Psychosis Prognosis Predictor: A continuous and uncertainty-aware prediction of treatment outcome in first-episode psychosis.

INTRODUCTION: Machine learning models have shown promising potential in individual-level outcome prediction for patients with psychosis, but also have several limitations. To address some of these limitations, we present a model that predicts multiple outcomes, based on longitudinal patient data, while integrating prediction uncertainty to facilitate more reliable clinical decision-making. MATERIAL AND METHODS: We devised a recurrent neural network architecture incorporating long short-term memory (LSTM) units to facilitate outcome prediction by leveraging multimodal baseline variables and clinical data collected at multiple time points. To account for model uncertainty, we employed a novel fuzzy logic approach to integrate the level of uncertainty into individual predictions. We predicted antipsychotic treatment outcomes in 446 first-episode psychosis patients in the OPTiMiSE study, for six different clinical scenarios. The treatment outcome measures assessed at both week 4 and week 10 encompassed symptomatic remission, clinical global remission, and functional remission. RESULTS: Using only baseline predictors to predict different outcomes at week 4, leave-one-site-out validation AUC ranged from 0.62 to 0.66; performance improved when clinical data from week 1 was added (AUC = 0.66-0.71). For outcome at week 10, using only baseline variables, the models achieved AUC = 0.56-0.64; using data from more time points (weeks 1, 4, and 6) improved the performance to AUC = 0.72-0.74. After incorporating prediction uncertainties and stratifying the model decisions based on model confidence, we could achieve accuracies above 0.8 for ~50% of patients in five out of the six clinical scenarios. CONCLUSION: We constructed prediction models utilizing a recurrent neural network architecture tailored to clinical scenarios derived from a time series dataset. One crucial aspect we incorporated was the consideration of uncertainty in individual predictions, which enhances the reliability of decision-making based on the model's output. We provided evidence showcasing the significance of leveraging time series data for achieving more accurate treatment outcome prediction in the field of psychiatry.

Negative symptoms in First-Episode Schizophrenia related to morphometric alterations in orbitofrontal and superior temporal cortex: the OPTiMiSE study.

BACKGROUND: Negative symptoms are one of the most incapacitating features of Schizophrenia but their pathophysiology remains unclear. They have been linked to alterations in grey matter in several brain regions, but findings have been inconsistent. This may reflect the investigation of relatively small patient samples, and the confounding effects of chronic illness and exposure to antipsychotic medication. We sought to address these issues by investigating concurrently grey matter volumes (GMV) and cortical thickness (CTh) in a large sample of antipsychotic-naïve or minimally treated patients with First-Episode Schizophrenia (FES). METHODS: T1-weighted structural MRI brain scans were acquired from 180 antipsychotic-naïve or minimally treated patients recruited as part of the OPTiMiSE study. The sample was stratified into subgroups with (N = 88) or without (N = 92) Prominent Negative Symptoms (PMN), based on PANSS ratings at presentation. Regional GMV and CTh in the two groups were compared using Voxel-Based Morphometry (VBM) and FreeSurfer (FS). Between-group differences were corrected for multiple comparisons via Family-Wise Error (FWE) and Monte Carlo z-field simulation respectively at p < 0.05 (2-tailed). RESULTS: The presence of PMN symptoms was associated with larger left inferior orbitofrontal volume (p = 0.03) and greater CTh in the left lateral orbitofrontal gyrus (p = 0.007), but reduced CTh in the left superior temporal gyrus (p = 0.009). CONCLUSIONS: The findings highlight the role of orbitofrontal and temporal cortices in the pathogenesis of negative symptoms of Schizophrenia. As they were evident in generally untreated FEP patients, the results are unlikely to be related to effects of previous treatment or illness chronicity.

The aperiodic exponent of neural activity varies with vigilance state in mice and men.

Recently the 1/f signal of human electroencephalography has attracted attention, as it could potentially reveal a quantitative measure of neural excitation and inhibition in the brain, that may be relevant in a clinical setting. The purpose of this short article is to show that the 1/f signal depends on the vigilance state of the brain in both humans and mice. Therefore, proper labelling of the EEG signal is important as improper labelling may obscure disease-related changes in the 1/f signal. We demonstrate this by comparing EEG results from a longitudinal study in a genetic mouse model for synaptic dysfunction in schizophrenia and autism spectrum disorders to results from a large European cohort study with schizophrenia and mild Alzheimer's disease patients. The comparison shows when the 1/f is corrected for vigilance state there is a difference between groups, and this effect disappears when vigilance state is not corrected for. In conclusion, more attention should be paid to the vigilance state during analysis of EEG signals regardless of the species.

Digital behavioural signatures reveal trans-diagnostic clusters of Schizophrenia and Alzheimer's disease patients.

