Decoding Chinese speech across multiple neural conditions via EEG: dataset construction and interpretability driven spatial optimization.

Severe mental illness is increasingly understood as a systemic condition in which immune, metabolic, and vascular pathways interact with brain function across diagnoses. Acute psychiatric in-patients represent a clinically informative population in which severe psychopathology, medication exposure, and systemic physiological stress commonly converge. People with severe mental illness (SMI) have high rates of physical comorbidity, modifiable cardiometabolic risk factors, and cardiovascular mortality, partly related to behavioral risk factors, healthcare inequalities, and adverse metabolic effects of psychotropic medication (;). Meta-analytic evidence supports that these physical health risks contribute substantially to excess premature mortality in both schizophrenia spectrum disorders (SSD) and mood disorders (MD) ( & 2023) in addition to marked reductions in life expectancy due to chronic medical conditions (). Circulating biomarkers may, therefore, provide clinically relevant information about biological burden. A systematic evaluation of inflammatory changes across multiple psychiatric disorders supported the possibility of differentiating psychiatric disorders by using inflammatory biomarkers and further claimed that inflammation may have a large enough effect size that these biomarkers could be detected in relatively small sample sizes (). However, it remains unclear whether many biomarkers map meaningfully onto categorical diagnoses or instead reflect transdiagnostic physiological processes. Growth differentiation factor 15 (GDF-15), also known as macrophage inhibitory cytokine-1 (MIC1), is a cytokine rising in response to cellular stress and regulated during inflammation, mitochondrial dysfunction, and tissue injury (; Jena et al., 2023; ; ). GDF-15 is overexpressed under a broad range of conditions, including cardiovascular disease, cancer, cachexia syndromes, and insulin sensitivity states (). GDF-15 increases with age and decreasing telomerase activity, is associated with suboptimal recovery following physical illness, high cardiovascular mortality, cancer, and poorer cognition. These features make GDF-15 a plausible candidate for capturing systemic burden in SMI, where inflammatory activation and renal-metabolic vulnerability commonly co-occur. Evidence for altered GDF-15 in SMI remains limited but a potential transdiagnostic signal has been suggested. In a Swedish SSD out-patient cohort, found higher plasma GDF-15 levels in patients compared to controls, and there were no differences in GDF-15 between psychosis diagnostic subgroups after adjustment for age and smoking. Within the psychosis cohort, GDF-15 correlated positively with age and was higher in smoking males. Importantly, in a multivariate regression model, greater psychosis severity was associated with lower GDF-15 levels. There was no significant correlation between GDF-15 and high-sensitivity C-reactive protein (hsCRP), however, suggesting that GDF-15 may capture a process partly distinct from CRP-indexed inflammation. A further study found circulating GDF-15 to be elevated in chronic, clinically stable male SSD in-patients compared to controls. GDF-15 was positively associated with Body Mass Index (BMI) and negatively with cognitive performance after adjusting for age, illness duration, smoking, education, and antipsychotic dose (). More recently, a cross-sectional case-control study from Brazil examined GDF-15 variation transdiagnostically across out-patients with SMI and healthy controls, and showed that GDF-15 was elevated in all psychiatric groups, suggesting that this marker reflects systemic stress load common to SMI across diagnoses rather than representing a signal specific to, for example, schizophrenia, which may add to physical health deterioration over the course of mental disorders. In this study, multivariate models identified diagnostic group and the number of psychiatric hospitalizations as significant main effects on GDF-15 levels. In contrast, age, sex, illness duration, and age of onset were not significant predictors of GDF-15 levels ().