Good afternoon, everyone. Welcome back to the HC Wainwright 27th Annual Global Investment Conference on September 8 to September 10, 2025. I'm Patrick Truchio, a Senior Healthcare Analyst at HC Wainwright. It's my pleasure to introduce our next company, Alto Neuroscience, a clinical-stage biopharmaceutical company pioneering precision psychiatry through biomarker-guided drug development. Alto is advancing a deep pipeline of CNS therapeutics using a proprietary platform that combines EEG, cognitive assessments, and wearable data to match the right patient to the right treatment. With that, it's my pleasure to introduce Amit Etkin, CEO of Alto Neuroscience. Welcome to the conference.
Thank you. Thank you for having us.
I think maybe first we can start by introducing Alto Neuroscience's Precision Psychiatry Platform, how it differentiates from the field in this area of neuropsychiatric drug development.
The bottom line of kind of clinical care as well as drug development these days is that there is no precision in psychiatry. That is the big step, I think, that we are trying to make that needs to be made. Right now, when we develop and we deploy drugs, it's based on clinical definitions of disorders that capture huge amounts of not just clinical but biological heterogeneity that results in generally failures in drug development and small effect sizes when they do succeed. What we're trying to do is anchor the development and deployment of drugs in biomarkers. What I mean by a biomarker here is a measurable attribute of somebody's brain function that tells us something about their disease.
Things like EEG or electroencephalography, brain wave recordings, which you could even do in patients' homes to get an idea of brain activity and connectivity, things like cognitive testing and looking at their memory and how they process information, all of which relates to brain function wearables. We do that systematically across our trials. A big focus then from a traditional precision medicine frame would be on using that information to select the patients most likely to respond. We think it's for a population like depression, around 30% to 50% of that broad population would qualify under any given biomarker for any particular drug as a rough estimate to give you a sense for proportions. We also use biomarkers as outcomes, as measurements of process, so that we're not reliant on waiting until really late-stage trials to show efficacy.
You can show that you're really getting the right engagement of the right target, frankly, in the way that the rest of medicine and even neurology employs. Think of neurofilament light chain as sort of a nice example there from neurodegenerative disorders. Applying that really for the first time here in psychiatry, doing it in a systematic way, we have a pipeline of drugs that all have some de-risking, in some cases a lot of de-risking, that allows us to say we can take this into a clinical program in phase two, learn a lot very quickly, embed biomarkers that are the same platform of biomarkers across our studies, understand placebo, understand drug effects, and as a consequence, hopefully bring differentiated drugs to market.
Can you talk about how a biomarker-driven approach could help address the high placebo response rates that have been present in neuropsychiatric drug development?
Yeah, I kind of joke that, you know, at some point we'll tell our kind of professional grandkids, as it were, remember there was a time where placebo was a big concern. Placebo is really only a concern when you don't have a lot of efficacy for your drug, right? Of course, there are high placebo rates. Another way of saying that is that there's low drug response rates. Part of the goal here is making sure we can predict better drug response, that our biomarkers predict better drug response in a reproducible manner. We do a lot where we have sort of a separate, untouched bit of data that we can replicate a biomarker on long before we then launch the large-scale efficacy study. That was the case for all of our programs. We also study placebo as an intervention with biomarkers.
We actually have a multiply now prospectively replicated placebo prediction biomarker. You can say, OK, I can predict drug. Does that predictor also work for placebo? Hopefully not. Then advance the program in a population more likely to respond to drug and potentially less likely to respond to placebo.
Can you talk to us about Alto 207? This is in treatment-resistant depression. Walk us through the mechanism of action, the therapeutic rationale, and how it came to be part of Alto's portfolio.
Yeah, this is a great example of where efficacy hopefully makes placebo concerns a thing of the past. We've been interested for a long time, really since our origin, in dopamine signaling. Dopamine is particularly important. It's thought for depression. It could be unipolar or bipolar depression. Looking for the right biomarkers to characterize it, once you have the right biomarkers, you can start to look for the right interventions. What that pointed to from a biomarker perspective is that not only in depression is there evidence of low dopamine, but in particular subpopulations like treatment-resistant depression, there's even lower dopamine. The question was, if we have that already biomarker definition of a population, even using clinical tools but backed up by a biomarker, what's the right way to address that deficit? That pointed us to direct dopamine stimulation, that is, of dopamine receptors themselves and a drug called pramipexole.
