Search

Background Early response to first-line antipsychotics is associated with positive long-term symptomatic outcome in psychosis. Unfortunately, attempts to identify reliable predictors of treatment response in first-episode psychosis (FEP) patients have not yet been successful. One reason for this could be that FEP patients are heterogeneous in terms of symptoms and underlying disease biological mechanisms, thereby impeding the identification of one-size-fits-all predictors of treatment response. Methods We have used a two-step hierarchical clustering approach to stratify 325 FEP patients into four subtypes, termed C1A, C1B, C2A and C2B, based on their symptoms assessed using the Positive and Negative Syndrome Scale. After validating our clustering approach on external measures, we then used a regularized logistic regression method to identify biological and clinical variables associated with increased odds of being non-remitters within each subtype. Results Compared to those from the other subtypes, C1A patients exhibited higher serum levels of several pro-inflammatory cytokines and were the most at risk of being non-remitters when treated with a second-generation antipsychotic drug. In C1A patients, but not in others, lower serum levels of interleukin-15, higher levels of C-X-C motif chemokine 12, previous exposure to cytomegalovirus, use of recreational drugs and a younger age were all associated with higher odds of being non-remitters 4 weeks after treatment. The predictive value of this model was 73%, and its specificity and sensitivity were 45% and 83% respectively. Conclusions Serum levels of two cytokines combined with three clinical variables predict remission in a subtype of FEP patients. [ABSTRACT FROM AUTHOR]

Background Using Magnetic Resonance Spectroscopy (MRS) we have recently shown that, in individuals with a Clinical High Risk (CHR), transition to psychosis is associated with increased left hippocampal glutamate levels at presentation. We have also shown that relative to healthy controls (HC), CHR participants show reduced hippocampal activity during a novelty salience task when processing pure stimulus novelty. However, the extent to which elevated glutamatergic metabolite levels in CHR cohorts are related to reduced hippocampal function has yet to be investigated. Methods Seventy-five individuals with a CHR for psychosis and 31 HC completed MRS and functional Magnetic Resonance Imaging during a novelty salience task to measure hippocampal glutamatergic metabolite levels and task-related activity (during pure stimulus novelty) respectively. In the CHR group 13 participants made a transition to psychosis at clinical follow-up (mean = 18.5 months). Three-way interactions were modeled between task-related activity, hippocampal glutamatergic metabolite levels and group (HC vs. CHR) using a Region of Interest (ROI) approach in bilateral hippocampus and striatum (which receives glutamatergic outputs from the hippocampus and is also activated during a novelty salience task). Results The CHR group were significantly younger and had received fewer years of education than HC. CHR and HC groups were matched for gender, estimated IQ scores, tobacco and alcohol use. Ten CHR participants were receiving low doses of antipsychotic medication. There was no interaction between left hippocampal glutamate levels, task-related activity and group in either ROI. There was a significant interaction between Glx levels (combined glutamate and glutamine peaks), task related activity and group in the right hippocampus (x = 28, y = -30, z = -6; Z =3.62; p FWE =.02) and the right dorsal striatum (x = 6, y = 16, z =6; Z = 3.67 p FWE =.01). In both hippocampal and striatal ROIs the 3-way interaction was driven by a positive association between hippocampal Glx levels and activity during pure stimulus novelty in the HC group. In the CHR group however, the relationship between Glx and task-related activity was negative and largely driven by CHR participants that had transitioned to psychosis at follow-up. Discussion Hippocampal glutamatergic metabolite levels are associated with altered functional activation in the hippocampus and dorsal striatum during a novelty salience task. These findings are consistent with previous clinical and pre-clinical work that posit a role for glutamate in hippocampal dysfunction and associated risk for psychosis. [ABSTRACT FROM AUTHOR]

