T81. DOES COGNITIVE CONTROL PERFORMANCE PREDICT TREATMENT RESPONSE IN PSYCHOSIS?. (9th April 2019)

• Record Type: Journal Article
• Authors: Szentgyorgyi, Timea; Vanes, Lucy; Mouchlianitis, Elias; Patel, Krisna; Wong, Katie; Barry, Erica; Joyce, Dan; Collier, Tracy; Shergill, Sukhi
• Is Part Of: Schizophrenia bulletin. Volume 45(2019)Supplement 2
• Page Start: S234
• Page End: S235
• Publication Date: 2019-04-09
• Subjects: Schizophrenia -- Periodicals; Schizophrenia -- Research -- Periodicals; 616.898005
• DOI: 10.1093/schbul/sbz019.361
• Languages: English
• ISSNs: 0586-7614

Abstract

Abstract: Background: Treatment resistant schizophrenia remains a salient issue in psychiatric practice with approximately 30% of patients not responding to conventional antipsychotic medication, which exerts its effect by targeting striatal dopaminergic neurons. This has led to the emergence of theories suggesting that mechanisms other than hyperdopaminergia could explain psychotic symptoms in treatment resistant cases. One such theory is that cognitive control – mediated by frontal mechanisms and known to be impaired in schizophrenia – is further impaired in treatment resistant patients. To test this hypothesis, we employed a classical cognitive control task, the Stroop task, employing an fMRI design to compare BOLD signal and task performance in a first-episode psychosis cohort stratified into treatment resistant (TR) and non-treatment resistant (NTR) groups. Methods: 50 first-episode psychosis patients with a maximum illness onset interval of 2-years were recruited from the South London area. Patients were invited for two separate assessments, one at baseline and one at 12-month follow up, during which they underwent an fMRI scan performing the Stroop task. Symptoms were assessed using the Positive and Negative Syndrome Scale (PANSS) and IQ was assessed with the Wechsler Abbreviated Scale of Intelligence (WASI). Here, we report the preliminary findings of the baseline visit. The event-related Stroop paradigm involved two conditions: Congruent and Incongruent trials. Neural activation associated with the Stroop contrast (Incongruent > Congruent) was compared between TR (N=17) and NTR (N=33) patients. Voxel-based whole brain and region of interest (anterior cingulate cortex) analyses were performed. Behavioural analyses involved the comparison of the two groups on Reaction Time (RT) Stroop-effect (RT Incongruent – RT Congruent) and Accuracy Stroop-effect (Accuracy Congruent – Accuracy Incongruent). Unpaired t-tests were performed to compare the task performance of each group. Treatment resistance was determined after the completion of the follow-up visit based upon the TRRIP criteria: at least 2 different antipsychotic courses of therapeutic dose for at least 6 weeks, at least two PANSS positive symptom items with a score of 'moderate' or above, and total PANSS score reduction of less than 20%. Results: TR (M=200.83, SD=29.23) patients showed a significantly greater RT Stroop-effect than NTR patients (M=121.33, SD= 83.19): t=3.16, p=0.004, d=0.98. There were no significant group differences for the Accuracy Stroop-effect. Furthermore, the two patient groups showed no significant differences in age or IQ scores, hence these variables were not included as covariates. There were no significant mean activation differences between TR and NTR patients on a whole brain level or within the anterior cingulate cortex ROI. Discussion: These results indicate an observable behavioural discrepancy between the two groups at baseline indicating an underlying cognitive deficit that might play a role in the treatment resistance of psychotic symptoms. Yet this finding is not paralleled by differential mean brain activation between groups. A possible explanation is that similar brain activation in TR patients compared to NTR patients is insufficient for the exertion of comparable cognitive control. However, a more complex cognitive control task may be more suitable to detect subtle differences in brain activation. A strength of this study is the robust and prospective classification criteria used at the completion of the final visit to stratify patients. Future analyses will focus on longitudinal change in behaviour and brain activation as a predictor of treatment response; a larger sample will add more power to detect between group differences.