Brain Imaging Identifies Six Subtypes of Depression

Precision psychiatry – urgent unmet need

Depression affects 3.8% of the world’s population. Approximately 30% of those diagnosed with depression have a form that is treatment-resistant, meaning that psychological and/or pharmacological interventions have proven unsuccessful in alleviating depressive symptoms. Unfortunately, such symptoms can lead to other comorbidities and, in some cases, suicide.

Depression treatment is an area of urgent unmet clinical need, but it’s one fraught with challenges. There are still so many unknowns regarding the underlying biology of the disorder and other psychiatric illnesses. Medications that prove effective in one patient may not yield any relief for another, leading some researchers to transition away from a one-size-fits-all approach to an alternative known as precision psychiatry, a subfield of precision medicine.

Dr. Leanne Williams, director of Stanford Medicine’s Center for Precision Mental Health and Wellness, the Vincent V.C. Woo Professor and a professor of psychiatry and behavioral sciences, is one such researcher. In 2015, Williams sadly lost her partner Jack to suicide after a decades-long battle with MDD. Now, her research focuses on pioneering efforts to generate personalized treatments for psychiatric conditions.

William’s latest study, published in Nature Medicine, identified six biological subtypes – or “biotypes” of depression by using machine learning (ML) approaches to group 801 depressed patients’ functional MRI (fMRI) scan data.

New image-processing procedure probes regions of the brain

Williams and colleagues implemented a standardized image-processing procedure, called the Stanford Et Cere Image Processing System, which probes six brain circuits previously associated with depression: the default mode circuit, salience circuit, attention circuit, negative affect circuit, positive affect circuit and the cognitive control circuit.

The Stanford Et Cere Image Processing System was used to analyze brain function in participants while they were at rest and while they completed tasks that tested their cognitive and emotional functioning. It was applied to brain scan images from 801 participants who had been diagnosed with depression and anxiety and 137 healthy controls. Approximately 95% of participants diagnosed with depression or anxiety were not receiving any medication for their illness at the time of the study.

Later, the same method was applied to brain scans from 250 participants taken after they had completed trials, which randomly allocated them to receive either one of three types of antidepressant medications, escitalopram, sertraline or venlafaxine (164 participants), or behavioral therapy used in conjunction with problem-solving techniques (86 participants). A machine learning method called cluster analysis was used to group the patients’ brain imaging data.

Six biotypes of depression could explain varying treatment responses

Williams and colleagues discovered six distinct subtypes of depression with distinct brain activity patterns that correlated with differences in symptoms, task performance and responses to treatment.

Comparing the baseline and post-treatment imaging data, the research team identified patterns of brain activity at baseline that were associated with improved responses to specific treatments. For instance, participants with a subtype of depression associated with overactivity in cognitive regions of the brain at rest responded well to venlafaxine compared to other biotypes.

Participants who had lower levels of activity in the brain circuit associated with attention control at rest were less inclined to benefit from talk therapy compared to other biotypes. This result makes sense, Dr. Jun Ma, the Beth Fowler Vitoux and George Vitoux Distinguished Professor in Geriatrics at the University of Illinois Chicago and a co-author on the study, said, because talking therapy equips patients with skills that they can utilize to address daily problems. Higher activity in brain areas associated with attention at rest might mean that these individuals are more inclined to adopt those new skills. One solution here, Ma added, could be to target lower brain activity with pharmacological methods prior to talking therapy.

“To our knowledge, this is the first time we’ve been able to demonstrate that depression can be explained by different disruptions to the functioning of the brain,” Williams said. “In essence, it’s a demonstration of a personalized medicine approach for mental health based on objective measures of brain function.”

Looking at symptoms and task performance, Williams and colleagues found that participants with overactivity in cognitive regions of the brain presented with higher levels of anhedonia – an ability to experience pleasure – and poorer performances on executive function tasks, compared to those with other biotypes.

Interestingly, one of the six biotypes did not present with any differences in brain activity compared to healthy controls. Williams said that, as the study focused on regions known to be associated with depression and anxiety, it’s possible that there are other regions affected in this biotype that the project failed to capture.

Testing new drugs for depression in different biotypes

In future studies with larger cohorts, the researchers plan to test more treatments – including drugs that are not traditionally prescribed for depression – across the six biotypes identified. They are also exploring ways to translate the research into clinical practice by creating easy-to-follow and easy-to-implement standards for the image-processing procedure.

“To really move the field toward precision psychiatry, we need to identify treatments most likely to be effective for patients and get them on that treatment as soon as possible,” Ma said. “Having information on their brain function, in particular the validated signatures we evaluated in this study, would help inform more precise treatment and prescriptions for individuals.”

Reference: Tozzi L, Zhang X, Pines A, et al. Personalized brain circuit scores identify clinically distinct biotypes in depression and anxiety. Nat Med. 2024. doi:10.1038/s41591-024-03057-9

This article is a rework of a press release issued by Stanford Medicine. Material has been edited for length and content.