Fully Automated Measurement of GFAP in CSF Using the LUMIPULSE System: Implications for Alzheimer's Disease Diagnosis and Staging.

Glial fibrillary acidic protein (GFAP) has been shown to be a reliable biomarker for detecting neurological disorders. Recently, we developed the Lumipulse GFAP plasma assay, which is a commercially available tool. Compared to existing assays, the LUMIPLSE platform offers the high-throughput, rapid, and fully automated quantification of biomarkers, enabling more standardized and accessible clinical study. In this study, we evaluated this assay using cerebrospinal fluid (CSF) samples. Assessing GFAP in CSF may provide more direct insights into central nervous system pathology than plasma and could improve the characterization of Alzheimer's disease (AD) stages and support treatment monitoring. The LUMIPULSE G system is a chemiluminescent enzyme immunoassay (CLEIA) platform equipped with full automation, utilizing specialized cartridges to process samples within 30 min. The assay, which employs a pair of proprietary monoclonal antibodies targeting GFAP, was evaluated for clinical performance using 30 CSF samples from patients diagnosed with AD, patients with mild cognitive impairment (MCI), and cognitively unimpaired (CU) individuals, with 10 samples from each group. In addition, levels of β-amyloid 1-40 (Aβ40), β-amyloid 1-42 (Aβ42), and pTau181 were simultaneously measured. The Lumipulse GFAP assay significantly differentiated ( < 0.05) between the amyloid accumulation and non-amyloid accumulation groups, as classified based on the CSF Aβ test. Furthermore, GFAP showed a moderate correlation with pTau181 (r = 0.588), as determined based on Spearman's rank correlation coefficient. Moreover, receiver operating characteristic (ROC) analysis was performed to determine the performance of GFAP in distinguishing amyloid-positive and amyloid-negative subjects, with an area under the curve (AUC) of 0.72 (0.50-0.93). When stratified by CSF pTau181 positivity, GFAP demonstrated an improved diagnostic accuracy, achieving an AUC of 0.86 (95% CI: 0.68-1.00). This study demonstrates that the Lumipulse GFAP assay, when applied to CSF samples, has the potential to differentiate AD from non-AD cases, particularly suggesting its utility in detecting tau-related pathology. While GFAP has previously been established as a biomarker for AD, our findings highlight that combining GFAP with other biomarkers such as Aβ40, Aβ42, and pTau181 may enhance the understanding of AD pathogenesis, disease staging, and possibly treatment responses. These findings suggest that GFAP may serve as a complementary biomarker reflecting astroglial reactivity associated with tau positivity, alongside established biomarkers such as Aβ40, Aβ42, and pTau181. However, since GFAP levels may also be elevated in other neurological disorders beyond AD, further investigation into these conditions is required.