Healthcare Conference. My name is Lucy Codrington. I'm one of the European Pharma and Biotech team based here in London. It's my pleasure to introduce Dr. Frank Weber, the CEO and CMO of Vivoryon, who's going to give us a presentation. With that, over to you.
Wonderful. Thanks for the kind introduction. Thank you for joining this conference. My name is Frank Weber. I'm the CEO, a minority shareholder, and employee of Vivoryon Therapeutics, and I will lead you through a presentation about early Alzheimer's and the small molecule glutaminyl cyclase inhibitor for the treatment of early Alzheimer's disease, and that molecule is called varoglutamstat. Please note, take notice of our disclaimer and the forward-looking statements. And then in one shot, what is unique about us and what sets us apart from everything else, which is in the space of early Alzheimer's disease therapeutics for development or already launched, and that is that we are the only mechanism which stops the production of neurotoxic pGlu-Aβ, while inhibiting an enzyme, which is the rate-limiting step to make the production of that enzyme.
That enzyme is called glutaminyl cyclase, and the neurotoxic peptide is called pGlu-Aβ. Many, many years ago, pGlu-Aβ was an unknown to everybody. Now, it's, for the insiders at least, known as it is a target also for donanemab, which is currently in approval for an antibody. But there are, huge differences between stopping the production of a neurotoxin and clearing it with an antibody. And the advantages are likely higher efficacy and better safety if you stop the production, rather, you clear it later. And I come into depth in that, overall, later in the presentation. We are at the end of phase II, and we expect the results of that large phase II study, phase II-B study, to be available in the Q1 , 2024.
We have a parallel study in the U.S., where we also look with different endpoints and a different methodology in early Alzheimer disease, and I give you the rationale why having two parallel studies in a second. And finally, I mean, that sets us probably not so much apart because we have a huge talent here at Jefferies. Everybody is a great entrepreneur, but we also have a good management team, and I think we don't need to hide, but of course, there is probably people more clever than us also around here. So, in that sense, I think we are good, but good as anybody else, probably. We have a financial runway. Sorry, I got a cold, and it is not very healthy here.
And we got our runway well beyond our value inflection point, so we have a runway until the second half of 2024, as the results come in the Q1 . There's ample time to profit from the data. Now, this is the pipeline, and you see us today laser sharp focused on doing only one thing, delivering varoglutamstat into a phase II-B study in early Alzheimer's disease. We have a partner in China, Simcere. We licensed the right for Greater China to that company. We have other assets in research, but we are currently not moving it because we have fully focused as a small company on delivering that asset to the market and to the patients. Let's talk a little bit about early Alzheimer's or Alzheimer's in general.
I think it's not needed to be said that it's a huge market and the medical need is significant. What most of you may not know is that it's largely driven by the aging population. It's not that we are suddenly a pandemic like in COVID, and everybody gets it. It's just that with the age, your risk of having it just increases, and as people getting healthier and healthier and live longer and longer, more and more of these people get it. And that is true for the Western developed countries, but it's also true in the future for less developed countries, where the life expectancy these times is probably in the sixties or and not in the eighties, and they will get there relatively quickly.
So it is a social, economic problem, and as the birth rates drop down, the number of people who can take care of advanced Alzheimer's patients is limited, and the cost thereof is huge. And so we need to stop that disease as soon as possible, otherwise, we get in really in trouble. And now, I want to switch a little bit what is there, because as a company working in the field of Alzheimer's disease, we cannot be agnostic to what others do. You know, we need to, of course, understand where the market goes and what is there already. And there are two antibodies: lecanemab from Eisai and Biogen is already launched, and donanemab from Lilly is also monoclonal antibody, both directed to neurotoxic peptides. They are... Donanemab is under registration in the U.S.
And the interesting thing is that the data are published for their phase III, and the interesting thing is that they say pretty much the same thing. And when you look at it, donanemab made a huge effort in stratifying the patient in tau and amyloid. Everybody has an amyloid-beta positive signature. And then the tau signature, they stratified in low, medium, and high, and tau is a peptide which goes out of the cell if the cell is damaged. And so tau is a surrogate for an intracellular pathology of the disorder. And when you see what the efficacy in the low tau population is, it's 35%, whereas what it is in the high tau population is 6% and not significant.
It tells you what everybody already knows, that monoclonal antibodies get into the brain with some difficulties, but never get in the cell. In other words, once Alzheimer's has cellular pathology, antibodies are probably not a good choice. In the very early phases, these patients actually progress much more on the high tau than the low tau. These are patients which, in the early space, are already a little bit more advanced and have probably twice to 3x the progression rate than patients who have no tau positivity. This tells you that there is a huge unmet medical need for the patients who have already intracellular pathology and are a little bit more advanced and progressing faster. The other point is that while at least the very earliest patients, we have a big impact with the monoclonal antibodies, and 35% is not negligible.
