Autoimmune responses to myelin-associated proteins as diagnostic and prognostic biomarkers of relapsing-remitting multiple sclerosis: associations with human herpesvirus-6 and Epstein-Barr Virus reactivation

Abbas F. Almulla, Michael Maes, Aristo Vojdani, Elroy Vojdani, Jing Li, Yingqian ZhangmedRxiv 2024 · DOI

This study, recently shared as a preprint on medRxiv, provides compelling evidence that specific autoantibody profiles can serve as high-precision tools for both diagnosing relapsing-remitting multiple sclerosis (RRMS) and predicting its clinical course. By linking these autoimmune attacks to the reactivation of common herpesviruses, the researchers offer a clearer picture of the "viral trigger" hypothesis in MS pathogenesis.

TL;DR

Researchers identified a specific signature of IgG, IgA, and IgM autoantibodies against myelin proteins that distinguishes RRMS patients from healthy controls with over 96% accuracy. Crucially, the study demonstrates that nearly half of the variance in these autoimmune responses is explained by the reactivation of Epstein-Barr Virus (EBV) and Human Herpesvirus 6 (HHV-6), suggesting these viruses drive the progression of the disease.

What Problem Does This Paper Address?

Multiple Sclerosis (MS) is a chronic autoimmune disorder where the immune system attacks the protective myelin sheath of the central nervous system. While we have long suspected a link between viral infections (like EBV) and MS, the exact relationship between viral reactivation and specific autoimmune biomarkers has remained elusive.

Furthermore, diagnosing the most common form—relapsing-remitting MS (RRMS)—and predicting how quickly a patient’s disability will progress is challenging. Current diagnostic tools often lack the sensitivity to capture the full breadth of the immune response, particularly the roles of IgA and IgM isotypes, which are often overlooked in favor of IgG.

Methods

The researchers conducted a case-control study involving 55 RRMS patients (in the remission phase) and 63 healthy controls. They employed several sophisticated analytical techniques:

  • Biomarker Assays: They measured IgG, IgA, and IgM responses against seven key myelin-related components (including MBP, MOG, and MAG) and viral markers for EBV (EBNA-1, dUTPase) and HHV-6 (dUTPase) using ELISA.
  • Clinical Correlation: Disability was measured using the Expanded Disability Status Scale (EDSS), and disease progression was tracked via the Multiple Sclerosis Severity Score (MSSS).
  • Machine Learning: Beyond standard regression, the team used Multilayer Perceptron (MLP) neural networks to classify patients.
  • Statistical Modeling: They utilized Principal Component Analysis (PCA) to group biomarkers and Canonical Correlation Analysis (CCA) to explore the complex relationship between viral reactivation and autoimmune responses.

Datasets

The study utilized a primary clinical dataset of 118 participants recruited from the Neuroscience Centre at Alsader Medical City in Al-Najaf, Iraq. While no external public benchmarks were used, the study provides a detailed intercorrelation matrix (ESF Table 2) that serves as a valuable reference for the relationship between viral antigens and neuropsychiatric rating scales.

Key Insights and Ideas

  • The "Isotype Matters" Revelation: While most MS research focuses on IgG, this study found that IgA and IgM responses to proteins like MOG and MAG are critical. In fact, combining IgG-MBP with IgM-MAG-37-60 provided the best discrimination between patients and controls.
  • The Viral Driver: A staggering 47.4% of the variance in autoimmune responses to myelin was explained by immune responses to EBV and HHV-6. This supports the theory of "molecular mimicry," where the immune system confuses viral proteins with the body's own myelin.
  • GLIAL-CAM and EBNA: The study confirmed a strong association between IgA directed at GLIAL-CAM and responses to EBV nuclear antigen (EBNA), suggesting a specific pathway for viral-induced neurodegeneration.
  • Prognostic Power: The identified biomarkers explained roughly 75% of the variance in disability (EDSS) and 63% of the variance in disease progression (MSSS), making them potentially superior to current prognostic markers.

