Immune-based Machine learning Prediction of Diagnosis and Illness State in Schizophrenia and Bipolar Disorder.

BACKGROUND: Schizophrenia and bipolar disorder frequently face significant delay in diagnosis, leading to being missed or misdiagnosed in early stages. Both disorders have also been associated with trait and state immune abnormalities. Recent machine learning-based studies have shown encouraging results using diagnostic biomarkers in predictive models, but few have focused on immune-based markers. Our main objective was to develop supervised machine learning models to predict diagnosis and illness state in schizophrenia and bipolar disorder using only a panel of peripheral kynurenine metabolites and cytokines. METHODS: The cross-sectional I-GIVE cohort included hospitalized acute bipolar patients (n = 205), stable bipolar outpatients (n = 116), hospitalized acute schizophrenia patients (n = 111), stable schizophrenia outpatients (n = 75) and healthy controls (n = 185). Serum kynurenine metabolites, namely tryptophan (TRP), kynurenine (KYN), kynurenic acid (KA), quinaldic acid (QUINA), xanthurenic acid (XA), quinolinic acid (QUINO) and picolinic acid (PICO) were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS), while V-plex Human Cytokine Assays were used to measure cytokines (interleukin-6 (IL-6), IL-8, IL-17, IL-12/IL23-P40, tumor necrosis factor-alpha (TNF-ɑ), interferon-gamma (IFN-γ)). Supervised machine learning models were performed using JMP Pro 17.0.0. We compared a primary analysis using nested cross-validation to a split set as sensitivity analysis. Post-hoc, we re-ran the models using only the significant features to obtain the key markers. RESULTS: The models yielded a good Area Under the Curve (AUC) (0.804, Positive Prediction Value (PPV) = 86.95; Negative Prediction Value (NPV) = 54.61) for distinguishing all patients from controls. This implies that a positive test is highly accurate in identifying the patients, but a negative test is inconclusive. Both schizophrenia patients and bipolar patients could each be separated from controls with a good accuracy (SCZ AUC 0.824; BD AUC 0.802). Overall, increased levels of IL-6, TNF-ɑ and PICO and decreased levels of IFN-γ and QUINO were predictive for an individual being classified as a patient. Classification of acute versus stable patients reached a fair AUC of 0.713. The differentiation between schizophrenia and bipolar disorder yielded a poor AUC of 0.627. CONCLUSIONS: This study highlights the potential of using immune-based measures to build predictive classification models in schizophrenia and bipolar disorder, with IL-6, TNF-ɑ, IFN-γ, QUINO and PICO as key candidates. While machine learning models successfully distinguished schizophrenia and bipolar disorder from controls, the challenges in differentiating schizophrenic from bipolar patients likely reflect shared immunological pathways by the both disorders and confounding by a larger state-specific effect. Larger multi-centric studies and multi-domain models are needed to enhance reliability and translation into clinic.

Anti-spike antibody responses to SARS-CoV-2 mRNA vaccines in people with schizophrenia and schizoaffective disorder.

Importance: Individuals with schizophrenia are at higher risk for severe COVID-19 illness and severe breakthrough infection following vaccination. It is unclear whether immune response to vaccination differs in this population. Objective: To assess whether anti-SARS-CoV-2 spike antibody titers after vaccination differ in people with a diagnosis of schizophrenia or schizoaffective disorder (SZ) compared to controls without a psychiatric disorder. Design: This cohort study assessed antibody response following the first and second dose of mRNA vaccines at longitudinal timepoints, up to 7 weeks following the first dose of vaccine. Setting: A multi-center study including psychiatric healthcare settings in the United States and Europe. Participants: 205 adults with no history of COVID-19 infection, including 106 individuals with SZ and 99 controls without a psychiatric disorder, who received their first dose of SARS-CoV-2 mRNA vaccine between December 20, 2020 and May 27, 2021. Main outcomes and measures: Mean SARS-CoV-2 anti-Spike IgG antibody levels within 7 weeks after the first dose of vaccination. Results: A total of 205 individuals (mean [SD] age, 44.7 [12.0] years; 90 [43.9%] male) were included, of which 106 (51.7%) were diagnosed with SZ. SZ was associated with lower mean log antibody levels (-0.15; 95% CI, -0.27 to -0.03, P = 0.016) after adjusting for age, sex, body mass index, smoking, days since vaccination, and vaccine manufacturer. In secondary analyses of dose-specific responses, SZ was associated with a lower mean log antibody level after the second dose of vaccine (-0.23; 95% CI -0.39 to -0.06, P = 0.006), but not the first dose of vaccine (0.00; 95% CI -0.18- 0.19, P = 0.96). Conclusions and Relevance: In this cohort study of individuals with SZ and a control group without psychiatric disorders, SZ was associated with lower SARS-CoV-2 anti-spike antibody levels following 2 doses of SARS-CoV-2 mRNA vaccination. This highlights the need for further studies assessing vaccine immunogenicity in individuals with schizophrenia.

