Blood Biomarkers Could Aid Prediction of Psychosis Risk

Peripheral blood biomarkers of inflammation and other processes appear to help indicate which individuals with a high risk for psychosis will actually “convert” to psychosis.

In a small sample of subjects from the North American Prodrome Longitudinal Study (NAPLS), laboratory tests analyzing multiple molecules and compounds in peripheral blood distinguished individuals who would go on to convert to psychosis from unaffected controls and from subjects deemed initially at high risk but who did not later develop psychosis.

The findings were presented at a symposium during a day-long satellite conference on prodromal psychosis that preceded the International Congress on Schizophrenia Research last month in Orlando, Fla.

Diana Perkins, M.D., M.P.H., a professor of psychiatry at the University of North Carolina, Chapel Hill, said the biomarker test included compounds reflecting altered immune status and oxidative stress, suggesting that these factors could drive the development of psychosis.

The study results advance the effort to discover biomarkers that point to risk of developing schizophrenia. According to Perkins, that’s important because the clinical risk criteria focusing on behavioral symptoms and patient self-reports of subjective experiences identify individuals at elevated risk for psychosis, but less than one-third will ultimately develop psychosis. Accurate diagnostic tests in those exhibiting symptoms that point to psychosis risk could guide clinicians and patients in making choices about preventative interventions.

Also at the conference, Daniel Mathalon, Ph.D., M.D., and Rachel Loewy, Ph.D., both of the University of California, San Francisco, presented research indicating that environmental stress and early-life trauma can induce inflammatory responses that may be implicated in loss of gray matter related to diminished synaptic plasticity. This means that growth in strength and efficacy of synaptic connections derived from repeated experience and learning is diminished in high-risk patients who convert to psychosis.

Symposium chair Tyrone Cannon, M.D., of Yale University, said that taken together the presentations offer a speculative model for understanding the progression to psychosis—from environment to biology to neurobiology—and opportunities for psychosocial intervention prior to the onset of psychosis.

But it is the advance toward finding biomarkers of risk that holds out the most promise for enhancing prediction of psychosis in those deemed at high risk on the basis of clinical criteria. Perkins said the findings on peripheral blood biomarkers build on research by Sabine Bahn, M.D., director of the Cambridge Centre for Neuropsychiatric Research, who has shown that plasma biomarkers of inflammation are significantly correlated with disease among first-episode psychosis patients.

Perkins said other biomarkers detectable in the laboratory—oxidative stress, HPA-axis dysfunction, and elevated cortisol levels—are also promising, but inflammation appears to be the leading candidate. “There is an emerging story that inflammation is really important,” she said. “I think these blood tests show promise in helping us to diagnose patients who are already at high risk according to clinical criteria. I don’t think they are going to be useful as a screening tool in the general population, but if you have an individual you are worried about, these tests may be able to help clinicians hone in on who is really at risk.”

In their study, Perkins and colleagues used a commercially available panel of tests to analyze blood samples from 32 clinically high-risk patients who progressed to psychosis, 40 high-risk patients who were followed for two years but did not progress to psychosis by February 2012, and 35 unaffected controls. (Plasma draws had been built into initial assessments of patients entering the NAPLS study; for information about NAPLS, see Psychiatric News, May 20, 2011).

The tests—weighted toward finding markers of inflammation, since inflammation is critical in many disease processes—reliably distinguished, with a high degree of sensitivity and specificity, those who converted to psychosis from the unaffected controls and from those at high risk who didn’t convert.

As an example Perkins presented case studies of presenting symptoms from two patients seen in her clinical practice. Both experienced symptoms that were rated high according to clinical risk criteria used for the NAPLS project, but one progressed to psychosis and the other did not.

While the two patients were difficult to distinguish on the basis of clinical criteria alone, the blood tests reliably predicted the patient who ultimately converted. “Having this test potentially could have been useful in deciding that this is the patient I should be more concerned about,” Perkins said.

And she said the laboratory findings correlating inflammation with risk for conversion to psychosis were externally validated; among converters to psychosis, positive blood tests were significantly correlated with high scores on the Scale of Prodromal Symptoms and on negative symptoms.

She said that blood biomarkers can be used in combination with clinical criteria to create psychosis-risk algorithms for patients, similar to algorithms developed for cancer or diabetes risk. For instance, body mass index is a reasonable clinical indicator of risk for diabetes; but predictive capacity is greatly enhanced when clinical criteria are combined with hemoglobin A1C or blood-pressure measures.

“I believe there is tremendous clinical utility in peripheral blood tests for predicting risk for psychosis,” she said. “We can target persons who might best benefit from intervention. We can predict morbidity and mortality. And there is the potential to learn something about disease etiology and to predict treatment response.” ■