Blood-Based Biomarker Panel Aids Bipolar Disorder Diagnosis

Researchers at the University of Cambridge have developed a way of improving diagnosis of bipolar disorder (BD), which uses a simple dried blood spot (DBS) test to identify biomarkers associated with the condition. While the team’s newly reported study suggested that the blood test on its own could diagnose up to 30% of patients with bipolar disorder, their research, involving data from the UK Delta study, showed that combining the blood-based biomarker test with a digital mental health assessment is even more effective in diagnosing the condition among patients who may have previously been misdiagnosed with major depressive disorder (MDD). The study results also pointed to the potential involvement of ceramides in the pathomechanistic basis of mood disorders.

Although the blood test is still a proof of concept, the researchers suggested it could be an effective complement to existing psychiatric diagnosis and could help scientists better understand the biological origins of mental health conditions. Incorporating biomarker testing could also help physicians differentiate between major depressive disorder and bipolar disorder, which have overlapping symptoms but require different pharmacological treatments.

“When someone with bipolar disorder is experiencing a period of low mood, to a physician, it can look very similar to someone with major depressive disorder,” said research lead Sabine Bahn, MD, PhD. “However, the two conditions need to be treated differently: if someone with bipolar disorder is prescribed antidepressants without the addition of a mood stabilizer, it can trigger a manic episode.” Bahn is a professor of neurotechnology at the department of chemical engineering and biotechnology and is a fellow of Lucy Cavendish College, Cambridge.

In their published paper in Jama Psychiatry (“Metabolomic Biomarker Signatures for Bipolar and Unipolar Depression”), Bahn and colleagues concluded, “Our results emphasize the potential of blood biomarkers to successfully complement psychometric assessments for mood disorders and improve disease biological understanding …This diagnostic study identified a reproducible metabolomic biomarker signature in patients’ DBSs that effectively differentiated misdiagnosed BD form MDD during episodes of low mood, and enhanced predictive value of diagnostic models based on self-reported patient information.”

Bipolar disorder affects approximately one percent of the population—as many as 80 million people worldwide—but, as the authors noted, for nearly 40% of patients the condition is misdiagnosed as major depressive disorder. “BD is frequently misdiagnosed as MDD because of overlapping symptoms and the lack of objective diagnostic tools,” they wrote. First author Jakub Tomasik, PhD, from Cambridge’s department of chemical engineering and biotechnology, added, “People with bipolar disorder will experience periods of low mood and periods of very high mood or mania. But patients will often only see a doctor when they’re experiencing low mood, which is why bipolar disorder frequently gets misdiagnosed as major depressive disorder.”

The most effective way to get an accurate diagnosis of bipolar disorder is a full psychiatric assessment. However, patients often face long waits to get these assessments, and they take time to carry out. “Psychiatric assessments are highly effective, but the ability to diagnose bipolar disorder with a simple blood test could ensure that patients get the right treatment the first time and alleviate some of the pressures on medical professionals,” said Tomasik.

Biomarker profiling represents a promising approach to enable earlier, more objective differential diagnosis of mood disorders, but to date, studies have faced a number of limitations, “such as identifying disease biomarkers in patients with established BD confounded by the effect of mood-stabilizing medication, using samples from patients with varying symptom polarities and other uncontrollable factors, and lacking independent validation cohorts,” the team noted. The scientists aimed to address these limitations by identifying and validating a DBS metabolomic signature that could distinguish BD from MDD, particularly among misdiagnosed BD patients who had presented with depressive symptoms.

For their study the researchers used samples and data from the Delta study, conducted in the U.K. between 2018 and 2020, to identify bipolar disorder in patients who had received a diagnosis of major depressive disorder within the previous five years and had current depressive symptoms. Participants were recruited online through voluntary response sampling.

More than 3,000 participants were recruited, and they each completed an online mental health assessment of more than 600 questions. The assessment covered a range of topics that may be relevant to mental health disorders, including past or current depressive episodes, generalized anxiety, symptoms of mania, family history, or substance abuse.

Of the participants who completed the online assessment, around 1,000 were selected to send in a dried blood sample from a simple finger prick, which the researchers analyzed for more than 600 different metabolites using mass spectrometry. After completing the Composite International Diagnostic Interview (CIDI), a fully structured and validated diagnostic tool to establish mood disorder diagnoses, 241 participants were included in the study. “The discovery cohort consisted of participants who had current depressive symptoms who had been diagnosed with MDD within the prior five years, whose diagnosis was either confirmed as MDD or altered to BD I by the CIDI,” the investigators stated.

Analysis of the data showed a significant biomarker signal for bipolar disorder, even after accounting for confounding factors such as medication. The identified biomarkers were correlated primarily with lifetime manic symptoms and were validated in a separate group of patients who received a new clinical diagnosis of major depressive disorder or bipolar disorder during the study’s one-year follow-up period.

The results pointed to the potential involvement of sphingolipids, and particularly ceramides, in the distinct pathomechanisms of mood disorders. “This may signify intrinsic immunometabolic or signaling differences between BD and MDD or the body’s adaptive responses to these conditions,” the investigators wrote. “The final model consisted of 17 biomarkers, among which ceramide d18:0/24:1 was the most important … Likewise the most relevant analyte class was ceramides.”

The researchers found that the combination of patient-reported information and the biomarker test significantly improved diagnostic outcomes for people with bipolar disorder, especially in those where the diagnosis was not obvious. “The online assessment was more effective overall, but the biomarker test performs well and is much faster,” said Bahn. “A combination of both approaches would be ideal, as they’re complementary.”

The authors also noted, “Decision curve analysis showed that at relevant diagnostic thresholds, the inclusion of biomarkers may result in the additional identification of up to 30% of patients with BD, accounting for false positives … The added value of biomarkers was particularly evident in scenarios where data on psychiatric symptoms were unavailable and at intermediate diagnostic thresholds, suggesting that biomarker tests may especially benefit patients who do not report their symptoms and whose diagnoses are uncertain.”

Commented Tomasik, “We found that some patients preferred the biomarker test because it was an objective result that they could see. Mental illness has a biological basis, and it’s important for patients to know it’s not in their mind. It’s an illness that affects the body like any other.”

Bahn added, “In addition to the diagnostic capabilities of biomarkers, they could also be used to identify potential drug targets for mood disorders, which could lead to better treatments. It’s an exciting time to be in this area of research.”

A patent has been filed on the research by Cambridge Enterprise, the University’s commercialization arm.