Measuring the activity of a panel of genes may provide enough information to diagnose bipolar disorder from a blood sample, according to a study published August 4th in the journal Translational Psychiatry. Currently, bipolar disorder is diagnosed by experienced clincians based on a subjective evaluation of a patient's symptoms; there is no blood test or brain scan that can diagnose the disorder.

In the study, which included 37 patients with rapid-cycling bipolar disorder and 40 healthy controls, researchers found that a composite measure of activity within a set of 19 genes offered diagnostic sensitivity and specificity corresponding to a “moderately accurate” test. Their analysis also identified specific genes whose activity is lower in people with bipolar disorder than in healthy individuals, as well as an additional gene whose activity spikes when patients with bipolar disorder are in a depressed state. Those findings call attention to biological pathways likely to be involved in the disorder.

To carry out the study, a team of scientists including first author Klaus Munkholm, M.D., at the University of Copenhagen, identified 19 genes that previous studies had indicated might behave differently in people with bipolar disorder, and for which relevant variations were likely to be detectable in blood cells. They took blood samples from the patients over a period of six to 12 months, keeping track of whether patients were in manic, depressed, or euthymic (neither manic nor depressed) states at the time each sample was collected.

The team, which included Lars Vedel Kessing, M.D., a 2012 NARSAD Distinguished Investigator and 2010 Foundation Colvin Prizewinner, then measured the activity of the 19 genes they had selected. Each sample received a gene expression score that considered the activity of all 19 genes together. These scores were sufficiently different between healthy participants and those with bipolar disorder to distinguish between the two, most of the time. For a portion of their blood samples, the scientists used the gene expression score to correctly identify 78 percent of samples from patients with bipolar disorder. Sixty percent of the samples from healthy individuals were correctly identified.

Further studies will be necessary to evaluate the gene panel as a potential biomarker of bipolar disorder. Including additional information in the score, such as levels of specific proteins, might improve its diagnostic sensitivity and specificity, the scientists say.

The study also generated some insight into the biology of bipolar disorder.  Two of the genes that were tracked in the study—POLG and OGG1—were significantly less active in people with bipolar disorder than they were in healthy controls. Another, NDUFV2, showed increased activity during depressed states among people with bipolar disorder, when gene activity was compared during individual patients' different mood states. Together, these findings point toward problems with mitochondrial function and DNA damage repair, the team says.

Read the paper.