35% in patients which are amyloid positive and have low, medium tau is a reasonably good value. But there is, of course, always the benefit-risk question, and it comes with ARIAs, it comes with infusion-site infusion related reactions. It comes with the monitoring of MRIs, and it comes with a very fragile elderly population, which is not mobile and running around and finding their therapeutic site. So it comes with limitations. Having said that, with varoglutamstat, we have an all-tau population, meaning that everybody needs to be positive on tau, but there is no upper limit. So we have a range of tau positivity. We selected the patients being amyloid positive and tau positive.
In both studies, the one which reads out in a couple of months and the one which is, a longer shot, potentially confirmatory study. So having said that, we have understood where, the unmet medical need in the area of, of Alzheimer's. What we need is improved safety and higher efficacy in early AD patients, and we need robust efficacy in more progressive, advanced mild AD patients with elevated tau levels. We don't want ARIAs, and of course, we want a convenient administration at home rather than sending elderly patients around the states and cities to get their infusion centers. Even subcutaneous may solve the problem partially, but only partially. Varoglutamstat has the full potential to fill these gaps.
We believe, based on preclinical research and some phase II-A data which we published, that we have the potential for higher safety and better efficacy compared to the standard of care. We have a novel mechanism preventing the production of neurotoxic peptides and, including the tau population, which is downstream from our research to neurotoxic Aβ. So neurotoxic Aβ in the cell creates the tau problem, the disassembly of the tubules in the axon, and if you don't have neurotoxic pGlu- Aβ, you don't have that problem anymore. We believe that we have a favorable product profile, and I will talk about this with no evidence of ARIA in clinical settings. We looked at this in the previous studies. We're continuously monitoring it. We have probably now 300 patients on it.
We never saw an ARIA which was symptomatic, and even in MRIs, we have done so far, we don't see anything, and it should also not be because it's antibody-related. And then, of course, our small, small molecule can be taken in the outpatient setting conveniently. No need for infusion. So how does that thing work? Varoglutamstat inhibits the rate-limiting enzyme, which is called QPCT, which is glutamine, glutamine peptide transferase or glutamine cyclase. That is an enzyme which everybody has in the brain and in the body, and is there for maturing peptides, any type of peptides. Usually, all these peptides have a glutamine amino acid at the N-terminal of their peptide.
So everything which is glutamine at the end, the enzyme comes, makes a ring form out of it, and that's the way the body actually efficiently can improve the potency of a peptide without increasing the amount. Because what it does with the pyroGlu, all these peptides get more potent, and all of these peptides get more resistance against the proteases digesting them again. So it is a physiological way the glutaminyl cyclase works on peptides which have a glutamine residue. Now, unfortunately, and this is, for example, the case for CCL2. This is one of the things, but there are neurohormones, there are endocrine hormones, there are immunocytokines, fractalkine, whatever. They all get that ring form, and that increases the potency of the peptide.
Now, unfortunately, with aging, the degradation process of APP, which is the biggest turnover molecule in the brain of the human, going to A-beta down, this degradation process is a little bit disordered. And you find over the time, A-beta versions with N-terminal 3 and 11 having a glutamate amino acid at 3 and 11 position of A-beta. Normally you have A-beta 1-42, 1-40, 1-37. There are, I don't know how many versions of A-beta at the C-terminal side, but suddenly with the age, the clearance process becomes less efficient, and also the N-terminal varies. And then you have a 3-42, 11-42, 3-40, 11-40, and so on. And at 3-11 position, at the N-terminus, there's a glutamate, and then this little enzyme comes and makes a ring form and creates this pGlu- Aβ .
It's a very bad substrate, these 3-11 Aβs, because it's a glutamate at the end and not a glutamine. But over the time, the enzyme gets there and does it. And it's a very long process, which explains why Alzheimer's takes so long from the biochemical state to the clinical state. It's a very slow, long process because the enzyme is not very efficient at doing it. But once these neurotoxic peptides are produced, they are very hard to clear, and when the production exceeds the clearance, natural clearance, of course, you get the neurotoxicity slowly up and running. And while then, the inflammation goes up, the body even upregulates this enzyme because it's primarily there normally, physiologically, to upregulate immune cytokines. So it upregulates and produces even more.
