Tired Man Sleeping in the Bad with a Clock Alarm on the Table

Here’s what’s happening this week in mental health research. New studies show a strong link between insomnia and depression in older people. Scientists are developing exciting new tools for tracking the activity of brain circuits. And new studies suggest that drugs may make therapy more effective at treating PTSD. Read on to learn more.

By Tom Insel, MD

Focusing on the Where and the How of Brain Function

Human brain stimulation or activity with neuron close-up 3D rendering illustration. Neurology, cognition, neuronal network, psychology, neuroscience scientific concepts.

The Holy Grail for neuroscientists has always been to understand how the brain processes information at the speed of thought. Sensation, emotion, memory – the where and the how of these subjective experiences remain mysterious. With modern imaging techniques, neuroscientists are beginning to solve the where mystery, identifying specific circuits that appear to encode or store relevant information. But as Katrin Amunts and Thomas Lippert point out in a Perspective this week in the journal Science, the pictures we have from human brain imaging, known as the macro-connectome, are far too coarse to understand the actual processing of information. That requires the micro-connectome, which maps brain connections at the level of individual cells with their millions of connections.

While techniques are improving to measure the micro-connectome in worms and flies, Amunts and Lippert remind us that even if we could make these kinds of measurements in the human brain, the required computing power to collect, store, and analyze the data from the whole human brain would be overwhelming, beyond current computing capabilities. They estimate that one current project, called Cellular Big Brain, which gets almost to the level of individual cells in a tiny cube of human brain will require two or three petabytes of data. To get to micro-connectome-level resolution would require close to 1,000 petabytes. A single petabyte is roughly the equivalent of 500 billion pages of standard text.

Even if neuroscientists can map the circuits where the activity is happening, answering the how question of brain function will require tools for modulating those circuits. There are already tools, such as optogenetics, that activate or inhibit an identified circuit by aiming light at cells that have been genetically engineered. These tools have revolutionized neuroscience by permitting individual circuits to be studied in freely behaving animals, revealing the neuroanatomy of sleep, feeding, fear, stress, and many other brain functions. The problem is that the activation is artificial and interferes with the normal timing or synchronization of circuit activity.

A new technique developed by Kafui Dzirasa and his colleagues at Duke University might help. They have designed a new way to regulate circuit activity in laboratory animals. The new technique, called LinCx (Longterm Integration of Circuits using Connexins), leverages the brain’s electrical connections, known as gap junctions, and the proteins, called Connexins, that are found there. By genetically engineering connexins from fish to express in mammalian neurons, Dzirasa and his colleagues were able to regulate the communication between cells in a mouse brain circuit through the docking of connexins with each other, respecting the natural context of the inputs and outputs of the circuit.

LinCx influences individual cells with a precision that has not been possible previously. We’re a long way from using this technique in humans, but LinCx brings scientists closer to the Holy Grail of mapping brain activity at the speed of thought.

Brain research challenges supercomputing  Science, Nov 26, 2021

Long-term precision editing of neural circuits in mammals using engineered gap junction hemichannels BioRxiv Nov 22, 2021


Want to Prevent Depression in Older Adults? Treat Insomnia. 

Tired Man Sleeping in the Bad with a Clock Alarm on the Table

So much progress in public health over the past century has been due to prevention. Better sanitation, improved nutrition, and vaccines are all part of the explanation for increases in life expectancy and reductions in disability. In mental health, prevention remains more of an aspiration.  There has been considerable research on prevention, much of which was noted in 2009 and 2019 reports from the National Academy of Medicine. In a recent essay in Nature, Ricardo Muñoz summarized the evidence for preventive interventions for post-partum depression, concluding that the efficacy was roughly equivalent to the influenza vaccine. But much of the literature on prevention has found small effects and, unlike preventive interventions in other areas of medicine, very few preventive interventions have been adopted in mental health care. 

So a new randomized clinical trial of a preventive intervention for depression in older adults deserves note. Unlike most previous studies that have focused on people with early symptoms of depression, this trial focused on insomnia. Nearly half of older people suffer from insomnia, and it doubles their risk of depression. In this new study, led by Michael Irwin of UCLA, nearly 300 older patients with insomnia were recruited and randomly assigned to either cognitive behavior therapy for insomnia or a control condition of sleep education therapy. They were not seeking treatment for depression (anyone with current depression was excluded) and the interventions were not framed as mental health treatments. This was strictly a treatment for insomnia, delivered in a group setting every week for two hours over two months. Outcomes were followed for three years.

Over the follow-up period, depression was detected in 25.9% of the control group but only 12.2% of the group getting cognitive behavior therapy. The lower rate of depression in the experimental group was associated with lower rates of insomnia, suggesting that treating insomnia was an effective preventive intervention for depression. Indeed, the incidence of depression in the treated group was about the same as in the general population without insomnia.  

