Psychiatry and Clinical Neurosciences

Brain-imaging models predicting psychiatric outcomes are rarely tested where they would be used. Clinical high risk for psychosis is a demanding test case because services differ in whom they recruit and how they record transition. We re-estimated a published structural magnetic resonance imaging (MRI) risk model and applied it unchanged to four independent cohorts (685 participants, 106 transitions). Discrimination was weak (area under the curve, 0.56) and near chance in the two cohorts not previously used to evaluate the model (0.53 and 0.56). Predicted risks did not transfer (calibration slope, 0.17). Exploratory performance landscapes, which map local discrimination across participant characteristics, showed that discrimination varied with negative-symptom severity, also in external cohorts, whereas apparent age- and brain-age-related variation depended on single cohorts; no global test survived multiple-testing correction. Reporting where a model works, not only its average accuracy, can reveal when one diagnostic label hides several prediction problems.

Purpose Delirium affects around 40% of intensive care unit (ICU) patients and adversely affects survival, length of stay, and cognitive recovery. Despite this, implementation of systematic clinical delirium assessments is variable. We sought to validate a point-of-care electroencephalography (EEG)-based system (Ceribell, Inc.) for the detection of delirium in the ICU. Methods In this multicenter, prospective cohort study, we monitored adult ICU patients with the Ceribell EEG system for 6-8 hours for a minimum of three days. Up to three times per day, expert clinicians assessed participants for delirium and research coordinators completed Confusion Assessment Method for the ICU (CAM-ICU) assessments, which we used to calculate CAM-ICU-7 severity scores. We compared expert clinician diagnoses with Ceribell EEG algorithm output using data from 30-minute periods surrounding each clinician assessment. Results Among 225 participants analyzed (median age: 64 years, 60% male, 21% mechanically ventilated), expert clinicians performed 857 assessments and found 24% of participants were delirious during at least one assessment. The Ceribell EEG system detected delirium with a sensitivity of 81% (95% CI: 76% to 87%) and specificity of 81% (95% CI: 78% to 84%). The continuous Ceribell EEG system score correlated with CAM-ICU-7 delirium severity scores (bootstrapped median rho 0.58, IQR: 0.54-0.62, p <0.001). Conclusion In a mixed adult ICU population, the Ceribell EEG system identified delirium with good sensitivity and specificity. Its continuous output was associated with delirium severity, highlighting its potential ability to characterize underlying physiological patterns across the clinical delirium spectrum.

Background: Serial glial fibrillary acidic protein (GFAP) trajectories have become an important framework for contextualizing evolving secondary-injury pathophysiology after moderate-to-severe traumatic brain injury (msTBI). However, total GFAP pools release and clearance signals that may be less useful for longitudinal bedside decisions than a proteoform-resolved assay. We compared total GFAP with neoGFAP, defined here as calpain-generated GFAP proteoforms intended to index active astroglial proteolysis during the subacute phase. Methods: We analyzed 651 serial serum samples from 95 msTBI patients from a previously described single-site cohort. Total GFAP and neoGFAP were measured on the same MSD platform from 6 to 240 hours after injury. Early (6 to 72 h) and late (96 to 240 h) windows, data-derived tertiles, and serial trajectory summaries were calculated directly from serial samples. Models were benchmarked against age plus admission post-resuscitation Glasgow Coma Scale (GCS) and the admission IMPACT extended risk score using five-fold stratified cross-validation. Outcomes were unfavorable outcome (GOSE 1 to 4), less-than-good recovery (GOSE 1 to 6), Disability Rating Scale (DRS) [&ge;]15, mortality, and neuroimaging worsening at 6 months. Results: The cohort contributed 95 serial biomarker profiles, with 90 participants evaluable for 6-month GOSE and 89 for DRS. Unfavorable outcome occurred in 57/90 (63.3%), and less-than-good recovery in 79/90 (87.8%). For unfavorable outcome, IMPACT plus early neoGFAP reached AUROC 0.85 versus 0.84 for IMPACT plus early total GFAP and 0.81 for IMPACT alone. For less-than-good recovery, IMPACT plus late neoGFAP achieved AUROC 0.90 versus 0.84 for late total GFAP and 0.82 for IMPACT alone. Secondary analyses for DRS, mortality, and neuroimaging worsening showed smaller differences. Conclusions: In this retrospective analysis, neoGFAP provided clearer incremental value than total GFAP for recovery-oriented monitoring, especially when late-window reassessment of patients who remained at risk for less-than-good recovery was required. Results support prospective testing of neoGFAP as a pathophysiology-informed adjunct to serial bedside decision making, repeat-assessment thresholds, and recovery stratification.