Background Exposure to infectious microorganisms has been implicated as an important risk factor in the development of psychotic disorders. Epidemiological and laboratory evidence suggests that exposure at various developmental periods may increase psychosis risk, from in utero through to childhood and adult life. Individuals who meet clinical high risk (CHR) criteria are at substantially increased risk of developing a psychotic disorder; however, no study to date has explored whether exposure to schizophrenia-associated pathogens is associated with clinical outcomes in this population. Methods Plasma samples from 254 subjects at ultra-high risk of psychosis from the EU-GEI cohort and 116 age- and sex-matched controls were tested for antibodies to the following infectious organisms: Toxoplasma gondii, herpes simplex virus type 1, cytomegalovirus, Epstein-Barr virus (EBV) and influenza. Samples were also tested for antibodies to the NMDA receptor using a fixed cell-based assay and for levels of S100B (a putative marker of blood-brain barrier disruption) and complement pathway protein expression. Only pathogens which predicted case-control status were taken through to further analysis looking at symptomatic associations and clinical outcome measures, which included the Positive and Negative Symptom Scale (PANSS), the Scale for Assessment of Negative Symptoms (SANS) and Global Assessment of Function (GAF). Results In logistic regression analysis, exposure to Toxoplasma was associated with increased risk of meeting CHR criteria (OR 2.66; p = 0.04), whereas EBV exposure was associated with decreased risk of meeting CHR criteria (OR 0.34; p = 0.006). Toxoplasma IgG antibody status was associated with higher SANS scores (p = 0.032) and higher GAF symptoms (p = 0.003) and disability scores (p = 0.003). EBV antibody status was associated with lower GAF scores (p = 0.007), indicating better functioning. In clinical outcome analysis, EBV antibody positivity was associated with a reduced risk of transition to psychosis (OR 0.44; p = 0.04), whereas Toxoplasma IgG seropositivity was associated with poorer functional outcomes at follow-up. The presence of NMDAR autoantibodies was associated with higher Toxoplasma IgM (p = 0.024), but not IgG (p = 0.972), titres. Furthermore, Toxoplasma IgM was associated with increased levels of serum S100B (p = 0.007) and with increased expression of Complement C4 (p = 0.018, FDR corrected), both of which have been independently implicated in psychosis risk. Discussion Toxoplasma exposure is associated with severity of negative symptoms and poor functional outcomes in CHR individuals, consistent with the known associations of Toxoplasma exposure in psychotic patients. Specific mechanisms by which Toxoplasma exposure might confer risk of poor outcomes include activation of specific complement pathway proteins, disruption of the blood-brain barrier and infection-induced production of brain-reactive autoantibodies. EBV exposure appears to have a mitigating effect, symptomatically and in terms of outcome, in this population, and as such may represent one of the first reported neurobiological protective factors for the CHR state and for the subsequent development of psychosis. [ABSTRACT FROM AUTHOR]

Background Individuals at clinical high risk (CHR) for psychosis have approximately a 25% chance of transitioning to psychosis in the first 2 years after first presentation to clinical services. A key aim in this field is to determine the risk of conversion for an individual meeting the criteria for CHR based on clinical, demographic and neuropsychological measures. An individualized risk calculator incorporating 8 risk factors based on these measures has recently been developed by the researchers from the North American Prodrome Longitudinal Study (NAPLS-2) and tested in their dataset of CHR participants. Methods We examined the risk factors from the NAPLS risk calculator in the EU-GEI (EU funded gene environment interaction) high risk study. The EU-GEI high risk study is a longitudinal multi-centre study including 9 sites in Europe, 1 site in Australia and 1 site in Brazil. At baseline, the study included 345 CHR individuals of which 65 later developed psychosis. The NAPLS risk calculator includes 8 risk factors and for each risk factor we attempted to harmonise the EU-GEI measures to the corresponding NAPLS measure. We used multivariate cox regression to determine the standardised hazard ratio for transition to psychosis for the 8 risk factors. To compare each of the NAPLS and EU-GEI hazard ratios we used a meta-analytical framework to calculate a combined hazard ratio and to examine heterogeneity between the effect sizes. Results In terms of the 8 risk factors, only the CAARMS unusual thought + non-bizarre ideas (corresponding to NAPLS SIPS P1+P2) was significantly associated with transition to psychosis, standardised hazard ratio = 1.51 (95%CI 1.01–2.23), p=0.046. Hazard ratios for the remaining risk factors varied from 0.73 to 1.41. The meta-analysis combining EU-GEI and NAPLS data did not indicate heterogeneity for any of the 8 measures, suggesting no systematic differences in the risk factors between the two studies. Discussion The test of the NAPLS risk factors in the independent EU-GEI high risk sample showed support for unusual thought content and non-bizarre ideas being predictive of later transition to psychosis. We were not able to replicate decline in functioning and lower scores in verbal learning memory as predictive of transition to psychosis as seen in the NAPLS study. However, our meta-analytical comparison showed no systematic differences in the hazard ratios for all 8 risk factors between both studies suggesting broad comparability between the samples. [ABSTRACT FROM AUTHOR]

