Novel Markers of Neuroinflammation in Cerebrospinal Fluid

Abstract

Diagnosis of autoimmune encephalitis (AE) is challenging due to diverse clinical presentations and limited sensitivity of current biomarkers, particularly in antibody-negative cases or those presenting solely with psychiatric symptoms. The overlap between AE and psychiatric disorders has renewed interest in immune contributions to psychiatric disease, highlighting the need for improved diagnostic tools. This thesis aims to identify novel cerebrospinal fluid (CSF) biomarkers to enhance diagnostic precision in AE and explore immune involvement in psychiatric illness.

Initial studies assessed conventional and emerging biomarkers in patients with high clinical suspicion of AE, compared with non-inflammatory (NI), viral infection, and inflammatory neurological disease control groups. These analyses confirmed the poor sensitivity of current tests but identified two cytokines, IL-21 and IP-10, as promising candidates for improving diagnostic yield.

Subsequent studies evaluated serum and routine CSF markers of autoimmunity and neuroinflammation in a psychiatric disease cohort, alongside CSF cytokine profiling, with comparison to with the AE and NI control groups. A subset of psychiatric patients demonstrated high CSF cytokine levels, suggestive of underlying immune dysregulation.

Finally, an unbiased CSF proteomic analysis using mass spectrometry across these cohorts revealed multiple proteins and pathways associated with immune and neural processes, providing a basis for future biomarker discovery. Several pre-analytical and post-analytical challenges were also identified, offering important insights for the design of future studies.

Collectively, the findings from these investigations provide new insights into the immune basis of CNS disease and underscore the potential of novel CSF biomarkers to refine diagnosis, guide therapeutic decisions, and improve outcomes. Further validation in larger, independent cohorts will be required to establish their clinical utility.