Background: Post-COVID Condition (PCC) affected 10-30% of individuals after SARS-CoV-2 infection, and pre-infection predictors of who goes on to develop it are not well established. We compared pre-infection structural brain variation and baseline psychiatric phenotype as candidate predictors of PCC within a common multi-modal framework in the German National Cohort (NAKO). Both are candidate "first hits" under the second-hit hypothesis, which motivates the comparison; the vulnerability-by-infection interaction that hypothesis turns on is not identifiable among infected participants, so this is a prediction study rather than a test of the hypothesis. Methods: In 8,464 SARS-CoV-2-infected NAKO neuroimaging participants, of whom 2,304 (27.2%) met PCC criteria (weighted post-COVID syndrome (PCS) score > 10.75) and the rest were symptom-free controls, ten pre-infection modalities (T1-weighted volumetric brain MRI across six parcellations, baseline mental health (PHQ-9, GAD-7), demographics, and seven further biomedical and socioeconomic domains) entered a stacked ensemble with 10x5 nested cross-validation (primary metric: the area under the precision-recall curve, PR-AUC). Pre-specified analyses assessed transportability to a non-imaging cohort and robustness to symptom trajectory and unmeasured confounding. SARS-CoV-2 infection status and the PCC outcome were ascertained by self-report; no clinically confirmed diagnoses were available. Results: Baseline mental health, assessed 3-8 years before infection, was the strongest predictor (38.2% of feature importance; standalone ROC-AUC = 0.640), whereas all six brain-MRI parcellations performed at or near chance (ROC-AUC 0.496-0.516). The multi-modal model achieved moderate, well-calibrated discrimination (ROC-AUC = 0.664, 95% CI 0.652-0.677; PR-AUC = 0.413, 0.392-0.434; ECE = 0.015). Applied without retraining to the non-imaging cohort it retained discrimination (ROC-AUC = 0.660, 0.654-0.665), meeting two of three pre-specified equivalence criteria. Adjusting for the full symptom trajectory shrank the baseline mental-health odds ratio from 2.06 (1.83-2.32) to a conservative lower bound of 1.38 (1.20-1.58; E-value 2.66) while leaving it independently significant, and was essentially unchanged under an alternative control definition (2.07). Baseline mental health did not predict objectively measured hyposmia in participants screened before their infection (0.89) while predicting self-reported smell loss in the same participants (1.92). Conclusions: Baseline mental health years before infection is the strongest pre-infection predictor of PCC among the infected, and the association held across the pre-specified sensitivity analyses. Whether it acts specifically on COVID-19 sequelae is a separate question this design cannot answer, and the indirect evidence points away from specificity: the association is undiminished after mild infection but absent among the hospitalised, and baseline mental health predicts current symptom load no more strongly in infected than in non-infected participants. The volumetric structural candidate is not supported: pre-pandemic T1-weighted volumetry carried no predictive signal in this single neuroimaging cohort, consistent with COVID-19-associated brain changes being acute-onset rather than pre-existing. Risk stratification may benefit from incorporating baseline psychiatric phenotype; whether treating it reduces PCC incidence requires interventional study.

