Using brain imaging and machine learning, researchers at Stanford Medicine have identified six biological subtypes of depression, paving the way for more precise psychiatric treatments. The results suggest that certain antidepressants and behavioral treatments are more effective for certain subtypes, offering promise for more targeted and effective treatment strategies. Credit: SciTechDaily.com
A Stanford Medical study reveals six different types of depression that can be identified by brain imaging Machine learning. These subtypes show distinct patterns of brain activity, which can help predict which patients will benefit from certain antidepressants or behavioral treatments. This approach aims to personalize and improve the effectiveness of depression treatment.
In the not-too-distant future, a quick brain scan during a screening evaluation for depression may identify the best treatment.
A new study led by Stanford Medicine researchers shows that brain imaging combined with a type of AI called machine learning can predict types of depression and anxiety. The study will be published today (June 17) in the journal Natural medicineIt separates depression into six biological subtypes, or “biotypes,” and identifies treatments that may or may not work for three of these subtypes.
The development of precision psychiatry
Better methods to match patients to treatments are essential, said the study’s senior author, Leanne Williams, PhD, Vincent V. Woo Professor, Professor of Psychiatry and Behavioral Sciences and Director of Stanford Medicine’s Center for Authentic Mental Health and Wellness. In the year Williams, who lost her partner to depression in 2015, has focused her career on pioneering the field of precision psychiatry.
30% of people with depression are known Treatment-resistant depressionMany medications or treatments failed to improve their symptoms. And for up to two-thirds of people with depression, treatment fails to fully return their symptoms to a healthy level.
This is because there is no good way to know which antidepressant or type of treatment will help a patient. Medications are prescribed by trial-and-error, so it can take months or years to land on a medication that works—if it ever does. And spending too much time trying treatment after treatment can only make depression symptoms worse without getting any relief.
“The goal of our work is to figure out how to get it right the first time,” Williams said. “It’s frustrating to be in the field of depression and not have a better alternative to this one-size-fits-all approach.”
Biotypes predict treatment response
To better understand the biology behind depression and anxiety, Williams and her colleagues used 801 study participants who had previously been diagnosed with depression or anxiety to undergo what’s known as a functional MRI, or fMRI, to measure brain activity. They scanned the brains of volunteers at rest and while engaged in various tasks to examine their cognitive and emotional functioning. Scientists narrowed down the areas of the brain previously known to play a role in depression and the connections between them.
Using a machine learning technique known as cluster analysis, they classified the patients’ brain images into groups and identified six distinct patterns of activity in the brain regions they studied.
The scientists also randomly assigned 250 study participants to receive one of three commonly used antidepressants or behavioral talk therapy. Patients with one subtype of hyperactivity in cognitive regions of the brain responded better to the antidepressant venlafaxine (commonly known as Effexor) compared to those with other biotypes. Those with another subtype, whose brains are at rest with greater activity among three regions associated with depression and problem solving, were better able to alleviate their symptoms with behavioral talk therapy. And those with the third subtype, which has low activity at rest in a brain circuit that controls attention, were less likely to experience symptoms with talk therapy than those with other biotypes.
Exploring the effectiveness of treatment based on brain activity
Biotypes and their response to behavioral therapy make sense based on what we know about these brain regions, said Jun Ma, MD, PHD, Beth and George Vitoux Professor of Medicine at the University of Illinois at Chicago and one of the authors. The study. The type of therapy used during their trial teaches patients skills to better solve everyday problems, so increased activity in these brain regions allows patients with that biotype to pick up new skills more quickly. As for those with lower activity levels related to attention and participation in the region, drug therapy may help those patients with lower activity levels better than talk therapy.
“To our knowledge, this is the first time we’ve been able to show that depression can be defined by different brain dysfunctions,” Williams said. “Essentially, it’s a demonstration of a personalized medicine approach to mental health based on objective measurements of brain function.”
Improving the prognosis of antidepressant treatment
in another A recently published study, Williams and her team have shown that using fMRI brain imaging can improve their ability to identify individuals who may respond to antidepressant treatment. In that study, the scientists focused on a subtype called the cognitive biotype of depression, which accounts for more than a quarter of people with depression and is less likely to respond to standard antidepressants. Using fMRI to identify those with a cognitive biotype, the researchers correctly predicted survival in 63% of patients compared to 36%. Accuracy Without using brain imaging. That improved accuracy means providers are more likely to get the treatment right the first time. Scientists are researching new treatments for this biotype in hopes of finding more options for those who don’t respond to standard antidepressants.
More searches for depression
Different biotypes are associated with differences in symptoms and task performance among trial participants. Areas of the brain with over-empathic abilities, for example, had higher levels of anhedonia than people with other biotypes. They also performed worse on executive functions. Those with the subtype who responded best to talk therapy performed poorly on executive tasks but did well on cognitive tasks.
None of the six biotypes included in the study showed any significant difference in brain activity in the imaged regions compared to non-depressed individuals. Williams believes that they probably haven’t fully explored the brain biology behind this disease – their research is focused on the regions involved in depression and anxiety, but there may be other abnormalities in this biotype that their images don’t capture. .
Williams and her team are expanding the imaging study to include more participants. She also wants to try more treatments with six biotypes, including drugs not traditionally used for depression.
Colleague Laura Hack, MD, PhD, assistant professor of psychiatry and behavioral sciences, has begun using the imaging technique in her clinical practice at Stanford Medicine in an experimental protocol. The team wants to develop easy-to-follow criteria for the method so that other psychiatrists can apply it.
“In order to really move the field toward proper psychiatry, we need to identify the most effective treatments for patients and get them to that treatment as quickly as possible,” Ma said. “Having information on brain function, especially the validated signatures we evaluated in this study, can help inform more accurate treatment and drug prescriptions for individuals.”
Reference: “Individual Brain Circuit Effects Distinguish Clinically Distinct Biotypes in Depression and Anxiety” by Leonardo Tozzi, Zhu Zhang, Adam Pines, Alyssa M. Olmsted, Emily S. Zihai, Esther T. Annen, Megan Chesnutt, Bailey Holt-Gosselyn, Sarah Chang, Patrick C. States, Carolina A. Ramirez, Laura M. Hack, Mayuresh S. Korgaonkar, Max Wintermark, Ian H. Gottlieb, Jun Ma, and Leanne M. Williams, 17 June 2024; Natural medicine.
DOI: 10.1038/s41591-024-03057-9
Researchers from Columbia University; Yale University School of Medicine; University of California, Los Angeles; UC San Francisco; of University of Sydney; University of Texas MD Anderson; And the University of Illinois at Chicago also contributed to the study.
Data sets in the study were funded by National Institutes of Health (grant numbers R01MH101496, UH2HL132368, U01MH109985 and U01MH136062) and by Brain Resource Ltd.