So you have a vicious circle, basically, with this enzyme producing more and more pGlu- Aβ. And this pGlu- Aβ sees the plaques, creates a synaptic impairment. This is how our memory works. The memory works not in the plaque. You know, in the plaque, there's no memory function. Plaque is a deposit. Memory works in the, in the neurons, in the synapses, and it creates tau pathology, and it creates neuroinflammation. So that is how the system works, and when you inhibit with an enzyme inhibitor, about 80%, 50%-80%, you basically get rid of the new pGlu- Aβs. The remaining in the brain and things get degraded, and the world is good again. At least that is the hypothesis. And we are not the only one who believe this.
We see that this pGlu- Aβ actually is in the cell and in the membrane. So the difference between an antibody who floats around in the CSF, actually, the toxins not necessarily come in the CSF first. They're secreted in the synapses. That is proven, and they sit in the cell walls in the cells. That's where they do the primary damage. Of course, they are also in the CSF. They are also in the plaque. But the antibodies only can address those in the plaques and those which are solidly floating in the CSF. Not those, not those in the cell wall, not those in the synapses, and not those in the cells, because antibodies don't go there in a sufficient amount.
So what you see is, and that is proven by others, I made some nice pictures here just to show you that, there is an electrophysiological change when you have pores. You know, all neurons have pores where they go through, and you have a pGlu-Aβ pore versus a normal pore. It changes electrical conductance and homeostasis. You see early accumulation in the neuropil around the synapses, which create a damage to the memory function in the synapses. And of course, it gradually accumulates in astrocytes because that's the clearance. Astrocytes are the clearance mechanism of the brain. Sorry. And this is where, it, at the end, accumulates because the body wants to get rid of it. And what the astrocytes cannot digest, it goes in the plaques. So development.
Going back a little bit more product-centric, having discussed a little bit the pathobiology, the environment of Alzheimer's, going about what we did in the development program. We did a large phase I. We don't do biotech studies; we do solid studies. So we have a 200-volunteer phase I study with complete PK/PD modeling. We're not doing 3 subjects here and 5 there. We wanna know with a new mechanism of action and with a new target, what's really going on and how we dose it, and not be proven stupid later. We did a phase II-A study for only 12 weeks treatment duration, limited by the tox data we had at that point in time, and we showed evidence of a positive effect on synaptic recovery.
We have an MTD defined there, and now we are here in the phase II-B with 260 patients. An average treatment of these patients of more than 80 weeks. So it's not a short study. That is a study which allows you to really read out the Alzheimer's disease. Nice, randomized, double-blind, and expected study. And we have another phase II study, with, in the U.S., partially sponsored by the NIH, actually, by a grant. And, where we look at, the same drug at the same patients, but with a different endpoint. Now, you will say: Why do they do three phase II studies? It's not 3 phase II studies.
Of course, it is, but the objective of this is if the drug works, we try to get accelerated approval with VIVIAD, and we try to get full approval with VIVA-MIND, transferring into a phase III study, just adding a couple of patients, because the endpoint is already there for full approval. This is an intermediate clinical endpoint. That is the final clinical endpoint. So this is what we have here. So we start small and nimble, but with a huge upside in our regulatory strategy. So it's not making 3 phase II studies, never thinking of confirmatory study. It's already inbuilt the pathway to the market in that program. And we hope if the drug works, or we're sure if the drug works, we don't need more study than this, at least double-blind studies providing evidence.
So what would you find in 12 weeks of treatment in the first setting in early AD patients? You can debate these results for a fact, but no disease-modifying drug has ever shown anything after 12 weeks. What we did here is we gave the drug and tried to measure whether pharmacologically the drug is active in the brain and thus the intended benefit. What you see on the left side is you see an improvement of working memory and attention after 12 weeks, and working memory and attention is probably the fastest you can react. So there's a real improvement after 12 weeks, as there is an improvement in the EEG synaptic activity. EEG is completely placebo-uncontrolled, unaffected, so it's not affected by any rater. You see a reduction of neural, neuronal injury biomarkers.
So after only 12 weeks, consistent with the hypothesis we had, we see signs of neuronal recovery, which says there's less pGlu-Aβ. The neurons are doing better, both from the electrical point of view, from the biochemical point of view, and from the short-term memory function the patient has. And of course, these data do not prove that the drug works in Alzheimer. Alzheimer is a disease of 5 years. You cannot, with a 12-week study, say, "ah, now you will..." No. But it shows that pharmacologically, the drug was what we expected doing it, and it's a sound basis for investing in the phase II-B program, which I showed you, which we have raw results, and then we have NIH funding. So it's an intermediate step from a patient, from a volunteer to a final big patient study. Study, by the way, is published.
