CSF Metabolomic Signatures Predict Shunt Outcomes for Patients with Normal Pressure Hydrocephalus (NPH)

Introduction: Molecular biomarkers for neurodegenerative diseases are essential for advancing diagnostics and treatment. NPH, a neurodegenerative disorder, is marked by progressive gait and cognitive impairment, and urinary incontinence. Identifying NPH patients who are likely to benefit from shunt surgery remains a significant clinical challenge. Recent genome-wide RNA sequencing of extracellular vesicles in CSF from NPH patients revealed genes and pathways associated with symptom improvement following surgery (Levin et al., Brain 2023). Here, we build on these findings by presenting metabolomic signatures in CSF that further predict outcomes of shunt surgery, providing an additional layer of biomarkers to support patient selection and enhance therapeutic strategies.

Methods: CSF samples for metabolomic profiling were obtained from 42 NPH patients at the time of ventricular catheter placement. Symptom improvement was clinically assessed before and 3 months after surgery. Metabolomic profiling was conducted by Metabolon. Data normalization and processing steps, such as mean-centering, log transformation, and Pareto scaling, were applied to account for variability and ensure accurate comparisons. This study was approved by the institutional review board (#1492994).

Results: CSF samples were analyzed for 456 metabolites spanning 96 chemical categories. Differential analysis revealed distinct metabolite profiles between patients who experienced symptom improvement and those who did not. Key metabolic pathways, such as kynurenine metabolism and single carbon metabolism, were altered, with potential links to neuroinflammation. Individual metabolites, like heme and bilirubin, also varied across patient groups, suggesting possible biomarkers for disease modulation. Principal component analysis (PCA) demonstrated class separability, and an SVM model yielded high AUC scores (0.79 for cognition, 0.78 for urinary symptoms, and 0.85 for gait), indicating good predictive performance.

Conclusion : These findings underscore the potential of CSF metabolite profiling to enhance patient selection and prognosis for shunt surgery, with significant implications for improving NPH diagnosis, treatment, and understanding of disease pathology.