Background Oculomotor abnormalities are among the most widely replicated findings in psychotic disorders with impairments in Smooth Pursuit Eye Movements (SPEM) and Antisaccades (AS) tasks. Cannabis-use may be a clinically relevant demarcating factor in early psychosis seen in around half of FEP patients at presentation and indicative of worse prognosis. No study to date has aimed at differentiating cannabis-using from non-using patients using AS and SPEM. Differences in eye movement measures between groups would indicate altered neurobiology between cannabis-using and non-using patients. We thus aimed to determine whether abnormalities in SPEM and AS are indicative of neurobiological differences between Early Psychosis patients who use cannabis (EPC) and those who do not use cannabis (EPNC). Methods We undertook the EfCiP study (the Effect of Cannabis in Psychosis) consisting of four groups: Early Psychosis patients who used cannabis (EPC) (n=28), Early Psychosis patients who did not use cannabis (EPNC) (n=25), healthy controls who used cannabis (HCC) (n=16) and healthy controls who did not use cannabis (HCNC) (n=22). The smooth pursuit stimulus used a sinusoidal waveform at three target frequencies (0.2, 0.4 and 0.6 Hz) across a horizontal range. Participants performed 40 antisaccade trials using a horizontal step task with an equal number of left and right stimuli in randomised order. As primary outcomes we tested measures robustly associated with abnormalities in psychosis: maintenance gain to index smooth pursuit and antisaccade percentage errors. We used one-way ANCOVA on four levels (EPC, EPNC, HCC and HCNC) with post-hoc Bonferroni correction. Covariates entered into the model were age, sex, AUDIT and Fagerstrom's scores. Results EPC and EPNC were clinically indistinguishable with no difference across clinical measures: including GAF, PANSS, duration since diagnosis, days spent in hospital, chlorpromazine equivalents, proportion with affective component to psychosis or schizophreniform spectrum disorder. Participants with EPC had higher indices of cannabis use and also scored higher on Fagerstrom's and AUDIT scores to index smoking and alcohol use respectively. Repeated measures ANCOVA revealed a significant effect of task, indicating that mean gain decreased as SPEM frequency increased. There was a significant Group effect (F3,83=3.514, p=0.019, ηp2=0.113). Post-hoc tests revealed significantly reduced mean gain for EPNC vs EPC (p=0.021). There was no significant difference between HCC and HCNC (p=1). There was no significant effect on SPEM gain for any covariate. When controls were combined there was a significant reduction in SPEM gain for EPNC vs EPC (Hedge's g=0.830, p=0.014) and trend level reduced gain for EPNC vs all controls (Hedge's g=0.594, p=0.058) but no difference between EPC and all controls (p=0.998) There was a numerical decrease in antisaccades percentage error rate such that EPC≈EPNC>HCC>HCNC but there was no significant group effect (p>0.35). Discussion This study demonstrates neurobiological distinction between cannabis using and non-using patients. There is evidence of impairment of SPEM in the EPNC group but not in the EPC group, whereas there is no discernable effect of cannabis use on SPEM gain in the healthy situation. This is consistent with lower psychosis liability in EPC than EPNC and suggests an additive component for cannabis use in the EPC group. Differential performance in SPEM and AS may indicate impairments in specific circuitry between groups. Further work should look to delineate the neurobiological substrates for this distinction. [ABSTRACT FROM AUTHOR]