The current neuropsychiatric nosological categories underlie pragmatic treatment choice, regulation and clinical research but does not encompass biological rationale. However, subgroups of patients suffering from schizophrenia or Alzheimer's disease have more in common than the neuropsychiatric nature of their condition, such as the expression of social dysfunction. The PRISM project presents here initial quantitative biological insights allowing the first steps toward a novel trans-diagnostic classification of psychiatric and neurological symptomatology intended to reinvigorate drug discovery in this area. In this study, we applied spectral clustering on digital behavioural endpoints derived from passive smartphone monitoring data in a subgroup of Schizophrenia and Alzheimer's disease patients, as well as age matched healthy controls, as part of the PRISM clinical study. This analysis provided an objective social functioning characterization with three differential clusters that transcended initial diagnostic classification and was shown to be linked to quantitative neurobiological parameters assessed. This emerging quantitative framework will both offer new ways to classify individuals in biologically homogenous clusters irrespective of their initial diagnosis, and also offer insights into the pathophysiological mechanisms underlying these clusters.

Clinical symptoms and psychosocial functioning in patients with schizophrenia spectrum disorders testing seropositive for anti-NMDAR antibodies: a case-control comparison with patients testing negative.

BACKGROUND: Antibodies against the N-methyl-D-aspartate receptor (NMDAR) have been described in the serum of people with schizophrenia spectrum disorders (schizophrenia). However, the prevalence and clinical relevance of these antibodies in schizophrenia is unclear. This knowledge gap includes the possibility of such antibodies being associated with a distinct clinical profile, which in turn might warrant a distinct treatment approach. We aimed to assess the seroprevalence of anti-NMDAR antibodies in schizophrenia, and compare symptoms and psychosocial functioning between patients with schizophrenia who were seropositive and seronegative for these antibodies. METHODS: In this case-control comparison, by combining new and existing studies, we included patients diagnosed with schizophrenia from four independent cohorts for whom anti-NMDAR serostatus had been assessed (or could be assessed by us) with live cell-based assays. Included cohorts were from the EULAST study (a trial conducted across 15 European countries and Israel), the OPTiMiSE study (an interventional study in Europe), and the PPiP1 and PPiP2 studies (conducted in the UK). Patients from these cohorts were analysed if they had complete Positive and Negative Syndrome Scale (PANSS) data. No participant had been diagnosed with autoimmune encephalitis or received treatment for this condition. After calculating the prevalence of serum anti-NMDAR antibodies, we examined possible differences in PANSS scores (negative, positive, and general symptom subscales, and total score) between anti-NMDAR-seropositive and anti-NMDAR-seronegative patients. Psychosocial functioning as measured by Personal Social Performance (PSP) score was also compared. All analyses were exploratory and no adjustment was done for multiple testing. People with lived experience were not involved in the conduct of this study. FINDINGS: We collected individual patient data from 1114 patients with schizophrenia across the four cohorts. The study population had a mean age of 28·6 years (SD 7·6) and comprised 382 (34·3%) women and 732 (65·7%) men, including patients of White (929 [83·4%]), Asian (54 [4·8%]), Black (68 [6·1%]), and other (62 [5·6%]) ethnicities. Overall, 41 (3·7%) participants (range 3·1-4·0% across cohorts) tested positive for serum anti-NMDAR antibodies. Lower symptom severity on the negative symptoms PANSS subscale was observed for anti-NMDAR-seropositive patients (mean score 15·8 [SD 6·4]) than for anti-NMDAR-seronegative patients (18·2 [6·8]; Cohen's d=0·36; p=0·026), as well as on the general symptoms PANSS subscale (32·9 [8·9] vs 36·1 [10·1]; d=0·33; p=0·029) and total PANSS score (65·5 [18·5] vs 72·6 [19·3]; d=0·37; p=0·013). Mean PSP score was better in anti-NMDAR-positive patients (62·0 [17·0]) than in anti-NMDAR-negative patients (53·5 [16·3]; d=0·52; p=0·014). INTERPRETATION: Serum NMDAR antibodies are present in 3-4% of patients with schizophrenia and are associated with relatively low severity of negative symptoms and relatively good psychosocial functioning. Thus, although the findings await replication in cohorts from other geographical regions, serum anti-NMDAR antibodies might be associated with a different form of psychotic illness. These findings could inform future prognostic and interventional studies examining whether anti-NMDAR antibodies are associated with a specific course of illness or with treatment response. FUNDING: None.

Revealing the impact of psychiatric comorbidities on treatment outcome in early psychosis using counterfactual model explanation.