Pramipexole stimulates a group of receptors called D2 receptors, and especially this D3 receptor, important for mood regulation. The problem with pramipexole, which is a drug approved for Parkinson's, is that it causes a lot of side effects, especially as you start to increase dose, which means that you have to do it very, very slowly and you're dose limited. Those side effects are nausea and vomiting principally. Chase Therapeutics, from whom we acquired this co-formulation of pramipexole and ondansetron, found that ondansetron, when paired with pramipexole, diminishes the nausea and vomiting, does not diminish, though, the antidepressant effect. They found in a phase 2A study a Cohen's D of 1.1 effect size, more than 8 points on the moderate difference between drug and placebo.
For frame of reference, in treatable depression, a 0.3 effect size, 2 to 3 points, is considered not only successful but can be very commercially successful, as many drugs have seen. This is three times that. We also saw data from a study in treatment-resistant depression that a group at Oxford led of very hardcore treatment-resistant populations, 10-year median duration of episodes, four failed drugs, dosing in that case pramipexole by itself very slowly, but seeing a Cohen's D of 0.9 effect size. Again, large effects. In that case, unfortunately, 20% of their population dropped out due to these side effects, only 5% with placebo. That's exactly the rationale for the combination approach, which also allows us, by the way, to dose five times faster. Chase was able to get to their therapeutic dose, which was higher than the Oxford study, within about a week.
That really has the potential to be a transformative medicine, not just for treatment-resistant major depression, but ultimately for treatment-resistant bipolar depression and other areas. We can further enrich even from there with biomarkers for either efficacy or tolerability. That's the kind of differentiated outcome we're hoping to bring to a field that's frankly desperately needed in.
What role do you see biomarkers playing in the phase 2B trial?
The phase 2B trial will predefine biomarkers for both efficacy and tolerability. By predefining them, that'll allow us to see which ones hold water in terms of their actual predictive ability. Because this is already a phase 2B that we think may be part of a registration package directly, we are anticipating that gives us one phase three that we would need to do afterwards, where we would take the biomarkers that we defined from the phase 2B, predefine that in a phase three, and then if positive, that would be an enrichment marker to complement a broad TRD approval strategy for that drug. It becomes a great pharmacoeconomic argument as well for why a branded therapeutic should, with that kind of effect size, be much further up in line than a typical branded would be in a large population like depression.
I want to move next to Alto 101 and cognitive impairment associated with schizophrenia. Just this morning, though, Alto announced independent replication of theta ITC as EEG biomarker in CIAS. Maybe if you could frame for us first Alto 101, the mechanism and the background on this drug, and then as well the publication from today, how does this impact the program?
Perfect. Cognitive impairment that's associated with schizophrenia is the basis of the disease. We all, as you kind of live life, right, associate psychosis as the cardinal symptom of schizophrenia. Actually, most of the time, these patients spend—they're not psychotic. They're cognitively impaired and disabled because of that cognitive impairment. It starts earliest in their life and is most persistent. It does not respond to any treatments. There are no drugs approved for cognition in schizophrenia. You're looking at a disorder that affects roughly 1% of the population. This is the basis of it. There's no established treatment. Huge opportunity, of course, not without risks and not without challenges. Our approach to that is, again, to take a biomarker perspective. What I'd like to do is have a drug where I can show the right target engagement with, in this case, an EEG biomarker.
I can select patients who are going to be the most likely to have an abnormality in that biomarker and to show a clinical effect. It starts to sound a lot like, actually, as I mentioned, stories from neurology where you're using that kind of a biomarker approach. We looked at large data sets of patients with schizophrenia and controls and put all of the described biomarkers head to head against each other and then replicated that and found that this one biomarker, theta ITC, which is basically the brain's kind of rhythmic response to getting sensory stimuli, is the most sensitive to the disorder, cases versus controls, and the best correlated with cognition.