Background: Alcohol use disorder (AUD) is associated with increased dementia risk, but its long-term relationship with dementia, variation by age and subtype, and how this risk compares with that associated with alcohol consumption remain poorly characterised. We investigated these questions in a large prospective cohort of UK women. Methods: We analysed Million Women Study participants recruited aged 50-64 years and followed for more than 20 years through linked health records. Women with hospital-recorded AUD were matched to up to five controls. Cox models estimated associations between time-varying AUD and incident dementia overall, by subtype and attained age, with exposure lags of up to 15 years and competing-risk analyses. Self-reported alcohol consumption was analysed separately. Findings: Among 1,288,359 women, 20,070 with AUD were matched to 100,339 controls. Dementia occurred in 3,774 women with AUD and 8,239 matched controls. AUD was associated with all-cause dementia (HR=3.73 [95% CI 3.52-3.95]), with strongest associations for mixed (5.03 [3.88-6.51]), unspecified (4.29 [3.92-4.70]), and vascular dementia (3.96 [3.44-4.56]). Associations were substantially stronger before age 70 years (10.68 [9.02-12.63]) than at ages 70-79 years (3.75 [3.47-4.04]) or 80 years or older (2.34 [2.11-2.60]). Risk remained elevated with a 15-year lag (2.16 [1.81-2.58]) and after accounting for competing mortality. Associations with self-reported alcohol consumption were substantially weaker. Interpretation: Clinically recognised AUD identifies women at markedly increased and prolonged dementia risk, particularly at younger ages and for vascular and mixed dementia, not captured by drinking quantity alone. People with AUD should be considered a priority population for earlier dementia prevention and risk reduction. Funding: Wellcome Trust Career Development Award to AT (306069/Z/23/Z); NIH grants P50AA012870 and R01DA054869 (JG). The Million Women Study is supported by the UK Medical Research Council (grant number UKRI1464).

Severe worry is a transdiagnostic syndrome associated with significant morbidity in older adults. In this study, we aim to infer worry-related mental states though brain activity timeseries. We acquired fMRI on two cohorts (N=116 and N=88), using an in-scanner worry induction and reappraisal task. We trained a recurrent long short-term memory (LSTM) neural network, using the first cohort as the train/validation and the second cohort as an independent test set. We predicted worry induction, reappraisal, and neutral states (area under the curve 0.89, 0.77, 0.91 for the test set and 0.78, 0.63, 0.81 for the independent set). The model was most accurate when participants reported high worry during the induction state. Dorsal attention network, and networks seeded on the anterior hippocampus, and supplementary motor area were most important for predicting worry states. The LSTM approach may have critical translational implications for identifying and treating severe worry in older adults.

Background and Hypothesis: First-line antipsychotics are ineffective in ~40% of people with first episode psychosis (FEP). Electroencephalogram (EEG)-based predictive biomarkers could be a scalable solution to reduce trial-and-error. Study Design: We acquired EEG in 56 minimally treated FEP individuals before treatment initiation, and 34 matched healthy controls (HC). Treatment response was defined as >50% improvement in BPRS scores by 12 weeks of treatment. Modified k-means clustering of resting-state EEG yielded six microstates including canonical topographies (A-D). The first principal component (PC1) of microstate properties derived from HCs served as a normative reference onto which FEP patients were projected. Baseline PC1 was used to predict treatment response in adjusted regression analyses. To test stability of results, an independent dataset of HCs (n=92) was fitted using the same microstate solution and compared to our native HC dataset. PC1 scores derived from this independent normative dataset were also used for predicting response in FEP. Study Results: Individual microstates replicated established patient-control differences (e.g., higher presence of microstate C compared to D, p<0.001). PC1 explained 34.8% of variance. Baseline PC1 predicted subsequent treatment response (adjusted OR=0.63, p<0.001), yielding AUC=72.3% in leave-one-out cross-validation (pperm<0.001) when combined with clinical information. PC1 from our HC transported well to the second healthy control EEG dataset (n=92) (r=0.9,p<0.001; r=0.89,p<0.001 ). PC1 in FEP derived using PC1 weights from a second normative dataset also predicted response in FEP (OR=0.70, p=0.03). Conclusions: EEG microstate architecture differing from the expected norm is associated with lower odds of treatment response in FEP.