Introduction: Psychiatric comorbidities have a significant impact on the course of illness in patients with schizophrenia spectrum disorders. To accurately predict outcomes for individual patients using computerized prognostic models, it is essential to consider these comorbidities and their influence. Methods: In our study, we utilized a multi-modal deep learning architecture to forecast symptomatic remission, focusing on a multicenter sample of patients with first-episode psychosis from the OPTiMiSE study. Additionally, we introduced a counterfactual model explanation technique to examine how scores on the Mini International Neuropsychiatric Interview (MINI) affected the likelihood of remission, both at the group level and for individual patients. Results: Our findings at the group level revealed that most comorbidities had a negative association with remission. Among them, current and recurrent depressive disorders consistently exerted the greatest negative impact on the probability of remission across patients. However, we made an interesting observation: current suicidality within the past month and substance abuse within the past 12 months were associated with an increased chance of remission in patients. We found a high degree of variability among patients at the individual level. Through hierarchical clustering analysis, we identified two subgroups: one in which comorbidities had a relatively limited effect on remission (approximately 45% of patients), and another in which comorbidities more strongly influenced remission. By incorporating comorbidities into individualized prognostic prediction models, we determined which specific comorbidities had the greatest impact on remission at both the group level and for individual patients. Discussion: These results highlight the importance of identifying and including relevant comorbidities in prediction models, providing valuable insights for improving the treatment and prognosis of patients with psychotic disorders. Furthermore, they open avenues for further research into the efficacy of treating these comorbidities to enhance overall patient outcomes.

Generalizability of the Results of Efficacy Trials in First-Episode Schizophrenia: Comparing Outcome and Study Discontinuation of Groups of Participants in the Optimization of Treatment and Management of Schizophrenia in Europe (OPTiMiSE) Trial.

Objective: In the majority of randomized controlled trials (RCTs) conducted in schizophrenia populations, patients suffering from a substance use disorder (SUD) or suicidality are excluded. Excluding these patients from RCTs might impact the generalizability of results. The aim of this study is to determine whether excluding patients with suicidality and/or SUD impacts RCT results on symptomatic remission, premature study discontinuation, symptom severity, and social functioning.Methods: Across Europe and Israel, 481 patients with first-episode schizophrenia, schizophreniform disorder, or schizoaffective disorder, based on DSM-IV criteria, were recruited between May 26, 2011, and May 15, 2016, for the Optimization of Treatment and Management of Schizophrenia in Europe (OPTiMiSE) trial. Baseline characteristics and follow-up assessments were compared between patients with versus without baseline SUD and/or suicidality.Results: A total of 446 patients met eligibility criteria for the OPTiMiSE trial and initiated amisulpride treatment, of whom 404 (91%) had data available on suicidality, SUD, duration of illness, and CDS score. Of the 360 eligible patients with baseline data on suicidality and SUD, 106 patients had comorbid suicidality and/or SUD while 254 patients had neither of these comorbidities. No significant differences in the likelihood to achieve symptomatic remission or to prematurely discontinue the study were found when comparing comorbid versus non-comorbid patients (P = .27). There were no significant differences in symptom severity and social functioning between the groups. Comorbid patients had a higher level of depressive symptoms and more impaired social functioning compared to non-comorbid patients.Discussion: Excluding first-episode schizophrenia patients with comorbidities from clinical trials unlikely affects key outcome measures. It is recommended to include patients with comorbidities in clinical trials while carefully monitoring suicidality and implementing safety plans to gain insight into efficacy and safety of treatment in this substantial patient population.Trial Registration: ClinicalTrials.gov identifier: NCT01248195.

Cannabis Use and Symptomatic Relapse in First Episode Schizophrenia: Trigger or Consequence? Data From the OPTIMISE Study.

BACKGROUND AND HYPOTHESIS: This analysis examined the relationship between cannabis use, compliance with antipsychotics and risk for relapse in patients in remission following a first episode of schizophrenia, schizophreniform, or schizoaffective disorder. STUDY DESIGN: Analyses were performed on data from a large European study on first episode of schizophrenia, schizophreniform, or schizoaffective disorder (OPTiMiSE). After 10 weeks of antipsychotic treatment, 282/446 patients (63%) met criteria for symptomatic remission; of whom 134/282 (47.5%) then completed a 1-year follow-up. Cross-lagged models and mediation models investigated the temporal relationships between cannabis use, compliance with antipsychotics, social functioning, and symptomatic worsening/relapse. STUDY RESULTS: Compared to nonusers, cannabis use increased risk for relapse, adjusted hazard ratio (HR) = 3.03 (SE = 0.32), P < .001, even in patients who were compliant with antipsychotic medication, adjusted HR = 2.89, (SE = 0.32), P < .001. Cannabis use preceded symptomatic worsening and was followed by worsening of Positive and Negative Syndrome Scale total score at the 1-year end-point (standardized ß = 0.62, SE = 0.19, P = .001) and by worsening of social functioning (coef = -0.66, P ≤ .001). CONCLUSIONS: In patients in remission from their first episode of schizophrenia, schizophreniform, or schizoaffective disorder, cannabis use increases the rate of relapse in both compliant and noncompliant individuals. Importantly, the temporal relationship between cannabis and relapse was that cannabis use preceded later relapse, noncompliance, and decrease in social functioning, and not that patients began to relapse, then used cannabis. Further research with a precision psychiatry approach might identify those patients in particular danger of relapse when using cannabis.

