Patterns of autoantibody expression in multiple sclerosis identified through development of an autoantigen discovery technology
Europe B. DiCillo, … , David Pisetsky, Thomas Tedder
Europe B. DiCillo, … , David Pisetsky, Thomas Tedder
Published March 3, 2025
Citation Information: J Clin Invest. 2025;135(5):e171948. https://doi.org/10.1172/JCI171948.
View: Text | PDF
Research Article Autoimmunity Neuroscience

Patterns of autoantibody expression in multiple sclerosis identified through development of an autoantigen discovery technology

Abstract

Multiple sclerosis (MS) is a debilitating autoimmune disease of the CNS, which is characterized by demyelination and axonal injury and frequently preceded by a demyelinating event called clinically isolated syndrome (CIS). Despite the importance of B cells and autoantibodies in MS pathology, their target specificities remain largely unknown. For an agnostic and comprehensive evaluation of autoantibodies in MS, we developed and employed what we believe to be a novel autoantigen discovery technology, the Antigenome Platform. This Platform is a high-throughput assay comprising large-fragment (approximately 100 amino acids) cDNA libraries, phage display, serum antibody screening technology, and robust bioinformatics analysis pipelines. For autoantibody discovery, we assayed serum samples from CIS patients who received either placebo or treatment who were enrolled in the REFLEX clinical trial, which assessed the effects of IFN-β-1a (Rebif) clinical and MRI activity in patients with CIS. Serum autoantibodies from patients with CIS were significantly and reproducibly enriched for known and previously unreported protein targets; 166 targets were selected by over 10% of patients’ sera. Further, 10 autoantibody biomarkers associated with disease activity and 17 associated with patient response to IFN-β-1a therapy. These findings indicate widespread autoantibody production in MS and provide biomarkers for continued study and prediction of disease progression.

Authors

Europe B. DiCillo, Evgueni Kountikov, Minghua Zhu, Stefan Lanker, Danielle E. Harlow, Elizabeth R. Piette, Weiguo Zhang, Brooke Hayward, Joshua Heuler, Julie Korich, Jeffrey L. Bennett, David Pisetsky, Thomas Tedder

×

Figure 5

Autoantigen selection by CIS antibodies may predict disease activity and therapeutic response.

Options: View larger image (or click on image) Download as PowerPoint
Autoantigen selection by CIS antibodies may predict disease activity and...
(A–G) LASSO model parameters for predicting disease activity in the absence of therapeutic intervention using REFLEX placebo samples. (A) The Solution Path Plot displays values of the estimated parameters, where each curve represents a predictive term in the model. (B) The Validation Plot includes a curve for both the training and validation sets at various magnitudes of scaled parameter estimates. In each plot (A and B), the x-axis represents the l1 norm, and the vertical red line represents the value of the l1 norm for the best and chosen solution. (C–E) The ROC curve for the (C) training, (D) validation, and (E) test samples and the associated AUC values. (F) Effects chart. “Main Effect” shows the relative contribution of the predictor to the model alone, and “Total Effect” shows the relative contribution of the predictor when other predictors are also taken into account. (G) Plot of Shapley values where each dot represents a patient sample in the model. The color of the dot represents their raw value. The x-axis represents the effect of the sample, where a negative effect indicates a contribution to the PBO-A outcome and a positive effect indicates a contribution to the PBO-NA outcome. (H–L) LASSO model parameters for predicting response to IFN-β-1a therapy; panels were constructed as described for A–G above. (H) Solution Path Plot, (I) The Validation Plot. (J–L) ROC curves for the (J) training, (K) validation, and (L) test samples and their AUC values. (M) Effects chart. (N) Plot of Shapley values, where a negative effect indicates a contribution to the RNF-A outcome and a positive effect indicates a contribution to the RNF-NA outcome.