Evaluation of the Fujifilm Wako β-Glucan assay in cerebrospinal fluid for the diagnosis of non-cryptococcal fungal infections of the central nervous system

Abstract

Background

(1→3)-β-D-glucan (BDG), a component of the fungal cell wall, has emerged as a potential cerebrospinal fluid (CSF) biomarker for fungal infections of the central nervous system (FI-CNS). While the Fungitell® assay has been well studied in this context, the diagnostic performance of the Fujifilm Wako β-glucan assay in CSF remains unknown. The present study aimed to evaluate the Fujifilm Wako assay for diagnosing non-cryptococcal FI-CNS, establish an optimal CSF-specific BDG cut-off value, and assess its concordance with the Fungitell® assay.

Methods

We prospectively analysed CSF samples from patients with suspected non-cryptococcal FI-CNS using the Fujifilm Wako assay. Cases were classified as proven or probable FI-CNS based on revised EORTC/MSGERC criteria, and non-fungal neurological cases served as controls. BDG levels were compared across groups, receiver operating characteristic (ROC) analysis was used to identify an optimal diagnostic threshold, and agreement with the Fungitell assay was assessed.

Results

Sixty-nine patients with suspected non-cryptococcal FI-CNS (4 proven, 65 probable) and 35 control patients were included. BDG was positive in 95.6% of the suspected cases, including all proven cases. An optimal cut-off of 10 pg/mL yielded 91.2% sensitivity, 100% specificity, 100% PPV, and 85.4% NPV, with an area under the curve (AUC) of 0.976. Agreement with the Fungitell assay was strong (Cohen’s κ = 0.877 at this threshold).

Conclusion

The Fujifilm Wako β-glucan assay demonstrates high diagnostic accuracy for non-cryptococcal FI-CNS when used on CSF, particularly at a 10 pg/mL cut-off. Larger multicentre studies with standardized protocols are needed to confirm these findings and support broader clinical adoption.