HERV-W association with serum biomarkers NfL and GFAP in multiple sclerosis

Background Serum biomarkers of multiple sclerosis (MS), such as glial fibrillary acidic protein (sGFAP) and neurofilament light chain (sNfL), are established indicators of disease progression and disability. Human endogenous retrovirus of the W family (HERV-W) is repeatedly associated with MS neuroinflammation, but its relationship with neural injury biomarkers is unclear. Methods We measured anti-pHERV-W, -syncytin-1 IgG and their Ratio (HERV-W) in 83 RR-MS patients and 112 healthy controls (HC). Age and sex-adjusted sNfL/sGFAP Z-scores were derived from HC-calibrated GAMLSS models. HERV-W humoral responses were correlated with MS severity score (MSSS), cytokines (Olink™), and clinical phases (acute MS [AMS], stable MS [SMS]) via non-parametric tests and discriminant analysis. Results HERV-W showed no difference between MS patients and HC, consistent with stable low-EDSS RR-MS predominance. HERV-W ratio correlated positively with Z-sNfL (ρ=0.67, p=0.012), Z-sGFAP (ρ=0.54, p=0.048), MSSS, and proinflammatory cytokines (IL-6, IL-1β and CXCL-9/10) specifically in AMS and SMS-EDSS>4 subgroups. AMS patients with elevated sNfL (>10 pg/ml; Z-score >1.5) exhibited markedly higher HERV-W than those with lower sNfl values. Discriminant models combining HERV-W, sNfL, and sGFAP achieved 82% of accuracy for MS-EDSS>4 classification; excluding HERV-W reduced the correct MS-EDSS>4 classification by 9.1%. Conclusions HERV-W humoral activity shows state-specific associations with sNfL and sGFAP and proinflammatory status in acute and high-disability MS phases. These findings support integrating HERV-W humoral response into dynamic biomarker panels to better stratify patients by inflammatory burden and disability trajectory, positioning it as a cofactor linking immune dysregulation to neurodegeneration rather than a singular MS marker.