Background The neurobiological mechanisms underlying the association between cannabis use and acute or long-lasting psychosis are not completely understood. While some evidence suggests altered striatal dopamine may underlie the association, direct evidence that cannabis use affects either acute or chronic striatal dopamine is inconclusive. In contrast, pre-clinical research suggests that cannabis may affect dopamine via modulation of glutamate signaling. Methods A double-blind, randomized, placebo-controlled, crossover design was used to investigate whether altered striatal glutamate, as measured using proton magnetic resonance spectroscopy, underlies the acute psychotomimetic effects of intravenously administered delta-9-tetrahydrocannabinol (Δ9-THC; 1.19 mg/2 ml), the key psychoactive ingredient in cannabis, in a set of 16 healthy participants (7 males) with modest previous cannabis exposure. Results Compared to placebo, acute administration of Δ9-THC significantly increased Glutamate (Glu) + Glutamine (Gln) metabolites (Glx) in the left caudate head (P=0.027). Furthermore, compared to individuals who weren't sensitive to the psychotomimetic effects of Δ9-THC, individuals who developed transient psychotic-like symptoms (~70% of the sample) had significantly lower baseline Glx (placebo; P=0.023) and a 2.27-times higher increase following Δ9-THC administration. Lower baseline Glx values (r=-0.55; P=0.026) and higher previous cannabis exposure (r=0.52; P=0.040) were associated with a higher Δ9-THC-induced Glx increase. Discussion These results suggest that increase in striatal glutamate levels may underlie acute cannabis-induced psychosis while lower baseline levels may be a marker of greater sensitivity to its acute psychotomimetic effects and may have important public health implications. [ABSTRACT FROM AUTHOR]

Background Accumulating research suggests that altered levels of neurochemical metabolites -such as glutamate- play a role in psychosis pathophysiology, are present prior to onset in people at Clinical High Risk (CHR-P) and predict clinical outcomes. Oxytocin, a neuropeptide with prosocial and anxiolytic properties, modulates glutamate neurotransmission in animals but its neurochemical effects in CHR-P individuals remain unknown. The present study examined the effects of intranasal oxytocin on glutamate and glutamate+glutamine (primary hypothesis) and other neurochemical metabolites (secondary hypothesis) in CHR-P individuals. Methods In a double-blind, placebo-controlled, crossover design, 30 CHR-P males underwent two MRI scans at 3 Tesla, once after 40IU intranasal oxytocin and once after matched placebo (one-week wash-out). Proton magnetic resonance spectroscopy (1H-MRS) was used to measure the effects of oxytocin on glutamate, glutamate+glutamine and other metabolite levels (choline, N-acetylaspartate, myo-inositol) scaled to creatine in the hippocampus, anterior cingulate cortex and thalamus. Results Data analysis is currently ongoing, and the results will be presented at the conference. Discussion These results will provide the first neurophysiological evidence for the potential effects of oxytocin on neurochemical metabolites in CHR-P individuals. Given the current lack of evidence for effective treatments in this patient group, continued efforts to find novel molecules with a relevant profile of effects remains a priority. [ABSTRACT FROM AUTHOR]