Caspr2 autoantibody-associated Morvan syndrome predating thymoma relapse by 30 months.

Morvan's syndrome (MoS) is a rare autoimmune disorder characterized by central nervous system involvement, autonomic dysfunction and peripheral nerve hyperexcitability. MoS is believed to be caused by autoantibodies targeting contactin-associated protein 2 (Caspr2), a subunit of the neuronal voltage-gated potassium channel (VGKC) complex, usually in association with thymoma, less commonly with other malignancies. This case highlights an exceptional case of severe sleep disturbances and behavioural changes due to MoS, in a patient who would present with and be treated successfully for a second relapse of thymoma 30 months later. Originally he suffered from ocular myasthenia, another autoimmune disorder, which led to diagnosis of his original thymoma and first relapse.

Perspective: Solving the Heterogeneity Conundrum of TSPO PET Imaging in Psychosis.

Positron emission tomography using ligands targeting translocator protein 18 kDa (TSPO PET) is an innovative method to visualize and quantify glial inflammatory responses in the central nervous system in vivo. Compared to some other neuropsychiatric disorders, findings of TSPO PET in schizophrenia and related psychotic disorders have been considerably more heterogeneous. Two conflicting meta-analyses have been published on the topic within the last year: one asserting evidence for decreased TSPO uptake, while the other observed increased TSPO uptake in a selection of studies. In this paper, we review and discuss five hypotheses which may explain the observed variability of TSPO PET findings in psychotic illness, namely that (1) an inflammatory phenotype is only present in a subgroup of psychosis patients; (2) heterogeneity is caused by interference of antipsychotic medication; (3) interference of other clinical confounders in the study populations (such as age, sex, BMI, smoking, and substance use); or (4) methodological variability between studies (such as choice of tracer and kinetic model, genotyping, study power, and diurnal effects); and (5) the glial responses underlying changes in TSPO expression are themselves heterogeneous and dynamic. Finally, we propose four key recommendations for future research proposals to mitigate these different causes of heterogeneity.

Multiprobe molecular imaging of an NMDA receptor hypofunction rat model for glutamatergic dysfunction.

There are many indications of a connection between abnormal glutamate transmission through N-methyl-d-aspartate (NMDA) receptor hypofunction and the occurrence of schizophrenia. The importance of metabotropic glutamate receptor subtype 5 (mGluR5) became generally recognized due to its physical link through anchor proteins with NMDAR. Neuroinflammation as well as the kynurenine (tryptophan catabolite; TRYCAT) pathway are equally considered as major contributors to the pathology. We aimed to investigate this interplay between glutamate release, neuronal activation and inflammatory markers, by using small-animal positron emission tomography (PET) in a rat model known to induce schizophrenia-like symptoms. Daily intraperitoneal injection of MK801 or saline were administered to induce the model together with N-Acetyl-cysteine (NAc) or saline as the treatment in 24 male Sprague Dawley rats for one month. Biweekly in vivo [(11)C]-ABP688 microPET was performed together with mGluR5 immunohistochemistry. Simultaneously, weekly in vivo [(18)F]-FDG microPET imaging data for glucose metabolism was acquired and microglial activation was investigated with biweekly in vivo [(18)F]-PBR111 scans versus OX42 immunohistochemistry. Finally, plasma samples were analyzed for TRYCAT metabolites. We show that chronic MK801 administration (and thus elevated endogenous glutamate) causes significant tissue loss in rat brain, enhances neuroinflammatory pathways and may upregulate mGluR5 expression.