Stanford researchers use imaging technology and machine learning to identify six subtypes of depression

We Need a Better Way to Match Patients depression With the right treatment, Stanford researchers found that Today's Report Nature MedicineIn their study, they say they identified six biological subtypes, or biotypes, of depression. The team also identified existing treatments that may be more or less effective against three of those biotypes.

According to statistics from the National Institute of Mental Health: 2021 National Survey on Drug Use and HealthApproximately 21 million people in the United States (8.3% of the population) suffer from major depression. This rate is significantly higher among women (10.3%) than men (6.2%). Major depression is defined as a period of at least two weeks of low mood, loss of interest or pleasure in daily activities, and a majority of specific symptoms, such as problems with sleeping, eating, energy, concentration, or self-esteem.

Approximately 30% of people with depression have treatment-resistant depression, meaning that their symptoms do not improve despite trying multiple types of medications and therapies. Additionally, two-thirds of people with depression never fully improve despite treatment. One reason for this is a lack of data showing which treatments work for which patients, and current practice relies on trial and error to find effective treatments. In some cases, depression can worsen while trying to find the right treatment.

“The goal of our study is to think about how we can do it right the first time,” said lead author Leanne Williams, PhD, professor of psychiatry and behavioral sciences and director of the Center for Precision Mental Health and Wellness at the Stanford University School of Medicine. “It's very frustrating to be in the field of depression and not have a better alternative than this one-size-fits-all approach.”

To better understand the different biotypes of the disease, Williams and his colleagues enrolled 801 people who had been diagnosed with depression or anxiety in their study, measuring their brain activity with functional magnetic resonance imaging (fMRI). Brain scans were conducted both while the subjects were resting, while they were engaged in an activity, and while they performed a variety of tasks that tested cognitive and emotional functioning. The imaging focused on areas of the brain already known to be involved in depression.

Using these data, the researchers applied a machine learning technique called cluster analysis to group the patients' brain images and identify six distinctive patterns of brain activity in the regions they studied.

The team also randomly assigned 250 patients to receive one of three commonly prescribed treatments for depression: medication or behavioral therapy. In one subtype, characterized by overactivity in cognitive areas of the brain, participants treated with venlafaxine (Effexor) fared best compared with patients with the other biotypes.

The second subtype, who showed higher levels of activity in three brain regions associated with depression and problem-solving while patients were at rest, showed better treatment outcomes when they participated in talk therapy. People with the third subtype, who showed lower resting levels of activity in the brain circuits that control attention, were less likely to see improvement in their symptoms with talk therapy than people with the other biotypes.

The biotypes and how they respond to different treatments correlate with what is already known about these areas of the brain, says Jun Ma, M.D., a professor of medicine at the University of Illinois at Chicago and one of the study's authors. But the research also breaks new ground.

“To our knowledge, this is the first time we've been able to demonstrate that depression can be explained by multiple disturbances in brain function,” Williams noted. “In essence, this is a demonstration of a personalized medicine approach to mental health based on objective measurements of brain function.”

The researchers also found that the different biotypes identified showed differences in the symptoms and task performance of study participants. For example, people with overactivity in cognitive areas of the brain had higher levels of anhedonia, the inability to experience pleasure, than other biotypes. People with anhedonia also performed worse on executive function tasks. Patients who responded best to talk therapy performed better on cognitive tasks but worse on executive tasks.

An interesting finding of the study is that one out of six biotypes showed no noticeable differences in brain activity compared to those in people without depression. Williams adds that this is likely due to gaps in our understanding and lack of data on the full range of brain activity that underlies depression, and that other areas of the brain beyond those studied are also involved in the illness.