Background The non-intoxicating cannabinoid compound cannabidiol (CBD) may have the potential to become a new effective, safe and well-tolerated antipsychotic drug. CBD may be particularly feasible as a treatment for people at clinical high risk (CHR) for the development of psychosis due to its lack of serious side effects. Preclinical models suggest that abnormalities in functioning of the medial temporal lobe and striatum are fundamental in the pathophysiology of psychosis. Recent neuroimaging studies showed increased resting state perfusion in these brain regions in CHR individuals, but it is unknown how this is affected by CBD treatment. Here we examined the impact of treatment with CBD on resting state perfusion in people at CHR for psychosis. Methods Using a parallel-group, double-blind, placebo-controlled, randomised design, 33 individuals who met CHR criteria were treated for three weeks with either a single daily oral dose of 600 mg CBD (N=16) or placebo (N=17). Resting state perfusion was measured three hours after the first administration (day 1) and after three weeks of treatment (day 21) using Arterial Spin Labelling (ASL). Complete ASL data sets were obtained for 23 CHR individuals (placebo N=10, CBD N=13). ASL images were preprocessed in SPM8 and normalised for global perfusion. Resting state perfusion was examined in a network relevant to psychosis comprising the parahippocampal gyrus, hippocampus and striatum. Mean perfusion values were extracted from these brain areas and analysed in SPSS. Symptomatology was assessed with the Comprehensive Assessment of At-Risk Mental State (CAARMS) at baseline and after three weeks of treatment. For both symptomatology and perfusion, difference scores were calculated by subtracting results from the first measurement from those obtained at the follow-up assessment. Group and treatment effects were determined with two-sample t-tests or χ2 tests and associations with Pearson's correlation. Results There were no significant differences between the two treatment groups in terms of age, gender, years of education, handedness or baseline symptomatology. Treatment with CBD significantly reduced resting state perfusion in the psychosis network compared to placebo (-0.009±0.034 and 0.028±0.042, p=0.031). Post hoc analysis revealed a trend towards a significant decrease in perfusion after CBD treatment in the parahippocampal gyrus (-0.015±0.057 and 0.030±0.052, p=0.063). No significant post hoc treatment effects were found on perfusion in the hippocampus (CBD -0.008±0.066, placebo 0.027±0.062, p=0.216) or striatum (CBD -0.006±0.047, placebo 0.026±0.050, p=0.126). Total symptom scores decreased with CBD treatment compared to placebo, although this did not reach significance (-58.6±54.4 and -28.9±62.5, p=0.157). There were no significant correlations between resting state perfusion and symptomatology. Discussion Three-week treatment with CBD reduced resting state perfusion of individuals at CHR for the development of psychosis in a brain network comprising the medial temporal lobe and striatum. This is consistent with previous neuroimaging work that showed that resolution of the high-risk state was associated with attenuation of resting perfusion in these areas. Our findings suggest that the therapeutic effects of CBD in CHR individuals may be mediated through normalisation of regional resting state perfusion. Results support further examination of CBD as a treatment for people at CHR for psychosis in large-scale multicentre clinical trials. [ABSTRACT FROM AUTHOR]

Background The inappropriate attribution of salience to what would normally be irrelevant or neutral stimuli is a robust feature of psychotic disorders and is linked to altered subcortical dopaminergic signaling. Preclinical models suggest that the latter is driven by interactions within a hippocampal-striatum-midbrain circuit. Although novelty has been less investigated as a dimension of salience processing in psychosis than reward, preclinical evidence indicates that dopaminergic neurons in the midbrain code the salience of unexpected stimuli and respond to novel stimuli. Our first aim was to examine whether subjects at clinical high risk (CHR) for psychosis show altered activation and effective connectivity in the putative hippocampal-striatum-midbrain circuit using fMRI during the processing of novelty salience. The second aim was to examine the relationship between activation and effective connectivity within this circuit and the subsequent onset of a psychotic disorder, by following our CHR participants clinically for around 5 years after scanning to determine their clinical outcome. Methods Seventy-six CHR participants and 31 healthy controls (HC) were studied using fMRI while performing a task that engaged novelty salience processing. The CHR sample was then followed clinically for a mean of 59.7 months (~5 years): during this period, 13 individuals (17%) developed a psychotic disorder (CHR-T) and 63 did not. Functional activation and effective connectivity within a hippocampal-striatal-midbrain circuit were compared between groups using the contrast Novel versus Neutral trials (indicating 'pure' novelty), using SPM12 and dynamic casual modelling (DCM12). Effects were considered significant at a voxel-wise threshold of p < 0.05 family-wise error corrected (FWE). Results In CHR individuals, the hippocampal response to pure novelty was significantly attenuated compared to that in HC (p = 0.041 FWE). DCM12 revealed that pure stimulus novelty modulated effective connectivity from the hippocampus to the striatum, and from the midbrain to the hippocampus, significantly more in CHR participants than in HC. Conversely, pure stimulus novelty modulated connectivity from the midbrain to the striatum significantly less in CHR participants than in HC, and less in CHR participants who subsequently developed psychosis (CHR-T) than in CHR subjects who did not become psychotic. Discussion These data indicate that both hippocampal activation and hippocampal-striatal-midbrain effective connectivity in the context of pure novelty salience are perturbed in people at CHR for psychosis, and that the later onset of psychosis is associated with alterations in midbrain-striatal connectivity. These findings are consistent with data from preclinical models of psychosis implicating alterations in a hippocampal–striatal-midbrain circuit in the development of psychosis. [ABSTRACT FROM AUTHOR]