Now, in the phase II-B, as I said, we are currently laser-focused on just delivering our phase II program. We look, and we published, of course, the baseline data of our study. As you see, we are exactly in the middle. I don't know how much you know about Alzheimer's, but the MMSE is a Mini-Mental scale, which is basically a very rough scale to say how disabled your disease is or you are already, and 25 is exactly in the middle between MCI and mild dementia. We are exactly in the middle. When you look at it, it's the same population baseline as lecanemab. It's not the same selection criteria. Don't, don't get me wrong, they are all subtle differences. For your information, donanemab had, I think, an MMSE of 22.
Lower is worse, meaning they probably had a little bit of a worse patient selection criteria there. And you see that, we are proud to announce that our inclusion/exclusion criteria exactly recruited the right population, being amyloid and tau positive, and are exactly in the middle of MCI and mild AD. And then, we had... You can consider it a minor problem in SAPPHIRE study because we didn't titrate in SAPPHIRE. The drug, we gave 800 milligram right away, with only one week of 400 milligram, and it led to a lot of tolerance problems. Patients discontinued the study, relatively high, in the blinded setting, 17%. But if you unblind the study, which was done already for SAPPHIRE, all of them were in the active arm, so we had more than 30% not tolerating the drug well in 12 weeks.
This is for an elderly and fragile population, not a good thing. So what we did in the VIVIAD study, and here we are still blinded, of course, we introduced a titration regime, and we did a minor dose reduction by about, probably 6% target occupancy. And you see that the blinded data discontinuation to AE reduced tenfold, which proves that we don't have an off-target toxicity. This proves that there is not a problem of giving a QPCTL inhibitor for, two years. We have patients now two years in the study, quite a lot. There is no problem. We asked the DSMB in Europe, we asked the DSMB in U.S., all is good. It's only the way how you treat these patients. You need a slow titration phase to get to the dose.
That shows you that we have a tenfold reduction in those, and it's probably the lowest discontinuations due to AE in the first half year you see ever in Alzheimer's. You're going to benchmark that. You can look at the publications. This is an extremely well-tolerated drug and data. I have only 2 minutes. Let me tell you what we do. We have the Cogstate Neuropsychological Test Battery as a primary endpoint in VIVIAD. We consider this an intermediate clinical endpoint, measuring very well cognition. We have also the Amsterdam IADL scale as a key secondary for measuring function in these patients. And with these two as an intermediate clinical endpoint, we believe we can, if the drug works adequately and the results are really positive, we can achieve accelerated approval.
Now, we have other endpoints like EEG and biomarker and safety, of course, which all will be released as headline data, either in the Q1 or at a conference later as the full data come out. And then, this one for the time, I think is already said. We have shown the progress, we have shown the regulatory strategy, and of course, we thinking about follow-up programs and other indications. We see China as an opportunity. There is the discussion of combination with monoclonal antibodies. I'm personally a little bit more skeptical on this, but it's more of a personal view on things because, if one drug works, first-line single agent, why would you take a monoclonal antibody, which just creates a problem if the efficacy is not synergistic, threefold higher?
Otherwise, you would do it stepwise. And, I want to end here, that our molecule and our company is very well positioned to improve health outcome in early AD. We work on a validated target in Alzheimer's. We work on a very logical and robust set of preclinical and clinical data. We are not limited to existing plaques. We are focusing on the neurotoxins where they actually are, and that is in the cells and in the synapses. And, we are aiming at very clear, that's the ambition, at first-line single-agent treatment of patients with early AD based on claims being well tolerated, no signs of ARIA, convenience, and a quite robust efficacy. Now, the data will tell in the Q1 2024. Thank you for your attendance. Questions? Critique. Yeah, please. Yeah, yeah, you. Please.
Maybe just one on the mechanism. I think you mentioned that the, the glutaminyl cyclase-
You get a mic, so that is... Maybe you-
Can you hear me? Yes. I think on the mechanism, you mentioned that the glutaminyl cyclase is not very efficient at catalyzing the cyclization of the N-terminal of the peptides. So does that mean that we should expect to be treating perhaps even longer, or what's the consequence of that?
Yeah, this is enzyme, enzymology at its purest. I'm very sorry. But the worse the substrate is, the less you need to inhibit the catalyzing enzyme to reduce the product. In other words, if you have a machine which very effectively manufactures new peptides, you need to. You can inhibit 80%, you still have enough product out there. If you have a machine which needs 10 attempts to make one efficient product, and you inhibit by 50%, you need already 20. You know, if you go to 80%, it needs 100 attempts to make one product, and then it is disappearing virtually. That means we don't need to inhibit this enzyme completely. And that's very important for the benefit-risk because you don't want to mess up with all the maturation of the other peptides.