This is one of the largest and best-designed studies of a preventive intervention for depression. The ease of delivery and the profound reduction in depression means this cost-effective intervention should be widely adopted. Depression in older adults is also associated with higher rates of dementia and suicide, so a safe, inexpensive intervention that can reduce depression by 50% deserves a close look by health systems.

Prevention of Incident and Recurrent Major Depression in Older Adults with Insomnia  JAMA Psychiatry Nov 24, 2021

Can drugs augment psychological treatments for PTSD?

One of the most effective psychological treatments for post-traumatic stress disorder (PTSD) is exposure therapy – putting a traumatized person in contact with the cues they associate with their original trauma. In the process, the person either becomes habituated to or is able to extinguish their fear. Unfortunately, many people with PTSD have trouble extinguishing fear and don’t benefit from this treatment.

Studies of fear extinction in laboratory animals have provided a useful way to understand the biology of this process. They show that the stress hormone cortisol facilitates fear extinction, possibly by stimulating NMDA receptors in the brain. Now a new study asks the question: Could cortisol (given as the drug hydrocortisone) or D-cycloserine (a drug that directly activates NMDA receptors) help extinguish people with PTSD?

Sabra Inslicht and colleagues from UCSF and the San Francisco VA Medical Center addressed this question in 90 people with PTSD. This wasn’t a treatment study but a rigorous laboratory investigation of the effects of these drugs on fear conditioning, using physiological responses as a measure of habituation. The results were surprisingly clear: a single dose of both drugs enhanced fear extinction compared to placebo. There was also a trend towards greater retention of extinction one week following d-cycloserine treatment.

So will people with PTSD who aren’t helped by exposure therapy do better if hydrocortisone or D-cycloserine is added to their psychological treatment? We will need a clinical trial to figure that out. But these new findings, limited as they are to laboratory responses, remind us that the combination of medical and psychological interventions may be more effective than either one alone.

Randomized controlled experimental study of hydrocortisone and D-cycloserine effects on fear extinction in PTSD   Neuropsychopharmacology, Nov 19, 2021

Diagnosis by Any Other Name

3D illustration of DSM-5 script on a book, isolated on green gradient.

Diagnosis has always been a contentious topic in mental health. Some of the founders of modern psychiatry, like Adolph Meyer of Johns Hopkins, rejected the very notion of diagnostic categories. But in the aftermath of World War II, as veterans were presenting with a range of mental conditions, the American Psychiatric Association pushed for a manual to bring some order to what was becoming a chaotic field. The first diagnostic and statistical manual, DSM-1, was released in 1952. The subsequent DSM editions have arrived with increasing debate about reliability (will two people seeing the same person make the same diagnosis?), validity (do the categories represent distinct disorders?), and relevance (do these diagnostic categories predict treatment response?). And then there is the Adolph Meyer question about the very concept of categories. Is there a bright line between having a disorder and not having a disorder?

We have data to inform this debate. An early study of DSM-5 showed a wide range of reliability, with little agreement for some of the most common disorders. Biological studies of genetic architecture and brain imaging demonstrated shared features across disorders, rather than abiding by the clinical categories of DSM-5. But the architects of DSM never intended the manual as a way to predict treatment response; it was always designed to describe symptoms.

A new study by Jennifer Newson and colleagues from Sapien Labs examines symptom clusters in over 100,000 adults from eight English-speaking countries. Using the Mental Health Quotient, an anonymous, online, self-report tool that evaluates symptom profiles across 10 common mental disorders, the team looked for a bright line between “normal” and “disordered,” but found a continuum with no clear separation. For people with symptoms that mapped on to a DSM diagnosis, the differences within a diagnostic category were almost as high as the differences between any two categories. In other words, the categories were not capturing a single group.  Indeed, the DSM clusters could not be separated from random clusters of symptoms.

This study, which used online self-reports from individuals in the Mental Health Million database is, of course, taking a different path to diagnosis than DSM-5, which used clinical assessments by trained diagnostic raters in treatment-seeking individuals. So perhaps we should not be surprised that these self-reported symptom profiles do not line up with DSM-5 categories. This study argues that an empirical approach that focuses on self-reported symptoms rather than diagnosis should improve clinical and research outcomes. Which approach will prove more clinically useful? That remains to be seen.

Poor Separation of Clinical Symptom Profiles by DSM-5 Disorder Criteria  Frontiers in Psychiatry Nov 29, 2021



Tom Insel, MD, is a psychiatrist, neuroscientist, and former director of the National Institute of Mental Health (NIMH). He is a donor to MindSite News and chair of its Editorial Advisory Board. Dr. Insel’s financial conflict of interest statement, which includes equity and advisory roles in several early-stage mental health technology companies, can be found here.