Key Results

  • Diagnostic Accuracy: A logistic regression model using just two markers (IgG-MBP and IgM-MAG-37-60) achieved 96.6% accuracy (95.7% sensitivity, 95.2% specificity).
  • Neural Network Performance: The MLP neural network achieved 100% accuracy in a holdout sample with an Area Under the ROC Curve (AUC) of 0.991.
  • Effect Sizes: The strongest individual discriminator was IgG against myelin basic protein (IgG-MBP), with a very high effect size (Partial Eta Squared = 0.661).

Limitations and Future Work

The authors acknowledge several areas for further exploration:

  • Subtype Generalization: The study focused exclusively on RRMS. It remains to be seen if these same biomarkers are effective for Primary Progressive (PPMS) or Secondary Progressive (SPMS) subtypes.
  • Disease Phase: Because the participants were in remission, future research is needed to see how these markers fluctuate during acute relapses.
  • Causality vs. Association: While the link to viral reactivation is strong, clinical trials investigating the impact of antiviral treatments on these specific autoimmune biomarkers are necessary to prove a causal link.
  • Geographic Diversity: Validation in cohorts from different ethnicities and geographic regions is required to ensure the global applicability of these biomarkers.

Summary updated

Abstract

Abstract Background The pathogenesis of relapsing-remitting multiple sclerosis (RRMS) is linked to autoimmune attacks against myelin proteins, and reactivation of Epstein-Barr virus (EBV) and human herpesvirus 6 (HHV-6). However, the connection between viral reactivation and autoimmune biomarkers has remained unclear. Objectives To investigate immunoglobulin (Ig)G/IgA/IgM responses targeting myelin-related proteins in association with EBV and HHV-6 replication markers in RRMS. Methods We recruited 55 patients with RRMS and 63 healthy controls and assessed IgG/IgA/IgM responses against seven myelin-related components, as well as EBV nuclear antigen 1 (EBNA-1) and deoxyuridine-triphosphate nucleotidohydrolase (dUTPases). Disability was evaluated using the Expanded Disability Status Scale (EDSS) and disease progression using the Multiple Sclerosis Severity Score (MSSS). Results IgG/IgA/IgM levels targeting seven myelin-related proteins were significantly higher in RRMS than in controls. IgG against myelin basic protein (MBP) (IgG-MBP), IgM-myelin-associated glycoprotein (IgM-MAG)-37-60, IgA-MBP, and IgA-myelin-oligodendrocyte-glycoprotein (IgA-MOG-31-55) distinguished RRMS from controls with a predictive accuracy of 96.6% (sensitivity = 95.7%, specificity = 95.2%) and an area under the ROC curve of 0.991. A large part of the variance in the EDSS (around 75%) and MSSS score (62.8%) was explained by IgG-MBP, IgM-MBP, IgA-MOG-31-55, and IgM-MAG. Part of the variance (47.4%) in the IgG/IgA/IgM responses to myelin-related proteins was explained by immune responses to EBNA and deoxyuridine-triphosphate nucleotidohydrolases of EBV and HHV-6. Conclusions Autoimmune reactivities targeting myelin-related proteins are valuable biomarkers of RRMS and the severity and progression of RRMS. Reactivation of EBV and HHV-6 may trigger or maintain these autoimmune responses thereby impacting disease progression.
BibTeX
@misc{almulla2024autoimmune,
  author = {Abbas F. Almulla and Michael Maes and Aristo Vojdani and Elroy Vojdani and Jing Li and Yingqian Zhang},
  title = {{Autoimmune responses to myelin-associated proteins as diagnostic and prognostic biomarkers of relapsing-remitting multiple sclerosis: associations with human herpesvirus-6 and Epstein-Barr Virus reactivation}},
  year = {2024},
  doi = {10.1101/2024.11.16.24317434},
  howpublished = {medRxiv},
}
Figure