We measure that in the study to be sure that we are not interfering with the thyroid pathway, with the GnRH pathway, with the sexual hormones, with the immune cytokines, and so on. So there must be enough QPCTL around to keep the physiological things, but enough inhibition to inhibit the unphysiologic thing, which we call therapeutic margin. And because this glutamate, luckily, is a bad substrate for that enzyme, we need not full inhibition and have a direct effect.
So you don't—I mean, it's a chronic treatment, by the way. I mean, if you stop this, it will start earlier or later. Of course, QC will down-regulate over the time. It will not start so rapidly, but it's considered a chronic treatment. But it is working from the first dose, so the inhibition of the pyroglutamate beta happens with the first dose w e don't need to accumulate over time. Actually, it's a positive thing that you need very little inhibition to get fully rid of the pyroglutamate, but have the pyroGlu of the other enzymes like CCL2, fractalkine , thyroid hormones, and so on, you still have.
Okay, thanks. And maybe the second one, if I may. Can you show the slide with the different endpoints? I wondered if you could guide us through the choice of the endpoints in the European study-
Yeah.
-and then the rationale for choosing a different endpoint in the second one.
It's a long debate and a long story, but first of all, we believe that rater-independent measuring of cognition, like with the Cogstate NTB, is much better in phase II-B studies than is doing ADAS-Cog, which is a completely insensitive and not a very good scale. This is a more sensitive and better scale from our point of view than the ADAS-Cog. The second thing is, if you target strategically accelerated approval with a study, you need to have robust evidence of efficacy, either on a surrogate endpoint or an intermediate endpoint. It is not having some trend on the CDR-SB on the final one; it is having robust efficacy on intermediate. We consider the NTB as an intermediate. The Cogstate and the intermediate one is approved in the US for measuring cognition, so the FDA knows that test.
But of course, as a full approval in the phase III study, it's not probably suitable. There's a debate whether or not, but would be risky. This is because. This is why in the second study, we have the CDR sum of boxes, which is a very convenient, easy 6-item scale, which is accepted by everybody as a primary endpoint for a full approval. But the advantage in going into an intermediate one is actually getting easier an accelerated approval. You probably know that lecanemab, how did they get it? They failed on the CDR sum of boxes in their phase II. It was not significant. They got it on robust evidence of efficacy of clearing plaques, which is a surrogate. They had p-values of 0.01 of clearing plaques at 1 year or 1.5, and this is why they got it.
So our strategy here is to go with an endpoint if we get robust evidence of efficacy on that endpoint as an intermediate endpoint, which gives probably much better trust and confidence that the product works on cognition than probably a PET. So it's probably better than a PET, but it's not as robust as the CDR sum of boxes. But we also go here in the hierarchical testing. So we look at function because you want to do two things in Alzheimer's. You want to see that the cognition, the slowing of the disease in terms of cognitive progression, is happening, and you want to see that it has a functional improvement for the patient. He needs functionally also to slow this, and we measure both, but in independent scale, the CDR sum of boxes takes this one.
Actually, this one already correlates with the correlation coefficient between 0.5 and 0.7, the CDR sum of boxes. So from our point of view, this is the smartest way to develop, going here with these endpoints in phase II-B, to have an opportunity, if things go well, for accelerated approval, and for this one, to upgrade it to phase III and go for full approval. That's the strategy.
Are you measuring CDR in the VIVIAD study? Are you measuring the CDR sum of boxes in the European trial as well, or?
No, no.
It is not relevant.
It's actually probably for 260 patients, a not suitable endpoint because the CDR sum of boxes, this study here is powered on 440 patients with a 35% efficacy and an 80% power. So if you have 260 patients here, you know what happens: You're totally underpowered because the CDR sum of boxes is not a very sensitive scale. It's not a genius scale. It's a 30,000-foot level scale, which has only six questions, three for function, three for cognition, and rating between zero and three, making you zero to 18 points. So it's a very rough scale, put it that way. It's a very...
It's good if you do a study in 20 countries with 2,000 patients, you have probably a quite good scale for high-level understanding it, and that's how you use it. You don't use it for sophisticated analysis of cognition in early studies or as intermediate endpoints.
Great. Thank you very much.
Very good question. Thank you. I think we're done with the timer. The guy says it's over. Thank you for your attendance, thank you for your questions, and hope to see you again.