Background Neuroimaging studies in schizophrenia have linked elevated glutamate metabolite levels to non-remission following antipsychotic treatment and have also indicated that antipsychotics can reduce glutamate metabolite levels. However, the relationship between symptomatic reduction and change in glutamate during initial antipsychotic treatment is unclear. Methods Proton magnetic resonance spectroscopy (1H-MRS) measured Glx and glutamate in the anterior cingulate cortex (ACC) and thalamus in patients with first episode psychosis (n=23) at clinical presentation, and after 6 weeks and 9 months of treatment with antipsychotic medication. At 9 months, patients were classified into Remission (n=12) and Non-Remission (n=11) subgroups. Healthy volunteers (n=15) were scanned at the same three time-points. Results Glx levels in the thalamus increased over time in Non-Remitters (P=0.008), such that after 9 months Glx levels were higher in patients who were not in remission than in those who were (P=0.033). No change in thalamic Glx levels over time were evident in the Remission subgroup. In addition, the change in Glx in the thalamus over the 9 months of treatment was positively correlated with the change in the severity of PANSS positive, total and general symptoms (P<0.05). There were no significant effects of group or time on glutamate metabolites in the ACC, and no differences between either patient subgroup and healthy volunteers. Discussion These data suggest that the nature of the response to antipsychotic medication may be related to the pattern of changes in glutamatergic metabolite levels over the course of treatment. The findings add to existing evidence linking the response to treatment in schizophrenia to alterations in central glutamate function. [ABSTRACT FROM AUTHOR]

Background People with psychotic disorders have a markedly reduced life expectancy compared to the general population. Comorbid substance use disorders (dual diagnosis) contribute towards poorer treatment engagement and worse clinical outcomes but their impact on mortality is uncertain. We estimated the association of comorbid substance use disorders with life expectancy in a large sample of patients with psychotic disorders using an electronic health record (EHR) case register. Methods Data were obtained from 29,412 people with schizophrenia (61.9%), schizoaffective disorder (4.9%) or bipolar disorder (33.2%) receiving mental healthcare from the South London and Maudsley NHS Foundation Trust between 1st January 2007 and 31st December 2017. The Clinical Record Interactive Search tool (CRIS) was used to extract data on age, gender, ethnicity, comorbid substance use disorders (ICD-10 F10-19) and mortality from anonymised EHRs. The Chiang II method was used to estimate life expectancy from birth associated with substance use disorders using life-table data obtained from the UK Office of National Statistics. Results 3,657 patients had a comorbid substance use disorder of whom 70.1% were male. The most frequently documented comorbid substance use disorders were related to alcohol (44.9%), multiple drug use (31.3%) and cannabis (25.2%). 3,396 patients died during the study period. The life expectancy for patients with a comorbid substance use disorder was 63.1 years (95% CI 60.7 – 65.5) for men and 63.4 years (95% CI 60.2 – 66.7) for women. The life expectancy for patients without a comorbid substance use disorder was 65.1 years (95% CI 64.1 – 66.0) for men and 69.8 years (95% CI 68.9 – 70.8) for women. Patients with a comorbid substance use disorder were more likely to die of unnatural causes (28.5%) compared to those without (10.0%). Discussion Comorbid substance use disorders were most frequently recorded among male patients but associated with greatest reduction in life expectancy among female patients. Patients with a comorbid substance use disorder were almost three times more likely to die from unnatural causes (accidental, suicide or homicide). These findings demonstrate the need to better address comorbid substance use disorder in people with psychotic disorders, particularly among female patients who experienced a marked reduction of around six years in life expectancy. [ABSTRACT FROM AUTHOR]