OBJECTIVE: Sepsis-associated encephalopathy (SAE) and sepsis-associated delirium (SAD) lack confirmatory laboratory tests or biomarkers. We evaluated whether the Th17/interleukin (IL)-17 axis or neuroglial-injury biomarkers reflect acute brain dysfunction beyond systemic inflammation and whether biomarker profiles differ between these phenotypes. DATA SOURCES: EMBASE, MEDLINE/PubMed, and the Cochrane Library from inception to August 1, 2026, supplemented by backward and forward citation searching, following PRISMA 2020. STUDY SELECTION: Observational studies of adults with sepsis classified by SAE or SAD status and reporting circulating biomarker concentrations by neurological group or biomarker-outcome associations. Eighteen studies (2,388 participants; 1,863 with sepsis) were included. DATA EXTRACTION: Two reviewers independently extracted study, phenotype, assay, and outcome data. Risk of bias was assessed with the Newcastle-Ottawa Scale and certainty with GRADE. DATA SYNTHESIS: Random-effects models pooled standardized mean differences (SMDs) separately against non-septic and septic controls and odds ratios (ORs) for biomarker-outcome associations. IL-6 was the only mediator elevated in both comparative frameworks and associated with outcomes (non-septic controls: SMD, 0.54 [0.25-0.82]; septic controls: SMD, 1.05 [0.51-1.59]; OR, 1.85 [1.50-2.29]). S100B, neuron-specific enolase, IL-10, tumor necrosis factor; and interferon-gamma; were elevated versus septic controls, whereas IL-17 was modestly elevated only versus non-septic controls (SMD, 0.37 [0.01-0.73]). The IL-6 septic-control effect was concentrated in SAE (p<0.0001) and remained robust in leave-one-out and risk-of-bias sensitivity analyses. Certainty was low to very low, and all 14 computable prediction intervals included the null. CONCLUSIONS: IL-6 emerges as the leading circulating candidate biomarker of sepsis-associated brain dysfunction, while S100B and NSE capture neuroglial injury beyond sepsis. Current evidence does not establish a discriminating role for the Th17 axis. Future studies require standardized neurological phenotyping and appropriately matched septic controls to establish diagnostic value.

Machine learning approaches may support individual-level classification in psychiatry, but many EEG-based schizophrenia studies have relied on small samples or conventional summary features. We evaluated whether trial-level auditory oddball event-related potential (ERP) waveforms could support schizophrenia versus healthy-control classification using deep learning. The study included 258 patients with schizophrenia and 142 healthy controls. EEG recordings from an auditory duration oddball paradigm were segmented into -100 to 500 ms epochs, and trial-level mismatch waveforms were generated by subtracting each participant's mean standard response from accepted deviant trials. Models were trained using a fixed participant-level training, validation, and test split, with demographic residualization fit only in the training set. Five deep learning architectures were trained on full residualized ERP waveforms and compared with classical machine learning models trained on 18 conventional ERP summary features. Deep learning models achieved higher test set discrimination than classical feature-based models, with AUROC values ranging from 0.797 to 0.857 versus 0.705 to 0.720. Benchmark analyses suggested that performance depended on the combination of waveform-level input and deep learning architecture. These findings support trial-level auditory oddball ERP waveforms as promising classification inputs and candidate electrophysiological biomarkers of schizophrenia-related neural information processing.

Neighbourhood deprivation is one of the few potential policy-modifiable risk factors for psychiatric and neurological disorders, but the neurobiological pathways underlying these associations remain unclear. We investigated these relationships across three cohorts spanning the life span: the Healthy Brain and Child Development (HBCD) Study (n = 84, aged 0 to 4 weeks postnatal), the Adolescent Brain Cognitive Development (ABCD) Study (n = 4,792, aged 9 to 10 years), and the UK Biobank (UKB; approximately 500,000 adults, aged 44 to 87 years). Neighbourhood deprivation was associated with elevated disease risk, and individual lifestyle factors accounted for only a small fraction of this burden, indicating that the much larger residual effect reflects broader contextual characteristics of deprived environments rather than individual behaviours alone. Across all cohorts, greater deprivation consistently predicted lower cortical and subcortical brain volume, with effects detectable in early development and substantially stronger in adulthood. Across disorders, regional brain volume emerged as a consistent neuroanatomical mediator linking neighbourhood deprivation to neuropsychiatric disease. We further showed that deprivation preferentially affects brain regions intrinsically vulnerable to neuropsychiatric disorders. Spatial decoding analyses implicated dopaminergic and serotonergic neurotransmitter systems together with specific excitatory and inhibitory neuronal classes. Importantly, both the deprivation effects and their neuroanatomical mediation patterns were replicated across independent populations. Our study delivers a translational framework linking neighbourhood deprivation to brain health, which could inform public health policies and preventive interventions.

