NEW IN SOLEIL RESEARCH™
A study published in Nature Medicine on September 16, 2026 reports that Epstein–Barr virus (EBV)-linked immune changes can be detected in peripheral blood before some clinical relapses of multiple sclerosis (MS). The work is notable because it combines longitudinal sampling with single-cell transcriptomics, bulk RNA sequencing, targeted viral RT-qPCR, flow cytometry and host-genetic analyses. The findings strengthen a biologically plausible hypothesis that EBV reactivation may participate in relapse-associated immune activation—but they do not establish that EBV causes relapse.
Plain-language summary
Researchers studied blood samples collected from people with relapsing-remitting MS, including rare samples obtained before a clinical relapse. They found changes in B cells and monocytes that appeared up to roughly three months before relapse and resembled host responses to EBV lytic reactivation. In additional analyses, the investigators detected higher EBV LMP-1 RNA in pre-relapse B cells and a small increase in B cells carrying a signal for the viral gp350 protein.
The study is important because the signal appeared before the clinical relapse and was supported by several different laboratory methods. However, this was an observational study. It cannot determine whether EBV reactivation triggers relapse, whether early inflammatory changes in MS permit EBV reactivation, or whether both processes reflect another underlying event. The results also do not show that antiviral treatment would prevent MS relapse.
Why this study matters
EBV has become central to modern research on MS risk, but risk of developing MS and the biology of an acute relapse are different questions. This study addresses the latter by asking whether EBV-related immune activity can be detected before a relapse becomes clinically apparent.
The investigators used longitudinal biospecimens from the CLIMB cohort at Brigham and Women’s Hospital. Fifteen participants contributed to the initial single-cell discovery analysis. Additional patients were evaluated with orthogonal methods, including bulk RNA sequencing, targeted viral RT-qPCR and multiparameter flow cytometry. This multimodal structure is one of the study’s most important strengths because the principal hypothesis did not rest on a single assay.
What the researchers found
Single-cell RNA sequencing identified a pre-relapse transcriptional program in monocytes and B cells. The program was enriched for host genes known to respond to EBV lytic-reactivation factors and was detectable within the months preceding clinical relapse.
Targeted viral RT-qPCR provided a more direct viral measure. Among paired samples that passed quality control, EBV LMP-1 transcripts were higher before relapse than during remission. The reported odds ratio was 1.50 per one-unit increase in 40−ΔCt, corresponding to approximately a twofold increase in transcript abundance (95% CI 1.26–1.93; adjusted P=0.0022).
Flow cytometry also identified a small but statistically significant increase in gp350-positive B cells before relapse. Across the modeled comparison, the odds ratio was 1.41 (95% CI 1.11–1.79; P=0.0043), with model-predicted frequencies of approximately 0.263% before relapse versus 0.186% in remission. The strongest association occurred in samples collected within 0–90 days before relapse.
An independent bulk RNA-sequencing analysis further supported the host-response signal. In the 0–90-day pre-relapse window, an LMP-1-responsive transcriptional signature was higher than during remission. The investigators also found overlap between pre-relapse gene modules, MS-associated genetic loci and enhancers bound by the EBV protein EBNA-2, providing an additional mechanistic bridge between host susceptibility and EBV-responsive immune programs.
Methodological quality
Overall assessment: moderate-to-strong evidence for a temporal association and for hypothesis generation; low certainty for causal inference.
The study has unusually good translational coherence for an observational immunology paper. Longitudinal sampling allowed the investigators to examine biology before relapse rather than only after symptoms appeared. The use of single-cell sequencing, bulk transcriptomics, targeted viral assays and flow cytometry reduces dependence on any one measurement platform. The authors also used mixed-effects models and multiple-testing correction in key analyses, and they did not treat technical replicates as independent biological observations.
At the same time, the strongest clinical claim supported by the design is that EBV-linked immune activity preceded and was associated with relapse. The design cannot establish direction of causality.
