Researchers are taking two paths toward the same goal: a blood test for depression.
Eva Redei’s quest to find blood-borne signs of depression and anxiety began in 1989—when the prevailing scientific view “was that psychiatric disease wasn’t like other diseases, and that using animals to find biomarkers for depression was shocking,” says Redei, professor of psychiatry, behavioral sciences and physiology at Northwestern University’s Feinberg School of Medicine. Yet she persevered, breeding two strains of rats that were almost identical, except that one group had severe symptoms mirroring human depression. She exposed other rats to chronic psychological stress by confining them in a tube.
Redei found distinct genetic markers in the blood that set apart depressed rats from healthy controls. Then 26 markers were tested in human adolescents, and 11 helped differentiate between depressed and non-depressed teens; 18 separated those with major depression from those with depression and anxiety disorder. In a separate study, these biomarkers are yielding promising results in adult depression, says Redei.
At the same time, scientists working with Ridge Diagnostics in San Diego have analyzed mathematical models of existing biomarker candidates and found nine that appear to have the strongest predictive power for depression. “It has been so difficult to develop robust depression biomarkers because each risk factor has a modest effect,” says George Papakostas, director of treatment-resistant depression studies at Massachusetts General Hospital. Ridge Diagnostics developed a blood test based on the nine biomarkers—associated with inflammation, neuron development and stress responses in the brain.
In a pilot study, the test accurately pinpointed depression in 90% of depressed patients. The next step is to gauge the test’s ability to detect hard-to-diagnose depression. “The test could help a primary care physician diagnose difficult cases,” says Papakostas, lead author of an account published in the journal Molecular Psychiatry. It might also be used to see whether a drug is working or to predict the likelihood of relapse—but more studies are needed.