CSF GFAP

Endogenous body fluid biomarkers are seldom disease-specific and have to be interpreted in the context of the clinical presentation and in parallel with other diagnostic tests. In this issue of Neurology®, Takano and colleagues¹ present compelling data on the utility of glial fibrillary acidic protein (GFAP), from a panel of cell-specific CSF biomarkers, in differentiating neuromyelitis optica (NMO) from multiple sclerosis (MS) and acute disseminated encephalomyelitis (ADEM), 2 demyelinating disorders that overlap phenotypically with NMO. In addition to GFAP, a marker typically associated with gliosis, their biomarker panel included S100B, an astrocyte activation marker, myelin basic protein (MBP) for demyelination, and neurofilament heavy chain (NFH), a marker of neuroaxonal damage.¹

The CSF GFAP levels were massively increased in patients with NMO-associated relapses compared to subjects presenting with MS relapses, ADEM, and other neurologic disorders.¹ The diagnostic sensitivity and specificity of CSF GFAP in differentiating NMO from these conditions was 90.9% and 77%, respectively.¹ Reassuringly, when limited to differentiating NMO from MS, the specificity increased to 97%.¹ CSF S100B showed a similar trend, but was less discriminating than GFAP.¹ In contrast, MBP and NFH did not differentiate NMO from MS.¹ These data support the hypothesis that an immunologic attack targeting astrocytes causes massively raised CSF GFAP levels. This is supported by recent pathologic studies demonstrating selective depletion of astrocytic foot processes and loss of aquaporin-4 …