Early detection of individuals at clinical high risk for psychosis (CHR-P) improves access to preventive care. We developed and internally validated the E-DetectioN Tool for Emerging mental disoRders (ENTER), a multimodal digital screener designed to improve community identification of CHR-P compared with an established approach (PQ-16 cut-off). Participants aged 12-35 years from the UK and Italy completed the online ENTER screener, assessing attenuated psychotic symptoms, processing speed and environmental risk factors. Participants above the PQ-16 cut-off and a random subset below were assessed with the Comprehensive Assessment of At Risk Mental States (CAARMS). PQ-16 cut-off and ENTER models were developed using (penalised) logistic regression, random forest and extreme gradient boosting (XGBoost) within a repeated nested cross-validation framework with semi-supervised learning incorporating unlabelled data. Models were evaluated by discrimination, calibration and clinical utility. 2,522 participants completed the ENTER screener. 495 underwent CAARMS assessments. 124 (25.1%) met CHR-P criteria; 2,027 participants were not assessed with the CAARMS (unlabelled). ENTER showed improved discrimination (C=0.744, 95%CI: 0.697-0.790) compared to the PQ-16 cut-off (C=0.625, 95%CI: 0.570-0.682, p=0.003) and similar calibration (ENTER: slope=1.02, 95%CI: 0.78-1.32; PQ-16 cut-off: slope=0.39, 95%CI: 0.19-0.52, p=0.074). At a clinically relevant threshold, ENTER identified approximately 2 additional CHR-P individuals per 100 assessments compared with the PQ-16 cut-off. Compared with established PQ-16 cut-off pre-screening, ENTER showed higher performance for community identification of CHR-P individuals. Multimodal digital screening might support more informative community triage before specialist assessment. External validation and implementation studies are required to establish real-world clinical utility.

**Importance** Average gray matter volume (GMV) loss in early psychosis may conceal clinically relevant differences in brain change. **Objective** To identify longitudinal GMV-change patterns, their cross-site correspondence, and clinical associations. **Design, Setting, and Participants** Prospective cohort with T1-weighted MRI acquired during 2011-2026 at 2 Czech sites; median follow-up, approximately 13 months. Participants were 222 patients with first-episode schizophrenia-spectrum psychosis and 178 healthy controls frequency-matched by site and baseline age. **Exposure** First-episode psychosis vs healthy control status. **Main Outcomes and Measures** Diagnosis-blind clustering by site of image-quality-corrected, subsequently annualized GMV change across 333 cortical parcels; longitudinal voxel-based morphometry (VBM); Global Assessment of Functioning and consensus 5-factor Positive and Negative Syndrome Scale negative-symptom scores. **Results** Among 400 participants (mean [SD] age, 28.2 [7.5] years), 208 (52.0%) were male. Each site independently yielded 3 clusters: C1 (near-stable GMV), C2 (pronounced decrease), and C3 (increase). Cross-site centroid correlations across 333 parcels were strong for C2 and C3 (Spearman rho = 0.865 and 0.861); C1 correlation was weak (rho = 0.138), as expected for near-zero profiles with little regional variation. C2/C3 correspondence supported pooling matched site-specific clusters for post hoc analyses. Pooled clusters contained 77 patients and 125 controls in C1, 66 patients and 16 controls in C2, and 79 patients and 37 controls in C3. Suffixes -P and -HC identify patient and control subgroups. To situate patient changes relative to controls, post hoc VBM revealed an unanticipated gradient of increasingly pronounced GMV reduction across C1-HC to C1-P to C2-P, prominently involving association cortex, and bilateral ventral temporal GMV increase in C3-P (familywise error-corrected P < .05). The opposite contrast along this ordering showed periventricular changes consistent with progressively greater ventricular enlargement. C2-P had the lowest mean follow-up functioning and highest mean negative-symptom severity (adjusted omnibus P = .005 and .02); clinical changes from baseline did not differ significantly among patient groups. **Conclusions and Relevance** Early GMV change in psychosis was not uniform but comprised two contrasting directions: a clinically relevant, control-anchored pattern of increasingly pronounced GMV loss involving default mode regions and a distinct, directionally opposing ventral temporal increase.

