 'loss leader.' Even then, it would have to get through multiple ethics committees to approve what would effectively amount to a sponsorship deal across the line. Believe me when I say that Human Research Ethics Committees are not easy to get on side - I have dealt with many in my time.

10. A mega multisite RCT is still a limited TAM. The large TAM of millions of consumers is for the drugs, not for the computer based screener.

11. So where is the growth coming from to justify a price/sales of 7-8x? That's in the same league as Apple, Tesla, and Microsoft, and double Amazon, all of which have TAMs of hundreds of millions, if not billions, of customers. Cogstate, on the other hand, has perhaps dozens of customers with pockets deep enough to invest in mindbogglingly expensive clinical trials.

12. For the bulls to justify the lofty valuation, points 5, 6, 9, and 11 need to be wrong. They could be wrong. My guess is that they are not.

-----

I'll do some further reading when I get the time, and if I see anything from the monoclonal antibody studies that changes my mind about Cogstate's investment prospects, I will post it here.

Disc: Not held, and my training as psychiatrist gives me zero expertise as a professional investor when it comes to valuation.

@mkkle

If a clinician can't a MMSE or RUDAS in 5 minutes, or MoCA in less than 10 minutes, they've never done one before. A good well trained clinician can do a brief cognitive screen that tells you more than what Cogstate's brief batteries do in less than 5 minutes.

Having a brief look at the videos on Cogstate's website, the Cogstate Batteries are essentially modern day interpretations of long well established and standardised tests that can be done on an iPad. The long established neurospcyhological assessments have already transitioned to iPad based delivery in any case.

I'll have a good look at what tests were used in the aducanumab trials and come back when I've had a chance to make sense of what's being used and if there is any merit to the promised growth, but what I have seen thus far does not justify the bull case.

To be clear, this isn't the sort of test you can just leave somebody to do on their own and have it be a reliable measure.

Cogstate might argue you can DIY, but to meaningfully interpret any cogntiive test, you have to understand what the testing conditions were, and be able to interpret the nuance in why there might have been a delay in flipping a card. 

Even if self-directed testing were to become commonplace, the ultimate advice would be 'go and see your doctor' if there were any concerns about performance.

I will reiterate, the price optimism on the basis of expanded testing using Cogstate's tools is misplaced. There will be some growth in trials, but I don't think it will be the rivers of gold some are arguing to justify the current valuation. I certainly don't see it taking up widespread adoption in clinical practice.

The big money is in the drugs, if they indeed work at all. The cognitive testing is a sideshow.

@chagsy

Price/Sales in my previous straw was trailing multiple.

Market Cap = A$395m
FY21 revenue = US$32.6m = A$46.6m

Even if revneues for FY22 double, and then FY23 doubles again, you've still got forward Price/Sales FY22 = 4.2, and FY23 of 2.1

In short, CGS is currently priced for perfection.

I've been trying to wrap my head around Cogstate as a company, and as an investment.

I am a psychiatrist by training. I work in both public and private sectors, and like to think i know a thing or two about brain health.

A lot has been written about Cogstate's clinical trials business, which has been the bulk of their income. Quite a few straws have argued that because the trials involve Cogstate's tools it follows that those tools go into clinical practice.

I disagree. 

Without delving too much into the ins and outs of clinical trials, once you are looking at data in granular detail, the depth of information offered by say Cogstate's tools become relevant over and above others. That applies only to the clinical trial phase. It doesn't necessarily apply in clinical practice.

Once you can prove an effect, there is then no need to use the same assessment tool provided you can demonstrate equivalence.

There are many free or low cost assessment tools that are used in diagnosis and monitoring of dementia and cognitive impairment, e.g., MMSE, MoCA, RUDAS, to name a few. These are essentially screeners (i.e. rough guesses) that are easy to administer with reasonable predictive validity.

Once a medication is proven to be efficacious, you can bet that the clinicians, the PBS (in Australia) or HMOs (in the US) will be working out what the equivalent threshold MMSE/MoCA/RUDAS score you need to be at before recommending a new, fancy and horrendously expensive drug, to keep costs down instead of using a set of tools that comes at a cost.

The only area where Cogstate's tool might become a standard will be if it can be proven that poor performance on of the subscales or subtests is predictive of future cognitive decline. I think this is unlikely, given the lack of specificity and level of high level clinicial skill required to interpret the information we can get from complex testing batteries that are available today (something I do on a day to day basis).

My view is that such an early predictor is more likely to come in the form of a novel biomarker, which will more than likely a blood test, imaging or nuclear medicine scan, or combination of the above.

If there is, then there will be a short term bump in that respect, until an equivalent, free/low cost neuropsychological assessment tool is developed and tested for face validity, inter-rater reliability and positive predictive value.

So in short, price optimism based on the increasing utilisation of Cogstate's suite of assessment tools is misplaced.

So that leaves us with where does the growth come from?

Clinical trials is, in my view, a growth area, but it's limited. Health care utilisation is limited. Sports - maybe - professional sports will always go for the fanciest and perceived best tool, but even then the market is small.

So that leaves us with valuation.

Cogstate is currently trading on approx 7-8x sales, it's way too expensive on any sensible metric, even for a fast growing company.

I for one will wait for the price to fall back, or for more evidence of increasing top line revenue that flows to bottom line.

Disc: not held.

ICYMI, there has been some insider selling in recent days. Martyn Myer has had a net disposal of 250k shares, but still has a very substantial holding. No reason for sale announced. David Dolby's net ownership is unchanged, despite the ASX announcements.

Not necessarily deal breaking news, but clearly something to watch as the price rises in case other directors start disposing.