What we reported this morning was in yet another independent large data set, exactly the same held true that it was the most robust marker of the disease, more so than any other established biomarker for schizophrenia. That makes it a great primary outcome for a proof of concept trial, which is what we're currently engaged in targeting later this year as the readout in patients with schizophrenia with this drug. Now, what is that drug? Alto 101 is a PDE4 inhibitor. PDE4 inhibitors, you I'm sure know from the immune system, drugs like Otezla, very successful anti-inflammatory drugs. This is an attempt to bring that to the brain, not so much for its anti-inflammatory effects. That's a wonderful upside, but more so because what a PDE4 inhibitor does is increase a second messenger called cyclic AMP that's important in plasticity and cognition.
I've actually studied that way back to my MD PhD days. It's been of tremendous interest for decades. The question is, why hasn't it been leveraged for the brain? That answer is actually quite simple, which is, number one, we've not known what the right targets are in the brain. Number two, tolerability has been a challenge because the more you give, again, more of that nausea-vomiting type effect that we talked about with Alto 207, that's really been dose limiting. Even for drugs like Otezla, that's been a challenge, and that barely gets into the brain. What we did is we reformulated the drug from oral to a transdermal patch, slowed down the absorption of the drug, and that really cut out these adverse events. The drug continues to be well tolerated. There's really been no meaningful PDE4 side effects in the ongoing phase 2 trial. That's great.
It means that we have a drug there, and obviously, we're hoping that's a drug for schizophrenia. In phase 1, we saw a dose-dependent effect of the drug on that theta ITC EEG biomarker. We also saw the exact same thing in animals. We saw an improvement in cognition in our phase 1 study, even with single doses in healthy people, and the improvement in cognition correlated with the improvement in the EEG biomarker. That's a pretty good setup for going into that proof of concept trial. The primary outcome, as I mentioned, is EEG. That's kind of the go/no-go from a mechanistic perspective here. We'll also look at cognition. That's a second set of secondary outcomes. 10 days of treatment as a crossover design, so you have drug and placebo all within the same person.
10 days of treatment is thought to be pretty short for moving cognition in schizophrenia. Normally, you'd think of like 12 weeks as the period of time you'd be treating patients. Our expectation is not that we change cognition. If we do, that's a phenomenal outcome. The expectation here is that we're engaging the right target with that EEG biomarker. Of course, tolerability is an important outcome. As I mentioned, it's a drug, right? At the very least, it's a drug for the immune system now that we've solved a longstanding tolerability challenge. Our hope is that we have a drug for schizophrenia, and then disorders from there become a really interesting opportunity set.
Terrific. Maybe moving to Alto 300, this is a major depressive disorder. Maybe talk to us about this drug, mechanism of action, how it came into the pipeline.
Alto 300 is a drug called agomelatine. You can tell some of our approach at de-risking and diversifying here. Agomelatine is an approved antidepressant in Europe and in Australia. It's a novel chemical entity here in the U.S., and our focus is on a U.S. development strategy. We know that it works for depression. What we are looking then for is the opportunity to develop it and enrich for a population defined by a biomarker. The reason that you'd want to do that in this case is because it has a unique mechanism of action relative to other antidepressants, which means you're not going to just enrich for something where you can replace it with Zoloft and you get the same effect. What we did here is study it adjunctively in depression.
We know in MDD, the greatest need is in those people who failed first-line treatments, of which that's the majority of the population. That's really where a branded therapeutic would be used. As a psychiatrist, I wouldn't obviously even have the opportunity to give somebody a branded therapeutic first-line. When they're coming and having failed something, I probably wouldn't be taking away that underlying antidepressant if it showed even any signal for them. That's why we study it as an adjunct. In this case, we found a biomarker using a machine learning approach that we developed in my lab at Stanford, where we found with EEG and the machine learning strategy, a biomarker that we could then replicate in a separate population to predict better response. Just like we were talking about earlier with placebo, it does not predict greater placebo response. That's how we're selecting patients in the study.
It describes about half of the MDD population as being positive for this biomarker. We're studying primarily patients who are positive. That's our powered primary outcome population, and then a small population that is negative to start to get at questions of enrichment, which ultimately are a phase three question where you more fully sample the range of people. There's an adjunct treatment that's extremely well tolerated, unlike antipsychotic medications, which are the primary adjunctive treatment in depression. We think it could make a very exciting product to bring into the clinic. That's in a phase 2B. That is on track for a mid-next year readout. We conducted an interim analysis in that study after our Alto 100 study didn't hit its primary endpoint.
