Analytical Validation of an ELISA Assay for Maternal Autoantibody-Related Autism

Background/Objectives: Genetic and environmental factors during early development contribute to autism pathogenesis. Maternal autoantibodies recognizing specific fetal brain proteins can predict autism risk in a subset of cases. These antibodies cross the placenta and bind to their target antigens, which play critical roles in neurodevelopment, thereby increasing autism risk in this mechanistically defined subtype, Maternal Autoantibody-Related Autism (MARA). A multi-ELISA assay that detects maternal autoantibody combinations associated with increased autism risk has been described in the literature. This study aimed to transfer the MARA related autoantibody component assays to a clinical development laboratory for optimization and performance characterization. Methods: Indirect ELISA assays for eight maternal autoantibodies targeting LDH-A, LDH-B, GAH, STI1, CRMP-1, CRMP-2, NSE, and YB-1 were transferred from an academic laboratory to a clinical development laboratory for optimization and determination of the analytical performance and preliminary assay cutoff values. Standard methodologies were used to assess linearity, sensitivity, specificity, precision, and stability. Predefined validation protocols based on professional guidelines with established acceptance criteria for each parameter were followed. Results: Optimized ELISAs met the acceptable analytical performance criteria. All assays except one demonstrated excellent linearity when diluted with buffer or non-reactive plasma. The sensitivity analysis showed the lower limit of quantification discretely above the limit of detection, and below the preliminary population-based threshold values. Coefficients of variation for within-lot reproducibility of positive samples were