Revisiting IgG antibody reactivity to Epstein-Barr virus in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome and its potential application to disease diagnosis

Nuno Sepúlveda, João Malato, Franziska Sotzny, Anna D. Grabowska, André Fonseca, Clara Cordeiro, Luís Graça, Przemysław Biecek, Uta Behrends, Josef Mautner, Francisco Westermeier, Eliana Lacerda, Carmen Scheibenbogen

Abstract

Infections by the Epstein-Barr virus (EBV) are frequently associated with the onset of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS). However, traditional serological analyses often fail to distinguish patients from healthy controls (HCs), leaving the biomarker potential of EBV-derived antigens unclear. In this study, we re-analyzed a high-resolution peptide microarray dataset of IgG antibody responses against 3,054 EBV-related antigens in 92 ME/CFS patients and 50 HCs. While no universal antibody signature was found across the entire cohort, a distinct signature emerged when patients were stratified by disease trigger. Specifically, in patients with an infectious disease onset, antibody responses to two antigens—EBNA4_0529 and EBNA6_0070—showed significant diagnostic potential. When combined with age and gender in a regression model, these markers achieved a sensitivity of 0.833 and a specificity of 0.720. These findings suggest that EBV antibody levels may serve as viable biomarkers for a specific subgroup of ME/CFS patients and point toward molecular mimicry as a potential pathological mechanism.

Introduction

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is a debilitating multi-systemic condition characterized by persistent fatigue, post-exertional malaise, and cognitive impairment. Despite its severity, the lack of definitive diagnostic biomarkers remains a major hurdle for clinical management. Epstein-Barr virus (EBV) has long been a primary suspect in ME/CFS pathogenesis, as many patients report their illness began with infectious mononucleosis or a similar viral event.

Previous research into EBV serology has yielded inconsistent results, likely due to the inherent heterogeneity of the ME/CFS population. This study addresses this challenge by re-evaluating high-resolution antibody data through patient stratification and the inclusion of demographic interactions, aiming to identify specific viral epitopes that can reliably distinguish patients from healthy individuals.

Methods

Dataset and Patient Stratification

The study utilized a dataset from 92 ME/CFS patients (recruited from the Charité outpatient clinic in Berlin) and 50 age- and gender-matched healthy controls. The patient cohort was divided into two groups based on disease onset:

  1. Infectious Trigger (n=54): Patients reporting an acute infection at the start of their illness.
  2. Non-infectious/Unknown Trigger (n=38): Patients with a gradual or non-viral onset.

Peptide Microarray and Statistical Modeling

Antibody reactivity was measured using a peptide microarray covering 3,054 overlapping 15-mer segments of 14 different EBV proteins. To analyze this complex data, researchers employed regression models (logistic, probit, and complementary log-log) to classify disease status.

Crucially, the models included interaction terms for age and gender, acknowledging that immune responses are heavily influenced by these factors. A backward stepwise selection process was used to identify the most predictive antigens while minimizing redundancy.

Key Results

The Importance of Disease Onset

The analysis revealed that the "antibody signature" is not universal across all ME/CFS patients. When the entire cohort was analyzed as a single group, no significant differences were found compared to controls. However, when the analysis was restricted to the infectious-trigger subgroup, a clear signal emerged.

Identified Biomarkers

Two specific EBV antigens were identified as key predictors of ME/CFS in the infectious-trigger group:

  • EBNA4_0529
  • EBNA6_0070

For the EBNA4 antigen, the odds of being a patient increased approximately 9.5 times for every fold-change in antibody levels. The interaction between EBNA6_0070 and gender was particularly notable; in women, the probability of a diagnosis increased steadily with both age and antibody levels, whereas men showed a more complex, non-linear pattern.

Diagnostic Performance

The final classification model for the infectious-trigger subgroup demonstrated robust performance:

  • Sensitivity: 0.833
  • Specificity: 0.720
  • Area Under the Curve (AUC): 0.835

In contrast, the same model performed no better than a random guess when applied to the non-infectious trigger group, highlighting the biological distinctness of these patient subsets.

Discussion and Molecular Mimicry

Bioinformatic analysis of the identified EBNA6 segments revealed they are rich in arginine and share structural similarities (homology) with several human proteins. This supports the hypothesis of molecular mimicry: the immune system, while attempting to fight EBV, may produce antibodies that cross-react with the body's own tissues. This accidental autoimmune response could explain the chronic nature of symptoms in the infectious-onset subgroup.

Conclusion and Future Directions

This study demonstrates that EBV antibody responses can serve as a diagnostic tool for ME/CFS, provided that patient heterogeneity is accounted for through stratification. The identification of EBNA4_0529 and EBNA6_0070 as candidate biomarkers offers a promising path toward a standardized diagnostic test, such as an ELISA kit.

Future work will focus on validating these findings in a larger, prospective cohort, such as the UK ME/CFS Biobank, to confirm the clinical utility of these markers and further explore the role of molecular mimicry in disease progression.

Summary updated

Abstract

Abstract Infections by the Epstein-Barr virus (EBV) are often at the disease onset of patients suffering from Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS). However, serological analyses of these infections remain inconclusive when comparing patients with healthy controls. In particular, it is unclear if certain EBV-derived antigens eliciting antibody responses have a biomarker potential for disease diagnosis. With this purpose, we re-analysed a previously published microarray data on the IgG antibody responses against 3,054 EBV-related antigens in 92 patients with ME/CFS and 50 HCs. This re-analysis consisted of constructing different regression models for binary outcomes with the ability to classify patients and HCs. In these models, we tested for a possible interaction of different antibodies with age and gender. When analyzing the whole data set, there were no antibody responses that could be used to distinguish patients from healthy controls. A similar finding was obtained when comparing patients with noninfectious or unknown disease trigger to healthy controls. However, when data analysis was restricted to the comparison between HCs and patients with a putative infection at disease onset, we could identify stronger antibody responses against two candidate antigens (EBNA4_0529 and EBNA6_0070). Using antibody responses to these two antigens together with age and gender, the final classification model had an estimated sensitivity and specificity of 0.833 and 0.720, respectively. This reliable case-control discrimination suggested the use of the antibody levels related to these candidate viral epitopes as biomarkers for disease diagnosis in this subgroup of patients. When a bioinformatic analysis was performed on these epitopes, it revealed a potential molecular mimicry with several human proteins. To confirm these promising findings, a follow-up study will be conducted in a separate cohort of patients.
BibTeX
@inproceedings{sepulveda2022revisiting,
  author = {Nuno Sepúlveda and João Malato and Franziska Sotzny and Anna D. Grabowska and André Fonseca and Clara Cordeiro and Luís Graça and Przemysław Biecek and Uta Behrends and Josef Mautner and Francisco Westermeier and Eliana Lacerda and Carmen Scheibenbogen},
  title = {{Revisiting IgG antibody reactivity to Epstein-Barr virus in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome and its potential application to disease diagnosis}},
  booktitle = {Frontiers in Medicine},
  year = {2022},
  doi = {10.3389/fmed.2022.921101},
}
Figure