Strengths
- Prospective longitudinal cohort: samples collected before relapse provide temporal information that cross-sectional studies cannot.
- Orthogonal validation: the central signal was examined using single-cell RNA sequencing, bulk RNA sequencing, targeted RT-qPCR and flow cytometry.
- Paired sampling in key analyses: comparing pre-relapse and remission samples from the same individuals helps reduce between-person variability.
- Host-genetic integration: overlap with MS risk loci and EBNA-2-bound regulatory regions provides biologic context rather than relying only on viral detection.
- Statistical safeguards: mixed models and false-discovery-rate corrections were used across important comparisons.
Limitations
- Observational design: temporal precedence does not prove that EBV reactivation causes relapse.
- Small discovery and viral-validation samples: the initial single-cell cohort included 15 participants, and the targeted viral analyses were based on relatively small paired subsets.
- Very low viral RNA abundance: viral transcripts required preamplification, and single-cell RNA sequencing did not reliably capture EBV transcripts directly.
- Assay-specific uncertainty: the authors note that the gp350 flow-cytometry assay did not include Fc-receptor blocking or EBV-negative cell-line controls at the very low observed positivity rate, so cross-reactivity cannot be completely excluded.
- Alternative biological explanations remain plausible: generalized B-cell activation could facilitate EBV reactivation rather than EBV initiating the immune event. Subclinical CNS inflammatory activity could also precede both peripheral immune changes and clinical relapse.
- Specificity requires further work: B-cell and T-cell receptor clonotype analyses and additional controls for other B-lymphotropic herpesviruses would strengthen the mechanistic interpretation.
What this study does—and does not—show
The study supports the idea that EBV-linked peripheral immune activity can emerge before an MS relapse and therefore deserves prospective investigation as a biomarker and potential mechanistic contributor. It does not establish EBV reactivation as the cause of relapse, and it does not demonstrate that antiviral therapy or EBV-directed therapy prevents relapse.
Future studies should prospectively sample patients at high frequency, incorporate MRI and other measures of subclinical CNS activity, use stronger viral-specific controls, examine clonotypes and compare EBV signals with other herpesviruses. Interventional evidence would ultimately be needed to determine whether modifying EBV activity changes relapse risk.
Soleil Insight
The strongest signal here is temporal and multimodal, not causal. EBV-linked immune changes appeared before relapse across several complementary assays. That convergence elevates the hypothesis from a simple association to a testable mechanistic model—but prospective validation and stronger viral-specific controls are still required before prevention or treatment claims are justified.
Funding and conflicts of interest
The study reports support from the U.S. Department of Defense, the Water Cove Charitable Foundation, the Cindy Larsen-Chugg Chair, a Broad Institute Founders Grant and support from George and Sandra K. Schussel. The authors state that the funders had no role in study design, data collection or analysis, publication decisions or manuscript preparation.
Tanuja Chitnis reported research support from the Department of Defense, NIH, National Multiple Sclerosis Society, Genentech, Novartis, Octave Biosciences, Sanofi and Tiziana, and consulting or advisory relationships with Genentech, Lilly, Novartis, Octave, Sanofi and Tiziana. Howard L. Weiner reported research support from the Cure Alzheimer’s Fund, NIH and the National Multiple Sclerosis Society, and advisory-board roles with Sanofi and Tiziana. These disclosures do not invalidate the findings, but they are relevant context when evaluating the work.
Original study
King DA, Saxena S, Caefer D, et al. EBV reactivation priming of the peripheral immune system in multiple sclerosis relapse. Nature Medicine. Published September 16, 2026. DOI: 10.1038/s41591-026-04665-3. Original publication.
Featured image: colorized electron micrograph of three Epstein–Barr virus particles. Credit: NIAID; source via Wikimedia Commons. The image is identified by the source as public domain in the United States and also available under CC BY 2.0.
Advancing Science. Empowering Humanity.