Plasma NfL and GFAP Fail as Biomarkers in Myasthenia Gravis, Controlled Study Finds
核心洞察
Plasma NfL and GFAP concentrations did not differ significantly between 90 MG patients and 40 healthy controls after age and sex adjustment (p > 0.05).
Neither biomarker correlated with disease severity as measured by MGFA class, MGC score, or MG-ADL scale, and no associations were found with myasthenic crisis history or treatment exposure.
The findings argue against adopting NfL or GFAP for MG monitoring and highlight the need for MG-specific biomarker strategies focused on immunological and molecular mechanisms.
A rigorously designed controlled cohort study has found that plasma neurofilament light chain (搜索) (NfL) and glial fibrillary acidic protein (搜索) (GFAP)—two biomarkers with established utility in neurodegenerative and neuroimmunological conditions—do not distinguish patients with myasthenia gravis (搜索) (MG) from healthy controls and show no association with disease severity. The findings, published July 7, 2026 in PLoS One, represent the first systematic evaluation of GFAP in MG and one of the largest controlled analyses of NfL in this patient population.
The study, led by Grosmane-Bataraga and colleagues, recruited 90 MG patients and 40 healthy controls at a single center between January and December 2023. Plasma NfL and GFAP were quantified using Single Molecule Array (Simoa) assays, with disease severity assessed via the Myasthenia Gravis (搜索) Foundation of America (MGFA) classification, Myasthenia Gravis Composite (MGC) score, and Myasthenia Gravis Activities of Daily Living (MG-ADL) scale.
No Significant Differences Between MG and Controls
After adjusting for age and sex, plasma concentrations of both NfL and GFAP did not differ significantly between MG patients and healthy controls (p > 0.05). The association between MG status and the studied outcomes appeared weak, with wide confidence intervals suggesting substantial uncertainty. Median concentrations for each group were reported in the study's Table 2.
Age emerged as a strong predictor of both biomarker levels. Each additional year of age predicted higher plasma NfL concentrations (estimate 0.33, 95% CI 0.17–0.49, p < 0.001) and GFAP concentrations (estimate 2.70, 95% CI 1.17–4.23, p < 0.001), relationships that remained significant after adjusting for sex, disease duration, MGFA class, and treatment status. No independent effect of sex was observed.
No Correlation with Disease Severity or Treatment History
Within the MG cohort, neither NfL nor GFAP correlated with any measure of clinical severity. No significant differences were observed across MGFA classes for NfL (p = 0.662) or GFAP (p = 0.331), and neither biomarker showed associations with MGC score or MG-ADL scale (p > 0.05 for all comparisons).
Secondary analyses evaluating associations with history of myasthenic crisis, exposure to rescue therapies including intravenous immunoglobulin (IVIg) and plasma exchange (PEX), and current maintenance treatment regimens all yielded non-significant results (p > 0.05).
Implications for Biomarker Development in MG
The findings highlight fundamental differences between MG and other neuroimmunological disorders such as multiple sclerosis and neuromyelitis optica spectrum disorders, where NfL reflects axonal injury and GFAP indicates astrocytic damage central to disease progression. By contrast, MG primarily affects the neuromuscular junction, where postsynaptic receptor dysfunction—rather than widespread axonal or astrocytic injury—drives clinical manifestations.
"Although secondary changes such as denervation and muscle atrophy may occur, they appear insufficient to produce robust or consistent increases in circulating NfL or GFAP," the authors note. They emphasize that "biomarkers validated in CNS-driven disorders cannot be assumed to translate to MG, where pathology is localised and immune-mediated in a distinct manner."
The study also addresses earlier reports suggesting elevated NfL in MG subgroups, noting that prior studies included older patients compared with controls. Because NfL strongly correlates with age, the authors suggest previously reported elevations may have been driven by demographic differences rather than MG pathology, underscoring the critical need for age adjustment when interpreting NfL levels.
Study Limitations
The authors acknowledge several limitations, including a relatively small and younger control group necessitating age adjustment, underrepresentation of seronegative and MuSK-positive subgroups, and imbalanced distribution across MGFA severity classes with small numbers in higher severity groups. These factors may have reduced the power to detect subtle subgroup differences.
The study concludes that future research "should prioritize the development of MG-specific biomarker strategies, focusing on immunological and molecular mechanisms more closely aligned with the pathophysiology of this rare disorder," including autoantibody profiles, immune cell phenotyping, cytokine signatures, and circulating microRNAs.