2303 A study of multimodality outcome biomarkers from the Australian autoimmune encephalitis consortium

Autoimmune encephalitis (AE) encompasses a number of neuroinflammatory conditions. Optimal biomarkers to guide therapeutic decision making and prognostication remain to be identified.ObjectiveTo identify biomarkers of disease severity and predictors of patient outcome.MethodsClinical, radiological, serological, CSF, EEG and neuropsychological data in patients with AE are being collected from (12 sites) around Australia. Samples are collected on a 6-monthly basis for 3.5 years for the prospective arm to determine disease biomarkers.ResultsWe have recruited 85 prospective, and 153 retrospective cases of AE. In seropositive cases, high neutrophil to lymphocyte ratio (NLR) and CSF WCC>5 cells/mm3 were associated with higher odds of first line treatment failure (OR 1.32, 95%CI 1.03, 1.69). Patients with a CSF WCC>20 cells/mm3 had a higher rate of admission to critical care unit (OR 17.2, CI 3.12–93.3). Superimposed fast activity on EEG (OR 2.31; 1.75 to 3.04) was more common in anti-NMDAR antibody associated AE compared to other subtypes. Abnormalities on MRI were detected in 35% of seropositive patients, while PET was abnormal in 46%. Patients who developed medial temporal lobe atrophy had higher final mRS (HR 0.15, 95%CI 0.03, 0.74, p=0.020). In a subgroup of 59 patients who underwent detailed neuropsycholgocial assessment, deficits in psychometric markers of executive dysfunction were the most commonly observed.ConclusionWe present the first data gathered as part of the Australian AE Consortium. Our findings identify multimodal biomarkers that could be used for the development of clinical guidelines for AE diagnosis and treatment and assist with predicting prognosis.