Brain Scans Show Why a Common Depression Therapy Doesn't Help Some Patients
A new tool allows doctors to predict which patients can benefit most.
Approximately 7.1 percent of all American adults have experienced at least one episode of major depression, a statistic that which means millions have struggled with their mental health. Some people respond well to certain therapies, while others do not. In a study released Wednesday in Science Advances, researchers, in an effort to help patients find the right therapy sooner, reveal why that disparity exists.
Co-author Dr. Filippo Queriazza is a postdoctoral fellow at the University of Glasgow and a practicing National Health Service consultant psychiatrist. He tells Inverse that he’s always been frustrated with the lack of objective tests that can help inform how he makes decisions in the clinic.
In the medical field more broadly, there is a growing focus on “precision medicine”, which tailors treatment options to each patient to improve their care. To implement it, medical practitioners rely on established predictive biomarkers, which allow them to choose the right treatment for the patient without having to do a lot of trial and error. In psychiatry, predictive biomarkers aren’t well established, so it’s difficult to tell in advance which treatment will work for each patient.
“Although there are effective treatments for mental disorders, our current understanding of how they actually work is poor,” Querizza explains. “In this study, we tried to shed some light on the cognitive mechanisms underlying responses to cognitive behavioral therapy, which is a recognized effective treatment for depression.”
Cognitive behavioral therapy (CBT) is a type of talk therapy that’s teaches patients to become aware of patterns of inaccurate or negative thinking. The goal is for patients to realize that psychological problems are based, in part, on these patterns and to learn better ways of coping. CBT is established as an effective treatment for depression and anxiety disorder and can sometimes be more effective than psychiatric medications. However, it only helps an average of 45 percent of depressed patients.
Queriazza and his colleagues explain that not understanding the reasons behind CBT’s inconsistency among patients, and the “mechanisms” behind it, has hindered our understanding of the diseases it can treat. In turn, the biomarkers scientists need to seek out unidentified targets for drug development also remain elusive.
What scientists do know is that when people are depressed, their reinforcement learning — a learning style driven by rewards and punishments — doesn’t quite work. Typically, the brain drives reward learning by activating dopamine neurons in the midbrain, which signal what’s a reward and what’s a punishment.
Because CBT relies on a patient’s ability to update negatively biased beliefs regarding themselves, the world, and the future as they learn the positive aspects of doing so, the scientists hypothesized that the parts of the brain linked to reinforcement learning would reveal the predictive biomarkers they were searching for.
So, the team looked at the brains of patients before and after CBT treatment using functional magnetic resonance imaging (fMRI). Out of 37 patients, 26 completed a six-module CBT course.
As they predicted, the patients who responded best to CBT showed greater neural activity before the treatment in the parts of the brain linked to reinforcement learning: the right striatum and the right amygdala. This brain activity also allowed the scientists to predict individual responses to CBT with predictive power of around 80 percent.
Queriazza says that the pattern of brain activity in the CBT-responsive group can be interpreted as “indicative of a comparatively more pronounced ruminative thinking style” that’s “more amendable to benefit from CBT.” It showed active information acquisition and processing ability, in turn suggesting that people who respond to CBT are better at appraising the world around them and drawing new conclusions.
“We speculate that depressed patients who are more adept at thoughtfully sieving through the barrage of noisy feedback information surrounding them may exhibit a greater disposition to critical thinking,” the scientists write. “This may translate into a greater ability to challenge maladaptive thinking patterns.”
Queriazza emphasizes that this study is just “the first step in the development of an imaging biomarker that is predictive response to CBT in depression”; the results must still be replicated and validated in a larger sample of patients. Only then, he says, will we “have enough evidence to ascertain whether our biomarker can be adopted in a day-to-day clinical practice.”
While cognitive behavioral therapy (CBT) is an effective treatment for major depressive disorder, only up to 45% of depressed patients will respond to it. At present, there is no clinically viable neuroimaging predictor of CBT response. Notably, the lack of a mechanistic understanding of treatment response has hindered identification of predictive biomarkers. To obtain mechanistically meaningful fMRI predictors of CBT response, we capitalize on pretreatment neural activity encoding a weighted reward prediction error (RPE), which is implicated in the acquisition and processing of feedback information during probabilistic learning. Using a conventional mass-univariate fMRI analysis, we demonstrate that, at the group level, responders exhibit greater pretreatment neural activity encoding a weighted RPE in the right striatum and right amygdala. Crucially, using multivariate methods, we show that this activity offers significant out-of-sample classification of treatment response. Our findings support the feasibility and validity of neurocomputational approaches to treatment prediction in psychiatry.