Cerebrospinal fluid-based Onco-mNGS for the identification of neoplastic meningitis: CNV profiling and exploratory...

To assess the value of cerebrospinal fluid (CSF)-based Onco-mNGS for identifying neoplastic meningitis (NM) and to explore copy number variation (CNV) patterns and microbial sequencing signals generated from the same assay.

In this retrospective study, 15 patients who underwent CSF Onco-mNGS were included, comprising 10 patients with clinically diagnosed NM and 5 non-neoplastic controls. Human-derived reads were analyzed for CNV-based neoplasm assessment, whereas non-human reads were analyzed separately for exploratory microbial profiling. Diagnostic performance was evaluated against the clinical classification.

Two NM samples failed CNV quality control and were classified as indeterminate. Among 13 interpretable samples, CNV-based neoplasm assessment achieved a sensitivity of 87.5% (95% CI, 47.3%-99.7%), specificity of 100% (47.8%-100%), accuracy of 92.3% (64.0%-99.8%), positive predictive value of 100% (59.0%-100%), and negative predictive value of 83.3% (35.9%-99.6%). The overall interpretable yield was 86.7% (13/15). Exploratory analyses showed heterogeneous CNV patterns among NM cases and differences in microbial sequencing profiles between the neoplastic and non-neoplastic groups.

CSF-based Onco-mNGS may provide useful CNV-based evidence supporting the identification of NM. The tumor-origin and microbial findings should be considered exploratory because of the small cohort, limited subgroup sizes, and the potential for contamination in low-biomass CSF specimens.