Six Subtypes of Depression, Anxiety Identified Via MRI

Brain imaging combined with artificial intelligence has identified six distinct “biotypes” of depression and anxiety that may lead to more personalized and effective treatment.

This research has “immediate clinical implications,” study investigator Leanne Williams, PhD, director of the Stanford Medicine Center for Precision Mental Health and Wellness, told Medscape Medical News.

“At Stanford, we have started translating the imaging technology into use in a new precision mental health clinic. The technology is being actively developed for wider use in clinical settings, and we hope to make it accessible to more clinicians and patients,” Williams said.

The study was published online on June 17, 2024, in Nature Medicine.

No More Trial and Error?

Depression is a highly heterogeneous disease, with individual patients having different symptoms and treatment responses. About 30% of patients with major depression are resistant to treatment, and about half of patients with generalized anxiety disorder do not respond to first-line treatment.

“The dominant ‘one-size-fits-all’ diagnostic approach in psychiatry leads to cycling through treatment options by trial and error, which is lengthy, expensive, and frustrating, with 30-40% of patients not achieving remission after trying one treatment,” the authors noted.

“The goal of our work is figuring out how we can get it right the first time,” Williams said in a news release, and that requires a better understanding of the neurobiology of depression.

To that end, 801 adults diagnosed with depression and anxiety underwent functional MRI to measure brain activity at rest and when engaged in tasks designed to test cognitive and emotional functioning.

Researchers probed 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.

Using a machine learning technique known as cluster analysis to group the patients’ brain images, they identified six clinically distinct biotypes of depression and anxiety defined by specific profiles of dysfunction within both task-free and task-evoked brain circuits.

“Importantly for clinical translation, these biotypes predict response to different pharmacological and behavioral interventions,” investigators wrote.

For example, patients with a biotype characterized by overactivity in cognitive regions of the brain experienced the best response to the antidepressant venlafaxine compared with patients with other biotypes.

Patients with a different biotype, characterized by higher at-rest levels of activity in three regions associated with depression and problem-solving, responded better to behavioral therapy.

In addition, those with a third biotype, who had lower levels of activity at rest in the brain circuit that controls attention, were less apt to see improvement of their symptoms with behavioral therapy than those with other biotypes. The various biotypes also correlated with differences in symptoms and task performance.

For example, individuals with overactive cognitive regions of the brain had higher levels of anhedonia than those with other biotypes, and they also performed worse on tasks measuring executive function. Those with the biotype that responded best to behavioral therapy also made errors on executive function tasks but performed well on cognitive tasks.

A Work in Progress

The findings provide a deeper understanding of the neurobiological underpinnings of depression and anxiety and could lead to improved diagnostic accuracy and more tailored treatment approaches, the researchers noted.

Naming the biotypes is a work in progress, Williams said.

“We have thought a lot about the naming. In the Nature Medicine paper, we use a technical convention to name the biotypes based on which brain circuit problems define each of them,” she explained.

“For example, the first biotype is called D_C+S_C+A_C+ because it is defined by connectivity increases [C+] on three resting circuits — default mode [D], salience [S], and frontoparietal attention [A]. We are working with collaborators to generate biotype names that could be convergent across findings and labs. In the near future, we anticipate generating more descriptive medical names that clinicians could refer to alongside the technical names,” Williams said.

Commenting on the research for Medscape Medical News, James Murrough, MD, PhD, director of the Depression and Anxiety Center for Research and Treatment at the Icahn School of Medicine at Mount Sinai, New York City, called it “super exciting.”

“The work from this research group is an excellent example of where precision psychiatry research is right now, particularly with regard to the use of brain imaging to personalize treatment, and this paper gives us a glimpse of where we could be in the not-too-distant future,” Murrough said.

However, he cautioned that at this point, “we’re far from realizing the dream of precision psychiatry. We just don’t have robust evidence that brain imaging markers can really guide clinical decision making currently.”

Funding for the study was provided by the National Institutes of Health and by Brain Resource Ltd. Williams declared US patent applications numbered 10/034,645 and 15/820,338: “Systems and methods for detecting complex networks in MRI data.” Murrough had no relevant disclosures.