Background and Hypothesis: Although psychotic disorders are characterized by widespread cortical thinning, translating these findings into individualized biomarkers remains challenging. Regional Vulnerability Index (RVI) and Neuroscore are candidate individual-level biomarkers that quantify the similarity between an individual's structural brain pattern and generic psychosis-associated alterations. This study investigated cortical thickness (CT)-based RVI and Neuroscore, along with mean CT deviation from healthy individuals, across the psychosis spectrum. Study Design: Using structural 3T MRI data from 289 participants (clinical high risk for psychosis [CHR], n=39; first episode psychosis [FEP], n=44, chronic psychosis [CP], n=86; healthy controls [HC], n=120), RVI, Neuroscore, and mean CT deviation (z) scores were computed from region-wise CT measures quantified using FreeSurfer v7.2. Group differences and associations with symptoms were evaluated. Study Results: RVI and Neuroscore were significantly higher, and mean CT z-scores lower, in FEP and CP compared with CHR and HC, with moderate-to-large effect sizes (|Cohen's d| = 0.55-1.15). No differences were observed between FEP and CP or CHR and HC. Indices were not associated with psychosis-spectrum symptom severity in the CHR, FEP, or CP groups. RVI (r = -0.561, p < .001) and Neuroscore (r = -0.997, p < .001) were strongly negatively correlated with the mean CT z-score. Conclusions: CT-based RVI and Neuroscore appear to index established disease status but not risk status, while showing substantial overlap with mean CT deviation. These findings highlight the need to further refine these biomarkers through data-driven and multimodal approaches to improve their utility for risk and prognostic prediction.

Genetic studies of serious mental illness (SMI, including schizophrenia and bipolar disorder) have implicated two classes of risk variant: common variants1-3, each of small effect but collectively accounting for most population-level risk, and ultra-rare coding variants4,5 which confer large risk in a small number of individuals. Most such studies cannot directly assess the impact of specific variants on SMI, transdiagnostically, as the phenotypic information that they collect is focused on a single diagnosis. Here we report results of association analysis of predicted deleterious variants from exome sequencing of 21,958 cases (SMI and severe major depressive disorder) and 19,826 controls from the Mision Origen biobank in the Paisa genetic isolate of Colombia. We identified transdiagnostic associations to seven genes, three at an exome-wide and four at a false discovery rate significance threshold. Five of these genes have not previously been implicated in SMI (ADAP1, WIPI1, NFE2L1, FBXO10, and HLA-DPB1 ); for the first four of these the association is driven by variants enriched in the Paisa due to a founder effect, filling the gap between ultra-rare and common variants. The strongest association that we observe is to a frameshift variant (7:904169:T:TC) in ADAP1 that is genome-wide significant for SMI, schizophrenia, and bipolar disorder and increases risk for both diagnoses by more than threefold; it truncates ADAP1, a brain-enriched Arf6 GTPase-activating/scaffolding protein that has been mechanistically linked to post-mitotic neuronal morphogenesis and regulation of dendritic differentiation6. Variant carriers differ significantly from other SMI cases in Mision Origen for psychotic, manic, and cognitive item-level phenotypes assessed through analyses of electronic health records7-9 and displayed impaired executive function on neurocognitive testing. Variant carriers differ significantly from other SMI cases in Mision Origen for multiple item-level phenotypes (psychotic, manic, and cognitive) assigned through natural language processing of longitudinal electronic health records. Although the 7:904169:T:TC variant is not significantly associated with major depressive disorder, carriers compared to others with this diagnosis display a significant increase in symptoms typical of bipolar disorder. The variant's high allele frequency in Mision Origen (~0.4%) will enable population-level studies to stratify SMI based on causation and identify additional factors influencing risk and resilience.

Genetic studies of post-traumatic stress disorder (PTSD) implicate immune regulation, yet linking peripheral immune-related proteins to brain molecular regulation and experimentally tractable targets remains challenging. Cellular and spatial genetic analyses were used to characterize the cellular and spatial genetic context of these traits. We then conducted bidirectional Mendelian randomization (MR) across 30 predefined Ig-like plasma proteins and 16 behavioral and neuropsychiatric traits. The forward MR screen identified four associations meeting the prespecified screening criteria, all of which underwent the same post-screening robustness assessment. CD33-PTSD showed the most consistent support across instrument definitions and applicable robust estimators and was therefore advanced as the prioritized candidate to regional genetic, brain molecular QTL, and functional analyses. Standard CD33 pQTL-PTSD colocalization did not support a simple shared single-variant model, while multi-signal analyses remained scientifically unresolved under the available LD reference. Meanwhile, plasma CD33 pQTL and cortical CD33 eQTL showed strong regional sharing (PP4=0.96468). Regional localization, cross-analysis candidate selection, and functional annotation further suggested a complex regulatory context across the CD33-SIGLEC region. Among 284 compounds with prior CD33 binding evidence that entered structure-assisted screening, 11 were retained as candidates for experimental follow-up. Overall, CD33-PTSD was the prioritized association identified by our screening; subsequent regional genetic, brain molecular QTL, and functional annotation analyses further delineated the complex regulatory context of the CD33- SIGLEC locus, while the CD33-centered structure-assisted exploration yielded 11 compound candidates for subsequent experimental validation.