Extreme deviations from the normative model reveal cortical heterogeneity and associations with negative symptom severity in first-episode psychosis from the OPTiMiSE and GAP studies.

There is currently no quantifiable method to predict long-term clinical outcomes in patients presenting with a first episode of psychosis. A major barrier to developing useful markers for this is biological heterogeneity, where many different pathological mechanisms may underly the same set of symptoms in different individuals. Normative modelling has been used to quantify this heterogeneity in established psychotic disorders by identifying regions of the cortex which are thinner than expected based on a normative healthy population range. These brain atypicalities are measured at the individual level and therefore potentially useful in a clinical setting. However, it is still unclear whether alterations in individual brain structure can be detected at the time of the first psychotic episode, and whether they are associated with subsequent clinical outcomes. We applied normative modelling of cortical thickness to a sample of first-episode psychosis patients, with the aim of quantifying heterogeneity and to use any pattern of cortical atypicality to predict symptoms and response to antipsychotic medication at timepoints from baseline up to 95 weeks (median follow-ups = 4). T1-weighted brain magnetic resonance images from the GAP and OPTiMiSE samples were processed with Freesurfer V6.0.0 yielding 148 cortical thickness features. An existing normative model of cortical thickness (n = 37,126) was adapted to integrate data from each clinical site and account for effects of gender and site. Our test sample consisted of control participants (n = 149, mean age = 26, SD = 6.7) and patient data (n = 295, mean age = 26, SD = 6.7), this sample was used for estimating deviations from the normative model and subsequent statistical analysis. For each individual, the 148 cortical thickness features were mapped to centiles of the normative distribution and converted to z-scores reflecting the distance from the population mean. Individual cortical thickness metrics of +/- 2.6 standard deviations from the mean were considered extreme deviations from the norm. We found that no more than 6.4% of psychosis patients had extreme deviations in a single brain region (regional overlap) demonstrating a high degree of heterogeneity. Mann-Whitney U tests were run on z-scores for each region and significantly lower z-scores were observed in FEP patients in the frontal, temporal, parietal and occipital lobes. Finally, linear mixed-effects modelling showed that negative deviations in cortical thickness in parietal and temporal regions at baseline are related to more severe negative symptoms over the medium-term. This study shows that even at the early stage of symptom onset normative modelling provides a framework to identify individualised cortical markers which can be used for early personalised intervention and stratification.

Association of Complement and Coagulation Pathway Proteins With Treatment Response in First-Episode Psychosis: A Longitudinal Analysis of the OPTiMiSE Clinical Trial.

BACKGROUND AND HYPOTHESIS: Treatment response to specific antipsychotic medications is difficult to predict on clinical grounds alone. The current study hypothesizes that the baseline complement pathway activity predicts the treatment response and investigates the relationship between baseline plasma biomarkers with treatment response to antipsychotic medications. STUDY DESIGN: Baseline plasma samples were collected from first episode of psychosis patients (n = 243) from a multi-center clinical trial. The participants were treated with amisulpride for 4 weeks. Levels of complement and coagulation proteins at baseline were measured using both data-dependent and data-independent mass spectrometry approaches. The primary outcome was remission status at 4 weeks and the secondary outcomes included change in psychotic and functional symptoms over the period of treatment. In addition, immunoassays were performed at baseline for complement C1R, as well as for activation markers C4a and sC5b-9. STUDY RESULTS: The plasma level of complement variant C4A was significantly associated with remission at 4 weeks. Moreover, higher levels of several complement and coagulation pathway proteins were associated with a reduction in psychotic symptoms and an improvement in functioning. Immunoassays showed an association of baseline levels of C1R and C4a as well as complement activation marker sC5b-9 levels with treatment response. CONCLUSION: The results demonstrated that the response to antipsychotic treatment might be related to pre-treatment levels of plasma complement and coagulation pathway proteins. This is consistent with independent evidence associating immune dysfunction with the pathophysiology of psychosis. Moreover, these results inform the development of novel therapeutic approaches that target the complement system for psychosis.