Email Icon Facebook icon Twitter Icon GooglePlus Icon Contact

User Top Menu

ICOSR 2017: Schizophrenia Biomarkers Session Ranges Widely

2 May 2017

As part of our ongoing coverage of the 2017 International Congress on Schizophrenia Research (ICOSR), held March 25-28 in San Diego, we bring you session summaries from some of the awardees of the Young Investigator travel grants. We are, as always, grateful for the gracious assistance of YI program directors Laura Rowland and Scott Sponheim, as well as Michelle Tidwell of the ICOSR staff. For this report, we thank Sarah Haigh of the University of Pittsburgh Medical Center.


The Monday, March 27, biomarkers oral session at the International Congress on Schizophrenia Research focused on a variety of neurological and physiological biomarkers for schizophrenia. The majority of talks focused on immunological markers, with a few talks on electrophysiological and behavioral markers. The use of the term "biomarker" ranged from indicators of disease presence and mechanistic markers indicating where the deficit lies in schizophrenia to mechanisms for potential treatment.

The first speaker was Alp Üçok of Istanbul University, who discussed the presence of abnormal prepulse inhibition (PPI) in schizophrenia. PPI has been shown to be heritable and correlates with cognitive performance. He measured PPI in CHR (CHR) patients (those who show symptoms of schizophrenia below the diagnostic threshold) and FHR (FHR) patients (who have a first-degree relative with schizophrenia). Üçok showed significantly lower PPI in the CHR compared to the FHR and control groups. There was no significant difference between FHR subjects and controls. This indicates that abnormal PPI is related to the presence of symptoms; however, PPI did not correlate with symptoms (from the BPRS, SANS, or SAPS). He is still recruiting participants, so stay tuned for the completed study.

Next up, David Cotter of the Royal College of Surgeons in Dublin investigated the changes in plasma proteins in individuals who were 11 years old compared to those who were 18 years old, and related the proteins to presence (or absence) of psychotic features, as part of the Avon Longitudinal Study of Parents and Children. Thirty-four proteins were differentially expressed in those with psychotic symptoms and have a prediction accuracy of over 90 percent. Ten proteins were differentially expressed in those with full psychosis, but were less significant. This finding was verified in a second cohort. These proteins are inflammatory and immune markers (specifically, complement and coagulation systems), and suggest abnormal blood markers prior to psychosis onset.

Thomas Weickert of the University of New South Wales in Australia also focused on inflammatory markers, specifically, cytokines in prefrontal cortex, in schizophrenia. Cytokines have been linked to a variety of outcomes including cell death and neuropathology, or can be neuroprotective. Cytokine levels are related to the kynurenine pathway. mRNA proteins from the kynurenine pathway degrade tryptophan and link the neurotransmitter and immune systems. In postmortem brain tissue, Weickert divided the schizophrenia group into those with high cytokine levels and those with low levels. In the high cytokine group, he found a significant correlation between cytokine levels and KAT II mRNA expression. The metabolites of kynurenic acid were lower, but a higher ratio of mRNA and tryptophan were found in schizophrenia. The high cytokine group showed significant correlations between the mRNA/tryptophan ratio and attention measures, and with prefrontal cortex volume. Cytokines and the kynurenine pathway could provide a measure of prefrontal dysfunction in schizophrenia and further link the immune system with disease pathology.

The focus of the session changed with the presentation by Anna Docherty of the University of Utah, who investigated polygenic risk for schizophrenia. She used a subset of the Wellcome Trust’s case-control consortium data, and assessed risk based on GWAS data and the expression of certain predetermined gene and single nucleotide polymorphisms. Sex and ancestry were included in the analysis. Using Bayesian modeling, and a leave-one-out method to ascertain accuracy, she found polygenic risk factors related to symptoms. These risk factors were related to DNA and RNA repair. The highest risk factor resulted in a 50:50 risk of having schizophrenia―that is a 2.5 SD increase in risk compared to controls. Interestingly, the risk factors produced a spectrum of risk, unlike data in major depression, which was much more categorical and provided a clearer mechanistic pathway. While this method does not help with understanding the mechanisms underlying schizophrenia, it does provide a measure of risk that could help identify individuals before their first break.

We then switched gears again, as I presented a potential neural biomarker for schizophrenia. When a deviant stimulus is detected, the brain produces an extra negative event-related potential that can be detected in the electroencephalogram, called mismatch negativity (MMN). When the deviance is based on a change in a physical parameter of the stimulus, i.e., pitch, the amplitude of the MMN is reliably reduced in individuals with chronic schizophrenia but is not reliably reduced in individuals at their first episode of schizophrenia. One cause for this could be that the simple physical deviance is too easy for the subject and is not sensitive enough to task subtle deficits in novelty detection that might be present at first episode. Therefore, we measured novelty detection in complex patterns. We presented frequent groups of three identical tones, with the occasional group containing an additional fourth tone. We found a late MMN in healthy controls that was significantly blunted in chronic schizophrenia and in our first-episode schizophrenia group. This suggests that the late MMN to complex pattern deviance is more sensitive at detecting subtle abnormalities in auditory novelty detection when detection cannot rely on changes in physical stimulus parameters. In our future studies, we will test late complex MMN in those at ultra-high risk of developing schizophrenia.