Genomic studies have improved our understanding of posttraumatic stress disorder (PTSD) biology. Epigenetic differences in DNA methylation can reflect environmental influences, which are critical in PTSD etiology, and may additionally differentiate between PTSD cases and controls or distinguish severity of PTSD symptoms. We conducted PTSD epigenome-wide association studies (EWAS) in a large cohort, n = 22,141 subjects from the United States V.A. Million Veteran Program (MVP), with the PTSD Checklist (PCL-17), including 17,674 European ancestry (EUR), 3,430 African ancestry (AFR), and 1,037 admixed American descent (AMR). We evaluated the 17-item PCL-Total as well as symptom subdomains (re-experiencing, avoidance, and hyperarousal), and used lasso regression on electronic health records to define lifetime PTSD diagnoses. There were 9,413 total cases of Lifetime PTSD and 27,257 controls, with 5,184 cases and 19,382 controls in EUR, 3,095 cases and 6,240 controls in AFR, and 1,134 cases and 1,635 controls in AMR. We identified a total of 241 epigenome-wide-significant associated CpG sites across all PTSD traits. We replicated 5 out of 11 CpG sites reported in a prior (Psychiatric Genomics Consortium) PTSD EWAS. Overlap with differentially methylated CpG sites in subregions of the amygdala and hippocampus of human postmortem brains of individuals with PTSD vs. controls was evaluated at the CpG and gene levels. Ten genes of the 195 identified in the brain study overlapped with the 192 identified in our EWAS. There was significant enrichment of genes downregulated in somatostatin interneurons and brain endothelial cells. Regression analyses showed that PTSD case status was significantly associated with accelerated DNA methylation aging. Contrary to previous PTSD EWAS findings, smoking status stratification suggested that the association between PTSD and methylation of the AHRR gene may be confounded by smoking status. Our findings showcase the largest-scale PTSD EWAS in multiple ancestries and the first large PTSD EWAS on symptom severity. We replicated several findings from previous EWAS and greatly extended current knowledge of the relationship between epigenetics and PTSD.

Objective: To compare appointment completion and offered-slot utilization between virtual and in-person outpatient psychiatric care in Peru. Methods: We conducted a retrospective ecological longitudinal study using routinely collected monthly aggregate administrative indicators from January 2022 through June 2026, with modality-month as the unit. Virtual-minus-in-person differences and 95% confidence intervals used Newey-West standard errors with 12 lags. June 2026 was partial; slot utilization was restricted to 2024-2026. Results: Across 54 paired months, mean completion was 90.9% (SD=3.9) for virtual and 83.8% (SD=2.7) for in-person care; difference +7.2 points (95% CI=4.4-9.9). Across 30 paired months, slot utilization was 73.0% (SD=8.3) and 90.5% (SD=3.8), respectively; difference -17.5 points (95% CI=-24.5 to -10.4). Sensitivity analyses were similar. Conclusions: Virtual care had higher completion but lower slot utilization. Attendance and capacity use should be evaluated separately when planning hybrid psychiatric services. Outpatient psychiatric services increasingly operate as hybrid systems in which virtual and in-person visits coexist.

