Psychiatry needs new treatments that do not rely on guesswork. We are making that happen.
What if the basic assumptions behind our approach to mental health treatment are wrong?
What if the failure to develop effective therapies is a failure of imagination?
I left Stanford and founded Alto Neuroscience — a precision psychiatry biopharmaceutical company — to answer these questions, and by doing so, transform the treatment of mental health conditions. Work in my Stanford lab has pointed to the path forward, but to follow that path and bring entirely new treatments to patients, I had to bring it out of the lab and into the real world.
The mental health epidemic all around us
Mental health is the challenge of our day, yet we still rely on guesswork to treat patients. Our treatments are largely the same as those from 50 years ago. Something has to change. It’s time for a revolution. It’s time to bring precision to psychiatry.
Because of the pandemic, we’re hearing more about the mental health crisis. And yes, I’m pleased that the crisis is receiving some of the attention it deserves. But know this: It has been a crisis long before the first COVID-19 case was reported.
These numbers should shock and dismay, and I haven’t even mentioned other common conditions such as schizophrenia, bipolar disorder, anxiety, autism and ADHD – to name only a few. How can a group of conditions that are the single largest source for suffering, disability and cost in the world lag so far behind the rest of medicine?
It doesn’t have to be this way. For too long, we’ve matched the treatment to the diagnosis, not to the patient. And these diagnoses are by their very nature imprecise. Diagnoses are made based on a checklist of symptoms, with many different combinations still resulting in the same diagnosis. As just one example, the diagnosis of PTSD can be made in over 600,000 different ways. It simply makes more sense to focus on the patient, both in terms of advancing the science and translating to ultimate clinical care. We call this precision psychiatry.
In fact, in my experience, patients don’t care much about their diagnosis: they care about their treatment. They want to feel better.
And most are not feeling better.
A failed status quo: Trial and error
Finding the right therapy for all psychiatric conditions remains, as it has for decades, a matter of trial and error.
Trial and error is an imprecise process. Patients hate it. Clinicians hate it. It prolongs patient suffering, increases health care costs, and contributes to the general perception that antidepressants, for example, are ineffective.
Unfortunately, patients and clinicians don’t have many options. Look at PTSD, MDD, and substance use disorders: In more than 20 years, only one new drug mechanism has been approved: esketamine, for MDD. No new drug mechanisms have been developed for PTSD or substance use disorders. So clinicians work with what they have, which isn’t much.
Why so few options? It’s a confluence of factors, and the two most important are closely related.
After continual development failures, Big Pharma has for years been pulling out of psychiatric research, leaving the pipeline dry. Related to that is the apparent lack of efficacy of therapeutic candidates. The vast majority of potential therapies never made it out of clinical trials — not because they were unsafe, but because they were deemed ineffective.
But were they really ineffective? What if many of the drugs always worked, but we were thinking about mental illness the wrong way? What if the very way drug development has always been done destined them to fail?
Fundamentally, here’s what I know: The same drug can be both life-changing and useless – it just depends on whose brain it is in. By harnessing this insight, we can bring true innovation to patients.
Different brains, different responses – and a new opportunity
In therapies that won approval and in those that did not, only a minority of patients responded well to the therapy. Because the ultimate effect of a drug is judged based on the average across all patients in a trial, the overall effect seems small in both the approved and unapproved drugs. What differentiates the drugs that gained approval from those drugs that haven’t? In part slight differences in how big the proportion of responsive patients is – both in general and across individual studies performed.
All of this assumes, of course, that the drug was even tested in the right diagnosis. In many cases, the decision of which condition to test a drug in is based on assumptions developed from work in animals. Unfortunately, the past few decades have likewise taught us that there are no definitive animal models of human mental health conditions. As a consequence, preclinical work translates poorly into effective human treatments.
Clearly, we’ve been going about developing new medicines the wrong way. If patients respond to the same drug in different ways, then we need to find who the right people are for each drug instead of a one-size-fits-all approach for everybody with a particular diagnosis. Patients and clinicians are crying out for precision psychiatry to become a reality.
Seen this way, many of the drugs deemed ineffective may actually be quite effective, but only for specific and biologically definable subpopulations. Likewise, many drugs may have not even been tested in diagnoses with responsive subpopulations.
