By Jonathan D. Grinstein, PhD
As a tenured professor, pockets lined with funding, Amit Etkin, MD, PhD, had it all at Stanford. That is until he realized that what he really wanted was not just to publish papers that proposed therapeutic solutions for psychiatry but to actually see new treatments through to the finish line that changes people’s lives.
Etkin told GEN Edge that, around 2017, he understood that he would be unable to get the resources and run large-scale studies on humans to develop new interventions.
So, in 2019, he left Stanford to start Alto Neuroscience, and four years later, the company announced positive results from its third Phase II study in 2023.
Alto presented positive results from its Phase IIa trial for ALTO-300 in major depressive disorder (MDD) at the 62nd Annual Meeting of the American College of Neuropsychopharmacology (ACNP).
The eight-week clinical trial examined possible biomarkers that could show the safety and efficacy of ALTO-300 works as an adjunct treatment for people with MDD who were not responding well to an antidepressant. ALTO-300 demonstrated a favorable safety and tolerability profile with no unexpected adverse effects.
The company also said it initiated a Phase IIb study evaluating ALTO-300 in 200 patients with MDD, with results expected in the first half of 2025.
This is the third set of Phase II results from Alto in 2023. At J.P. Morgan 2023, Alto announced Phase IIb results for ALTO-100, a candidate that treats a different depression phenotype.
A hill to die on
The state of psychiatric medicine is nowhere near where other clinical disciplines like oncology are. That is, in part, due to the way the disease is defined.
For example, in oncology, a tumor sample is characterized, molecular pathways are identified, and drugs are developed to target those pathways.
But in psychiatric medicine, identifying patient populations with biomarkers and finding the drugs that drive clinical responses remain unsolved problems.
According to Etkin, this presents two distinct puzzles. If a patient arrives with depression, he wants to know if the patient can be classified into a subtype of depression that corresponds to a treatment that produces a biomarker response and clinical benefit. Similarly, if he has a drug with an entirely new mechanism of action but does not know who it is for, it is not very useful.
The answer for Etkin is tools that enable precision psychiatry so that patients can be diagnosed and treated with greater specificity and accuracy by virtue of a better understanding of their disease state at the biological level.
“The vision here was crystal clear, which is that precision psychiatry is going to have to be the way forward, and that has to be the hill that we’re willing to die on,” Etkin told GEN Edge.
The beauty of this, according to Etkin, is that the platform can create generalizable insights about biology rather than arbitrary definitions that started in the 1970s when the modern Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR) was introduced.
“We’re going to figure this out and hack our way into a solution that’s reproducible and generalizable,” said Etkin.
You can’t manage what you can’t measure
The team at Alto created a systematic, rigorous data science approach that employs machine learning to analyze highly robust, measurable, and meaningful biomarkers. Only a few things fit those criteria and have cross-sectional and longitudinal data from placebo and drugs for various disorders: electroencephalogram (EEG) activity, neurocognitive task performance, and wearable data.
“An old sort of adage from engineering is that [you can’t manage what you can’t measure],” said Etkin. “Being able to measure then instructs you on how you want to manipulate the system, and then it’s this dance between the drug side and the biomarker side.”
The team at Alto spent two years in stealth building this platform and looking through existing data sets for new drugs that target pathways not yet represented in the standard of care and have been developed and tested in humans through at least Phase I. These candidates could then be quickly entered into clinical trials and tested to prospectively replicate the ability to find those responders with Alto’s biomarkers.
“Despite where psychiatry is today, the really good news is that we’re starting with drugs in people in whom we’re looking at outcomes like a change in depression,” said Etkin. “So it’s inherently meaningful; there’s validity to that outcome.”
Today, Alto’s clinical-stage pipeline includes novel drug candidates for depression, PTSD, schizophrenia, and other mental health conditions.
Etkin said, “The thing that we put in the last paragraph of papers when I was in the lab, which is the thing you’d like to be able to do—now we’re doing it.”
The flywheel effect
Alto reported positive results from an eight-week clinical trial that looked at possible biomarkers for predicting the safety and effectiveness of ALTO-300 as an extra treatment for people with MDD who were not responding well to an antidepressant. The study enrolled 239 patients between the ages of 18 and 74 who remained on a background antidepressant while ALTO-300 was added as a new treatment. At the start of the study, 110 of these patients had an EEG.
Alto employed a strict data science method to analyze the EEG data based on machine learning, which necessitates future replicating a biomarker’s ability to predict how well an antidepressant will work in a different group of patients.
To do this, the Alto team looked at whether patients with this EEG biomarker had stronger clinical improvements in their depression symptoms and higher response rates to ALTO-300, as measured by the Montgomery-Sberg Depression Rating Scale (MADRS), compared to patients who did not have the EEG biomarker.
“That’s exactly what we saw—a substantial difference where those people who were predicted to do better did do better compared to those who were predicted not to do as well,” said Etkin.
The predictive biomarker only works for patients taking ALTO-300. It does not work for patients taking placebos or standard-of-care selective serotonin reuptake inhibitors (SSRIs) and serotonin and norepinephrine reuptake inhibitors (SNRIs). This is based on data from studies that used EEG measurements to tell the difference between patients with this biomarker and those who didn’t.
Alto Neuroscience now has two large Phase IIb studies, one for ALTO-100 and one for ALTO-300, in different subpopulations with depression. Each candidate has a biomarker to describe different aspects of phenotypes that all fit under the umbrella of depression.
ALTO-100 addresses reductions in neuroplasticity, which Alto is characterizing through a cognitive exam. ALTO-300 targets the reward system and circadian rhythms, which is easily measurable.
“In both cases, we are in near real-time selecting patients across sites based on cognition or EEG in a way that is precision psychiatry being done now,” said Etkin. “We’re very excited to keep pushing the approach, and then on top of that, we have two more drugs entering proof-of-concept trials next year that are different mechanisms, different populations.”
So far, Alto Neuroscience has demonstrated that the mechanics of its platform-driven flywheel are in place to realize precision psychiatry. We will soon find out what heights they will be able to reach.