Psychiatry · 2 h ago
Lipidomic models distinguish schizophrenia groups in CSF and serum
A cross-sectional study analyzed matched cerebrospinal fluid and serum samples from 49 antipsychotic-naïve patients with schizophrenia, 56 treated patients, and 51 healthy volunteers. Machine-learning models identified lipid patterns that distinguished groups, but the findings are exploratory and do not establish a diagnostic test.
- Matched CSF and serum lipid profiles were assessed in 156 participants.
- Model AUCs ranged from 0.87 to 0.94 across groups and fluids.
- Anandamide and d18:2 sphingolipids were prominent in naïve patients.
- Biomarker findings need independent validation before clinical use.
Researchers measured endocannabinoids, related N-acylethanolamines, ceramides, and sphingomyelins by liquid chromatography–tandem mass spectrometry in matched CSF and serum samples. The cohort included 49 antipsychotic-naïve patients with schizophrenia, 56 receiving antipsychotic treatment, and 51 healthy volunteers. The IDEARS machine-learning platform ranked lipid features and assessed their ability to discriminate between groups.
Reported areas under the curve were 0.87 and 0.94 in CSF for naïve and treated patients, respectively, and 0.87 and 0.90 in serum. Anandamide was the top CSF feature among naïve patients and was elevated versus healthy volunteers (q=0.010). Sphingolipids with a d18:2 backbone were elevated in both CSF and serum in naïve patients (q<0.05). Treated patients showed a pattern of higher C16–C18 and lower greater-than-C24 ceramides; the authors suggest this may reflect treatment-related metabolic effects.
The results support further investigation of lipid markers but do not establish clinical diagnostic or monitoring performance. The study compared selected patient groups with healthy volunteers, and the reported classification metrics require validation in independent, clinically representative cohorts before clinical use. The serum anandamide finding was not statistically significant by conventional analysis.
Is this summary clinically accurate?
Help fellow clinicians: your rating sends inaccurate summaries straight to our editors.
Sign in to rate this summary →Source
Molecular Psychiatry: Identification of fluid endocannabinoid and sphingolipid biomarkers in schizophrenia. A machine learning approach toward clinical value ↗This is an automated AI-condensed summary that has not yet been reviewed by an editor. Always consult the full item at the original source.