What this perspective also means is that we can accelerate the development of new precision medicines by starting with the right candidate drugs that have been brought into human trials by pharma; that means we already have safety data.
The differences between patients’ brains, long recognized but long ignored, is the basis of how we develop new medicines. Our mission is to redefine psychiatry using AI-enabled brain tests that identify responsive patients in order to develop personalized and highly effective medicines, helping patients get better faster. By starting with drug candidates about which we have already have safety data in humans, we can make this mission a clinical reality in the near term.
Think differently to do differently
It’s time to revolutionize psychiatry and transform mental health care.
Alto is developing a new generation of psychiatric drugs that are more effective because they are tailored to individual patients, who are identified based on their biology. We accomplish this by marrying our proprietary pipeline of potential therapies to a suite of biological tests of brain function (which we refer to as biomarkers) with AI; this helps us understand which biological tests best predict which patients will respond to a given drug. A biomarker can be as simple as reaction time, an electroencephalography (EEG) pattern, how much a patient sleeps, or a combination of these factors. All of these are measures of brain function.
Computerized tests of cognitive and emotion-related behavior, and wearables that measure sleep and activity patterns, can likewise be powerful tools for aligning the right Alto drug with the right patient. Our vision is that patients will be able to take biomarker tests in their doctor’s office or even at home.
This approach – precision psychiatry – yields benefits for the patient, the clinician, the payor, the health system, and society. Critically, it makes logical sense, after decades of not learning the lesson that faces us every day as we seek to improve the lives of our patients. The ultimate goal is simple: Provide the right treatments to the right patients at the right time.
Focus on what can be manipulated and measured
Our approach for both selecting candidate drugs and for identifying biomarker tests for drug responders focuses on core domains of mental functioning relevant to a broad range of mental health condition, and the brain circuits that underlie them:
- Cognition, which includes attention, decision-making, learning/memory, and neuroplasticity (i.e., the ability of the brain to change)
- Emotion and its regulation, which includes both excessive negative emotion (like anxiety) and impoverished positive emotion (including motivation), which are often difficult-to-treat features
- Sleep and activity, which includes alterations in either sleep or circadian rhythms
The brain circuits underlying each of these can be manipulated through a variety of novel drug mechanisms, and the functioning of these circuits can be assessed either directly (with EEG) or through their behavioral outputs (with computerized cognitive and emotional tests, and wearables). These biomarker tests tell us what our drugs do in the human brain, and which are the right patients for their therapeutic effects.
Given that all of our drug candidates are already clinical-stage, we’ll know soon what works and for whom. Just as important, because our approach gets stronger with every new dataset we collect, our precision psychiatry tools and insights become sharper regardless of whether a drug ultimately succeeds or fails. This can’t be said for the traditional ways drugs for the brain have been tested.
No more guessing games. We have been approaching mental health treatment in all the wrong ways until this point. We have failed patients, their families and their clinicians.
In almost every other area of medicine, we’ve seen tremendous advances. Mental health has been left behind. But no longer. Precision medicine for the brain is here.
Just watch what happens next.
1. Tam J, Mezuk B, Zivin K, Meza R. U.S. Simulation of Lifetime Major Depressive Episode Prevalence and Recall Error. Am J Prev Med. 2020;59(2):e39-e47. doi:10.1016/j.amepre.2020.03.021
2. Major Depression The National Institute of Mental Health
3. National Center for Health Statistics Data Brief No. 303, February 2018, CDC https://www.cdc.gov/nchs/products/databriefs/db303.htm
5. Key SAMHSA Substance Use Indicators in the US
6. Roehrig, C.: Mental Disorders Top the List of The Most Costly Conditions In The United States: $201 Billion. Health Affairs (2017)
7. Figueroa JF, Phelan J, Orav EJ, Patel V, Jha AK. Association of Mental Health Disorders with Health Care Spending in the Medicare Population. JAMA Netw Open. 2020;3(3):e201210 doi:10.1001/jamanetworkopen.2020.1210
8. Wu, W. et al.: An electroencephalographic signature predicts antidepressant response in major depression. Nature Biotechnology 38 (2020)
9. Zhang, Y., Wu, W., Toll, R.T. et al. Identification of psychiatric disorder subtypes from functional connectivity patterns in resting-state electroencephalography. Nat Biomed Eng5, 309–323 (2021). https://www.nature.com/articles/s41551-020-00614-8