Background The development of adverse clinical outcomes in psychosis has been associated with behavioral and neuroanatomical changes related to emotional processing (Aleman & Kahn, 2005; Philips & Seidman, 2008). However, the relationship between emotional processing deficits and neuroanatomy in subjects at clinical high risk of psychosis (CHR), and their relationship to longitudinal outcomes, remains unclear. The primary aim of the present study was to investigate the association between facial emotional recognition and grey matter volume (GMV) in a large cohort of CHR individuals compared to healthy controls, and their relationship with subsequent clinical outcomes. Methods Structural imaging scans at 3T and a computer-based measure of emotional processing (Degraded Facial Affect Recognition task, DFAR), were collected at baseline from 53 healthy controls (mean age: 23.3 SD ± 3.9; 52.8% male) and 213 CHR subjects (mean age: 22.9 SD ± 4.7; 50.7% male) participating in the multi-centre EU-GEI study (data pooled from 9 sites). CHR subjects were then clinically monitored for 12 months and clinical outcomes were assessed in terms of transition/non-transition to psychosis (CAARMS criteria) and the level of overall functioning (Global Assessment of Function scale; GAF). Structural images were preprocessed with voxel-based morphometry in SPM12. DFAR performance (total, happy, angry, fear and anger) was analyzed with repeated measures ANOVA at baseline (HC/CHR as between-group factor and DFAR as within-subject factor, adjusted for age, gender, IQ and site); and with binary logistic regression at the 12-month follow-up point (adjusting for the same variables). One-way ANOVAs in SPM12 were specified to examine interactions between group status (HC/CHR, CHR-T/CHR-NT, CHR-GO/CHR/PO) and DFAR performance (total, happy, angry, fear and anger), adjusting for the same variables. Statistical inferences were made at p<.05 after voxel-wise FWE correction within four pre-defined anatomical regions of interest (ROI: amygdala, hippocampus, insula, anterior cingulate cortex/medial prefrontal cortex -ACC/MPFC-). Results At baseline, DFAR performance did not differ between HC and CHR groups and was not significantly associated with GMV in our ROIs. At the 12-month follow-up point, 39 CHR subjects showed good overall functioning (GAF≥65), whereas 93 CHR subjects had a poor functional outcome (GAF score<65); 44 CHR subjects developed a psychotic disorder and 169 did not. Compared to subjects with a good functional outcome, CHR subjects with poor overall functioning showed exacerbated recognition of angry facial emotion (p=.030), decreased GMV in ACC and hippocampus (both p=.033 FWE), and a significantly different (inverse) association between angry facial emotion recognition and hippocampal volume compared to CHR subjects with good functioning (p=.037 FWE). CHR subjects who developed psychosis did not show significant differences in DFAR performance or significant associations between DFAR and GMV compared with CHR subjects who did not become psychotic. Discussion These findings indicate that adverse global functional outcomes in people at CHR of psychosis (but not transition to psychosis) are associated with baseline deficits in the recognition of angry facial emotion, which are in turn directly related to hippocampal volume decreases compared to CHR subjects with good functional outcomes. This has implications not only for stratification of CHR subjects according to baseline behavioral and volumetric characteristics, but also suggest potential preventative avenues for poorer outcomes focused on social cognitive and emotional processes. [ABSTRACT FROM AUTHOR]

Background Comorbid illicit substance use is associated with poor clinical outcomes in established psychotic disorders. However, less is known about the impact of substance misuse in first episode psychosis (FEP). We conducted a largescale electronic health record (EHR) study using text mining software to investigate the associations of cannabis, cocaine, amphetamines and MDMA with risk of psychiatric hospitalization. Methods Data were obtained from 1,835 people with FEP in South London using the Clinical Record Interactive Search tool (CRIS). TextHunter was used to identify exposure to cannabis, cocaine, amphetamines or MDMA documented in free text EHRs. Their relationship with number of psychiatric hospital admissions over a 5-year period was analyzed using multivariable negative binomial regression with age, gender and ethnicity as covariates. Results The most frequently used illicit substance was cannabis (58.5%) followed by cocaine (24.4%), amphetamines (3.7%) and MDMA (3.1%). Cannabis (IRR 1.58, 95% CI 1.37 to 1.83) and cocaine (1.36, 1.17 to 1.59) were associated with increased frequency of psychiatric hospital admission. Amphetamines showed no significant association (1.05, 0.77 to 1.45) while MDMA was associated with a reduction in frequency of admissions (0.59, 0.40 to 0.86). Discussion Cannabis and cocaine are associated with a significant increase in frequency of psychiatric hospital admissions. These findings highlight the importance of addressing comorbid illicit substance use in people with FEP. [ABSTRACT FROM AUTHOR]