One of the great challenges for psychiatric research is arriving at a reliable way of diagnosing bipolar disorder, and doing so as early in the course of the illness as possible so as to boost the chances of effective treatment and disease management.

An important problem relates to the fact that the first symptoms experienced by a patient may be those of depression. The thing that distinguishes a diagnosis of depressive disorder from one of bipolar disorder is the occurrence in the latter not only of a depressive episode, but also, at least once in the patient’s history, of symptoms of high mood called mania, or a less intense but just as serious form called hypomania. These symptoms are at the opposite end of the affective spectrum from symptoms of low mood that characterize depression. In mania and hypomania, people can be hyperenergetic and ambitious in unrealistic ways; they can function with very little sleep, may speak rapidly and form grandiose plans; and they may display reckless behavior involving activities such as sex, gambling, or substance use. Distinguishing bipolar disorder from unipolar depression is critical, since the treatments are different.

For those coming to the doctor with depression, is there a way to figure out which subset of patients are at high risk of also experiencing mania or hypomania, and thus receiving a bipolar diagnosis? One hope of researchers and doctors has been to discover one or more biomarkers that would signal elevated risk for mania, something that would help to identify bipolar individuals early, or indicate high risk before an initial manic or hypomanic episode.

In 2005, Mary L. Phillips, M.D., M.D. (Cantab) performed research under a BBRF Independent Investigator grant in which she began a search for neural biomarkers of bipolar disorder. Now a member of BBRF’s Scientific Council, a 2024 BBRF Distinguished Investigator, and in 2017 the winner of BBRF’s Colvin Prize for outstanding mood disorders research, Dr. Phillips and colleagues at the University of Pittsburgh recently published a paper in JAMA Psychiatry reporting on several patterns in brain activity in young adults that appear to distinguish elevated risk of mania/hypomania from elevated risk of depression. Michele A. Bertocci, Ph.D., a 2019 BBRF Young Investigator, and Henry W. Chase, Ph.D., a 2017 BBRF Young Investigator, were among members of the team. Maya C. Schumer, Ph.D., was first author of the paper.

The researchers recruited 299 young adults from the community between ages 18 and 30 (average age was early 20s), some of whom had psychiatric diagnoses and some who did not. The participants were divided into 3 independent subgroups. One, called the “discovery sample,” consisted of 114 individuals, 36 of whom had lifetime diagnoses of major depression, an anxiety disorder, or ADHD. The two other subgroups were named “test sample 1” and “test sample 2”; the first consisted of 103 people, 47 of whom had psychiatric diagnoses; the second had 82 individuals, 19 with psychiatric diagnoses. The intention of the researchers was to identify potential biomarkers of elevated mania/hypomania and/or depression risk in the “discovery” sample, and to test their reproducibility in the two “test” samples.

Several overlapping methods were employed in the search for biomarkers. Participants in all three samples completed a facial emotion-processing task involving response to various kinds of facial expressions (happy and angry faces) while undergoing a functional MRI scan. The intention was to measure patterns of neural response to approach-related emotions, to which heightened reactivity is associated with mania/hypomania risk. This includes goal-seeking behaviors; such behaviors are also associated with positive emotions, including reward cues, as well as anger and irritability, that can together indicate interest, curiosity, or the desire to engage. Individuals in all three samples were also assessed with a tool called the Mood Spectrum Self-Report (MOODS-SR), a standard questionnaire for the assessment of lifetime vulnerability to mood spectrum symptoms, including risk for subthreshold and spectrum-level mania/hypomania, as well as depression. The “manic mood domain” in the questionnaire assessed euphoria, inflated self-esteem, instability or irritability, creativity, sociability, extraversion, wastefulness and recklessness. A “depressive mood domain” assessed depressed mood and substance use-related depression.

 

In its analysis, the team was able to determine that certain brain activation and neural network connectivity patterns were specific to elevated risk for mania/hypomania, as well as for depression. These patterns were in some cases distinct, and thus were considered potential biomarkers. After making this determination, the team tested to see if the identified patterns also occurred in the 2 “test” groups. Results indicated that the risk associations specific to mania/hypomania and depression in the discovery sample could indeed be replicated in the two independent test samples.

Specifically, the team found that in all three samples of young adults, greater functional connectivity within the amygdala was associated with greater risk of both mania/hypomania and depression. This would not help distinguish mania from depression risk. In contrast, greater functional connectivity between the bilateral ventrolateral prefrontal cortex (vlPFC) and the right dorsolateral prefrontal cortex (dlPFC) was associated specifically with elevated mania/hypomania risk, while greater deactivation of the right caudate was specifically associated with elevated depression risk.

The findings relating greater inter-amygdala connectivity to risk of both mania and depression may reflect “greater attribution of salience to approach-related emotional stimuli, and is a potential biomarker of broader mood disorder risk,” the team said. Further consideration of the amygdala findings led the team to propose that depression risk might be associated with an individual’s assigning greater salience to all emotions, while mania/hypomania risk may be more specifically associated, they suggested, with assigning greater salience to “approach-related emotions,” i.e., emotions felt in situations where an individual is expressing interest or curiosity or desire to engage with a stimulus, be it a person, an object or a situation. The researchers also noted that neural signatures indicating greater engagement between the ventral attention network (VAN) and the central executive network (CEN) indicating greater attention or sensitivity to approach-related emotions appears to uniquely characterize elevated risk for mania/hypomania.

The “robust associations” they found in 3 independent samples that relate “specific neural markers” and risk for mania/hypomania and depression, the team said, may not only help distinguish the two conditions, but also “provide neural targets to guide and monitor interventions in individuals at risk for bipolar disorder and other affective disorders.”

Future studies will need to replicate these findings, and might also try to identify neural markers associated with future manic/hypomanic episodes, the team suggested.