Ikwunga Wonodi of the University of Maryland turned the conversation back to immune system markers. He investigated the use of clozapine in African-Americans, where there is a higher incidence of neutropenia (low white blood cell neutrophils). He measured absolute neutrophil counts (ANC) in schizophrenia and control participants recruited from Lagos, Nigeria, by measuring the Duffy antigen receptor for chemokines (DARC) gene, which is related to low neutrophil counts in African populations. Patients were stable on medication for over six months. A subsample of patients was on clozapine, and the majority were homozygous for DARC. Clozapine patients had higher ANC than other patients and controls. Together, this suggests a mechanism for the increased incidence of neutropenia.

In the next talk, Mami Saito of Niigata University in Japan investigated the contribution of anti-NMDA encephalitis to incidence of psychosis. Encephalitis can cause seizures, abnormal movements, autonomic failure, and hyperventilation. The incidence of psychosis from encephalitis is low, so participants were difficult to find. Cerebrospinal fluids (CSF) were tested for anti-NMDA receptor (GRIN1) antibodies to verify anti-NMDA encephalitis. Higher frequencies of antibodies were found in patients with typical psychosis symptoms, suggesting that testing for GRIN1 antibodies can be used to verify a diagnosis of psychosis and allow clinicians to offer different options for treatment.

Haley Demyanovich of the University of Maryland focused on the contribution of gluten-related proteins to inflammation triggering psychosis. Gluten sensitivity differs from celiac disease but can still impact the immune system, and previous studies have found high levels of the gliadin protein (gluten-related) in schizophrenia (about 30 percent higher). Inflammation is known to impact neural functioning via cytokines crossing the blood-brain barrier. Using an automated enzyme-linked immunosorbent assay, she found that more gliadin was associated with more cytokines in schizophrenia. Next, using magnetic resonance spectroscopy, Demyanovich found higher glial markers (suggesting neuroinflammation) in those who were positive for gliadin. Therefore, for a subset of individuals with schizophrenia who have sensitivity to gluten, their etiology could be different from others and require different treatment.

Katherine Thakkar of Michigan State University then moved the conversation toward a cognitive neuroscience marker, specifically, ocular-motor markers of prediction failures. The human brain is a statistical organ that uses prior predictions to guide interpretation of the environment. When the brain predicts an action and then carries it out, it sends out a corollary discharge signal to the sensory cortices to verify that this was a self-initiated action. The corollary discharge is related to the subjective sense of agency. The ability to predict actions is perturbed in schizophrenia, disrupting the corollary discharge signal and impacting the sense of agency over actions. Thakkar used two tasks based on the ocular-motor system where the participants had to predict gaze position after making an eye movement. The first was a double-step task where participants followed two targets with fast presentations. The second was a blinking task where the participants moved their eyes toward the target, whereupon the target disappears and then reappears. Participants have to decide whether the target reappeared closer to or farther away from the fixation point. Animal and human lesion studies have shown that the mediodorsal thalamus produces a corollary discharge, which helps predict and refine future predictions. The corollary discharge is also used to suppress the image during a saccade. The schizophrenia group performed worse in both of these tasks, and results from the second task correlated with symptom severity. Together, this demonstrates that ocular-motor prediction tasks are useful for probing corollary discharge signals from a known frontothalamic circuit in awake, functioning patients.

Finally, Neal Swerdlow of the University of California, San Diego, investigated early auditory information processing (EAIP) in schizophrenia and how improving EAIP can help improve neurocognition. Previous findings showed that negative symptoms can mediate this effect. Focusing on mechanisms that can improve long-term potentiation, he measured effects of an NMDA antagonist (memantine) on prepulse inhibition, mismatch negativity, and gamma-band synchronization, and found improvements on all three measures in both schizophrenia and healthy controls. Swerdlow then measured effects of amphetamine on behavioral measures of auditory processing speed. He used sound sweeps training to improve thresholds and found sustained effects of learning in schizophrenia. However, there were no neurocognitive improvements, suggesting that these effects might be domain specific.

Overall, this was an illuminating session on the variety of potential biomarkers that can be targeted to identify or help treat individuals with schizophrenia. In particular, the emphasis on markers of inflammation highlights how general health can impact neurological functioning, and should not be ignored when assessing individuals with psychosis.