The autoantibody-mediated diseases of the central nervous system are a recently described group of neuropsychiatric conditions. In an individual patient, the autoantibodies target a single antigen, and their direct pathogenicity has been shown both in vitro and in vivo. Patients with autoantibody-mediated encephalitides typically show moderate improvements with early immunotherapies. However, outcomes remain unsatisfactory with frequent, significant side effects from immunotherapies and >80% of patients unable to return to work due to cognitive disability. Despite its clear importance for future therapeutic advances, the development of autoantibody pathogenicity within B cell lineages has received very little attention. Furthermore, it remains unknown whether the intrathecal autoantibodies are generated exclusively by B cells that have entered the cerebrospinal fluid (CSF). Single-autoantigen driven diseases offer a unique opportunity to exclusively study the autoantigen-specific lineage towards a comprehensive mechanistic understanding of disease pathology, and to elucidate how pathogenic IgGs access the CNS. IgG4 is the predominant subclass of pathogenic autoantibodies in many neurological diseases. IgG4 molecules have a unique mechanism of action as they have the potential to form functionally monovalent antibodies, a process known as ‘Fab arm exchange’. This differs to the bivalency of other autoantibody subclasses. In my characterisation of patient sera, I have observed that non-IgG4 LGI1-specific antibodies are rare in patients with the best clinical outcomes, therefore a comprehensive functional characterisation of patient autoantibodies of IgG4 and non-IgG4 subclasses will inform the role of the antibody valency in disease pathogenesis. This study aims to characterise the development of IgG4 and non-IgG4 LGI1-specific antibodies in the periphery and CSF of our patients. We will clone these antibodies and determine their relative pathogenicity in vitro, and use prospective, longitudinal analyses to correlate these parameters with patient outcomes. This approach will allow the identification of the most pathogenic clones and determine their relative access to the CSF. These findings will have wider implications across IgG4-mediated neurological conditions.