We identified the reason why, not being a drug or biomarker issue actually in this case, but more of a site-level execution risk around bringing in professional patients. We did a blinded case review for Alto 300, found sites that have that risk profile, removed their data. That is out of the analyses. Conducted an interim analysis, showed drug-like signal supportive, continued to move forward with the study, obviously replaced those missing patients, and then reading out next year.
What are the expectations for that phase 2B trial? What would you need to show there to give confidence to move forward to phase three?
I think statistical significance would be the bar, just given the clinical need in terms of adjunctive treatment in depression. It would be a valuable asset to bring in. Obviously, the bigger the effect size, the better.
Alto 100 is being brought forward in bipolar depression. Maybe you can give some background, the learnings that were generated from that program, and how you see this next readout.
Yeah. To kind of double-click a bit on the learnings, as I mentioned, it was really a professional patient issue. That is, those people, in this case, who are taking the drug as monotherapy, brought in by certain sites, had a high non-compliance rate not due to side effects. The drug has been consistently very well tolerated. If you take out that component and you say, all the people who are on drug, did we see an effect in depression? The answer is yes. Actually, there was an effect and a biomarker signal. Also, the adjunctive population within that MDD study, which was a pre-specified stratified population of about 30% of the total sample, showed a Cohen's D of 0.47, nearly significant despite being not powered for significance. That told us that there's probably a signal there that merits continuing to study the drug.
Bipolar depression is an area where there are very few available treatments. The only available treatments are antipsychotic medications, which, as we've already discussed, have a tremendous side effect burden. The opportunity is really clear. You have a drug with a novel mechanism and an interesting, easily scalable biomarker using a memory test to identify people, a population in clinical need also that has a higher prevalence of cognitive impairment than depression. You already have a target population that's, if you will, more enriched. As well, a funder here in Wellcome Trust that helped fund a majority of that study. We brought in about $12 million in partnership with Wellcome. That's really an exciting proposition for us.
We think that the kind of underlying concept here of targeting neuroplasticity with a novel neuroplasticity-enhancing agent, a drug identified based on a functional screen for neuroplasticity done at the cellular level, looking at neurogenesis, with a way to identify those patients who have a more prominent neuroplasticity deficit and having a deficit in learning and memory is a pretty targeted approach for a well-tolerated drug that could be really transformative for this population. That's on track for a second half readout next year, also as a phase 2B. You can see a lot of really late-stage, potentially very catalytic events in high clinical need areas.
Here too, just given the unmet need in bipolar depression, statistical significance is that kind of is at the bar.
That's right.
For Alto 300, maybe you can talk to us about this program and MDD and kind of why it's unique and the data and the path forward.
Alto 203 is a drug we read out in June. It's an H3 inverse agonist that has unique effects on subcortical reward system dopamine, as well as H3 effects on things like arousal and wakefulness. That was a much, much earlier stage study. It was a single-dose pharmacodynamic study, just basically trying to understand what does the drug do in that case in patients with depression. What we found is that what the drug does is improve cognition and increase wakefulness, changes in EEG biomarker of cognition. It's well established. That's actually an FDA-cleared measure of ADHD. We found that that biomarker and, in fact, replicated that that biomarker also, in turn, predicts better cognitive effects of the drug. All of this is compared to placebo. Those are single-dose effects.
Where we currently are with that program is because we have a full pipeline and have been focusing on the later stage programs, as we've been discussing, we're not currently advancing that program on existing funds. That is not to say that the program doesn't have a lot of value. It's a cost of capital question, potentially options for partnership, other ways to develop that program. It's very clearly an active drug, has very clear effects that point to what the interesting populations to target here would be, given its effect on increasing both attention and wakefulness. That is something we'll look to do more with as time goes by. For the moment, really focusing on those things that will reach the clinics soonest, which is our latest stage program.
Just as a final question, just to reiterate, we've discussed throughout, but just to reiterate, the main milestones that investors need to be aware of over the next 6 to 12 months, and then also what you think investors may be missing about the story.
Yeah. So.