Skip to Main Content

Groundbreaking multi-site Yale study of psychosis engages biomarkers to reconfigure diagnoses

December 08, 2015

In a groundbreaking multi-site study, a comprehensive set of empirical biomarkers has been established to aid in diagnosis and treatment of psychosis -- psychiatric syndromes characterized by delusions, hallucinations and disordered language.

To date, the “gold standard” for the diagnosis of psychosis has been clinical observation, classifying patients into schizophrenia, schizoaffective, and bipolar disorders. But in this study, the Bipolar Schizophrenia Network for Intermediate Phenotypes (B-SNIP) identified three neurobiologically distinct “Biotypes” that do not match up with the conventional clinical diagnosis.

“Instead of constructing psychiatric diagnoses traditionally in a top-down fashion using symptoms and clinical course, we have redefined them from the bottom-up using multiple, reliable, biologically-based measures,” said Dr. Godfrey Pearlson, Director of the Olin Neuropsychiatry Research Center at the Institute of Living in Hartford, and Professor of Psychiatry and Neurobiology at Yale University School of Medicine. "This strategy offers promising preliminary data that suggests a means to derive a biologic redefinition of psychotic syndromes that may help improve both the search for risk genes for these disorders, and to understand their biological origin in order to develop rationally-based new treatments."

The B-SNIP consortium, which includes Yale, UT Southwestern in Dallas, Harvard University, the University of Chicago, and the University of Georgia, published its findings in the December issue of the American Journal of Psychiatry.

According to the study, "This process is one that has been typical for the history of medicine. For example, "dropsy" is now recognized not to be a distinct disease, but a syndrome with multiple causes, including a number of primary heart and kidney diseases, that require disease-specific treatments. ‘Psychosis’ may similarly designate a series of biologically distinct syndromes with different causes and molecular biological roots. Him"

Considerable evidence has shown that a symptom-based diagnosis of psychotic illness incompletely captures biologically meaningful differentiations, often resulting in less-than-satisfactory treatments.

B-SNIP took 2,700 psychotic individuals, their first-degree relatives, and a group of healthy control subjects through a biomarker battery of tests that included cognitive abilities, oculomotor movements, and electroencephalography (EEG) measures as well as several modalities of magnetic resonance imaging (MRI).

All of the subjects were genotyped. Plotting the results of the biomarker battery demonstrated three distinct clusters of psychotic individuals, or “Biotypes," each of which contained individuals from all three traditional psychiatric diagnostic types.

Biotype 1 was the most impaired group, demonstrating poor cognition and eye-tracking capabilities, and the least brain tissue compared to controls, primarily distributed over frontal, temporal and parietal regions of the brain. Their brainwave responses were significantly lower than those of controls.

Although all of the usual psychosis diagnoses appear in Biotype 1, there was a slight predominance (59 percent) of schizophrenia cases in this most impaired Biotype, and they tended to have more severe psychotic symptoms (hallucinations and delusions) than the other groups.

Biotype 2 demonstrated cognitive impairment and poor eye-tracking, but demonstrated higher brain wave response compared to controls as measured by EEG, something neuroscientists often call “noisy brain.” These individuals are often rated as overstimulated, hyperactive, or hypersensitive.

Biotype 2 also had reduced gray matter in frontal and temporal regions, but to a lesser degree than found in Biotype 1. Biotype 2 cases also had worse scores on mood scales, such as depression and mania.

Biotype 3 was the least impaired, with near-normal cognition, EEG function, and brain structure, and moderate severity of symptoms. Those in this group were slightly more likely to be diagnosed with bipolar disorder (60 percent).

“What’s at the same time puzzling and fascinating is that all three biologically driven disease constructs, or biotypes, might be clinically diagnosed as having schizophrenia, schizoaffective, or bipolar disorder,” said Carol Tamminga, MD, the study investigator at the University of Texas Southwestern in Dallas. “There are multiple examples in other fields of medicine where use of biomarkers has led to a distinction of unique diseases that overlap in their symptom presentations. Hopefully, this neurobiological examination of severe mental illness will lead to more precise, biologically meaningful diagnoses and novel treatments.”

B-SNIP was funded by the National Institute of Mental Health (NIMH), and is part of the NIMH Research Domain Criteria (RDoC) initiative. The RDoC initiative aims to develop the fundamental data for basing psychiatric diagnoses on biological characteristics, instead of merely on clinical symptoms. It attempts to take normal brain cognitive and affective functions and to describe the associated genes, molecular structures, and neural circuits. The goal is to develop a framework for modeling mechanisms of brain diseases.

NIMH recently funded the second wave of BSNIP, to assess whether the three Biotypes can be replicated in an independent sample, and are stable over time.

Submitted by Christopher Gardner on December 08, 2015