The potential role of the cerebellum in the pathophysiology of schizophrenia (SCZ) has received relatively insufficient attention. Prior studies characterized cerebellar abnormalities mainly at the group level, without fully considering the heterogeneity of the disease. Here, we applied a machine learning approach (Subtype and Stage Inference, SuStaIn) to cross-sectional 3D volumetric MRIs of the cerebellum, including 1,588 individuals with SCZ (638 females; mean age: 31.9 +/- 12.0 years) and 2,341 healthy controls (HC) (1,041 females; mean age: 33.3 +/- 13.6 years), from 17 sites worldwide. SuStaIn identified three distinct spatiotemporal trajectories in cerebellar gray matter volume (GMV) reduction, respectively originating in lobule X (subtype 1), lobule III (subtype 2) and lobule VIIb (subtype 3), with subtypes 1 and 3 corresponding to the posterior lobe and subtype 2 to the anterior lobe. These cross-sectionally inferred trajectories were replicated in two independent samples of 1,334 and 530 patients, respectively. Multimodal analyses using neuroimaging, transcriptomic and behavioral data revealed subtype-specific biological characteristics in brain morphological patterns, cerebellar-cortical connectivity, gene expression and clinical symptoms. Specifically, subtype-related genes were enriched in metabolism-related processes and immunity-related processes, respectively, in subtype 1 and subtype 2; subtype 3 showed more severe brain abnormalities and worse cognitive symptoms. Treatment data from 381 patients, with up to 12 months of follow-up, revealed poorer response to antipsychotic medications (APM) in subtype 2 and subtype 3, but better response to transcranial magnetic stimulation (TMS) in subtype 1, which also had worse emotion-related symptoms. Together, our findings offer a comprehensive characterization of the heterogeneity of cerebellar pathophysiological progresses in SCZ, which may help in developing future clinical stratification and targeted intervention strategies.

Background: Psychiatric disorders pose a significant public health burden, and their complex genetic architecture underlies multiple biological pathways, contributing to a lack of diagnostic biomarkers. Laboratory-Wide Association Studies (LabWAS) leverage electronic health record laboratory data as intermediary phenotypes to link genetic risk to physiology. Methods: Polygenic Scores (PGS) for 25 health-related, psychiatric, and substance use phenotypes were generated for individuals of African (AFR; N=10,473) and European (EUR, N=35,131) ancestry from the Penn Medicine Biobank. Laboratory data was prepared using the LabWAS pipeline, encompassing 78 laboratory tests and 6 vital signs. Linear regression models tested the association of each lab value with PGS adjusting for sex, age, and ten within-ancestry genetic principal components. Results: Four PGS were significantly associated with at least one laboratory value in AFR individuals and 22 PGS in EUR individuals. In EUR individuals, psychiatric PGS partitioned into three physiological profiles: anxiety, attention-deficit/hyperactivity disorder, major depressive disorder, and post-traumatic stress disorder shared a metabolic/inflammatory signature (higher BMI, glucose, and monocytes; lower high density lipoprotein cholesterol and oxygen saturation); anorexia nervosa, obsessive compulsive disorder, and schizophrenia were characterized by lower BMI; and bipolar disorder demonstrated a more circumscribed profile with lower electrolytes and renal makers. Substance use disorder PGS similarly clustered: tobacco use disorder and externalizing showed metabolic/inflammatory profiles; alcohol-related phenotypes showed lower red blood cell indices; and cannabis use disorder had a more circumscribed profile. Conclusions: LabWAS reveals physiologic signatures associated with genetic risk for psychiatric disorders, offering new biological insights into their pathology.

Psychosis has long been conceptualized as a disorder of disrupted hierarchical integration across distributed brain systems, yet it remains unclear whether alterations in macroscale cortical hierarchy are already present before illness onset and are associated with subsequent transition to psychosis. Using connectome gradient mapping, we characterized baseline cortical hierarchical architecture along the unimodal-to-transmodal axis in 580 participants from the NAPLS-3 cohort, including converters (CHR-C, n = 56), non-converters (CHR-NC, n = 434), and healthy controls (HC, n = 90). Group differences were assessed at regional, network, and global levels. Group comparisons revealed that CHR-C individuals, relative to the other two groups, exhibited bidirectional alterations selectively along the sensorimotor-to-association gradient, with reduced values in the visual network alongside elevated values in the default mode network, indicating greater separation between sensory and transmodal systems along the gradient. At the global level, CHR-C showed increased explained variance, range, and variation of this gradient, collectively indicating hierarchical expansion. Notably, greater explained variance of this gradient was associated with a shorter time to conversion to psychosis, while increased gradient range and variation were associated with higher positive symptom severity across CHR individuals. These findings indicate that expansion of the sensorimotor-to-association connectome hierarchy is already present before psychosis onset in individuals who subsequently convert to psychosis. This altered hierarchical organization may reflect greater decoupling between sensory and transmodal systems and may characterize neurobiological changes associated with progression from a clinical high-risk state to psychotic illness.