Metabolic Imaging for Diagnosis and Prognostication of Autoimmune encephalitiS

Autoimmune encephalitis (AE) is a rare neurological disorder mediated by autoimmune antibody response against neuronal cell surface and intraneuronal proteins associated with specific brain areas, resulting in severe inflammation and damage in the associated brain regions, all most frequently manifesting diverse cognition and memory impairment symptoms at follow-up. However, these symptoms may co-exist or mimic other CNS autoimmune and neurodegenerative disorders. The most common guideline for diagnosing autoimmune encephalitis relies on cerebrospinal fluid (CSF) antibody testing which might take several weeks to obtain, making it not optimal for the early diagnosis of AE. As for magnetic resonance imaging (MRI), which is the most common imaging tool utilized for aiding in the diagnosis of AE, can possess several limitations as some patients, like anti-NMDAr AE patients, can present memory and behavioral deficits even in the presence of normal brain MRI. Positron emission tomography (PET) with 2-deoxy-2-\[fluorine-18\] fluoro-D-glucose (18F-FDG) have been addressed by several studies as an important examination for the early diagnosis of AE . One study demonstrated that the fraction of having an abnormal MRI in AE patients is lower than having an abnormal PET, by which certain PET patterns were associated with autoantibody types of AE. Moreover, one report demonstrated that even with autoantibody negative test and normal brain MRI, FDG-PET examination showed abnormal hypometabolism and hypermetabolism patterns. More specifically, these distinct patterns include medial temporal and striatal hypermetabolism with cortical diffuse hypometabolism. Leiris et al. revealed that the methadology used for the analysis of these PET images is highly variable, especially intensity normalization methods, where most possess some limitations (e.g., proportional scaling) as they can impede the accurate differential diagnosis of autoimmune encephalitis (AE) by potentially indicating false hypermetabolism in otherwise preserved brain regions. Absolute quantification is not possible since the disease presents both diffuse hypometabolism and hypermetabolism on PET images. So, they suggested that it's best to parametrize the brain's activity by dividing it by that of the striatum. Their voxel-based analysis, comparing individuals with AE to both healthy subjects and those with mild cognitive impairment (MCI), demonstrated that a decrease in the cortex/striatal metabolic ratio is a robust biomarker for the early diagnosis of AE.