Сотрясение связано с множественным склерозом, если оно происходит до инфекционного мононуклеоза.

Introduction

Epstein–Barr virus (EBV) has long been implicated in the aetiology of multiple sclerosis (MS), and more recent evidence suggests it is an essential risk factor for MS to develop. 1 , 2 As around over 95% of the adult population are infected by EBV at some point, 3 , 4 timing of infection, host genetics and other environmental factors are likely to be involved. 5 One of the mechanisms put forward for the influence of EBV is molecular mimicry between EBV nuclear antigen 1 (EBNA1)—an essential protein for EBV replication—and host antigens like anoctamin 2, alpha-B crystallin and GlialCAM. 6 , 7 Interestingly, Jons et al. demonstrated that EBV serological conversion was followed by reactivity to the molecular mimic anoctamin: a rise in the nerve damage marker neurofilament light chain was observed over a period of 10 years or longer. 8 However, to what degree timing of risk factor exposure in relation to EBV status may also play a role has not been explored in detail. Identifying the precise timing of EBV seroconversion in the general population is challenging, but a first episode of infectious mononucleosis (IM) during adolescence will typically represent the primary EBV infection. 9

Multiple studies in different populations suggest that traumatic brain injury (TBI) increases the risk of a subsequent diagnosis of MS. 10-12 Typically, these register-based studies have found that TBI detected during adolescence (ages 11–20 years) is linked with MS that often is diagnosed more than 10 years later. 10 Most of these events represent milder trauma defined as brain concussion that causes short-term impairment of neurological function not explained by structural injury on routine brain imaging. 13 , 14 However, in many cases, long-term alterations in the CNS structure or milieu can be detected with sensitive techniques such as with advanced magnetic resonance protocols or soluble biomarkers. 13 , 15 It is also known that brain trauma may lead to release of antigens that cause an adaptive immune response against CNS autoantigens that may influence TBI outcomes. 16 , 17 Interestingly, studies across several experimental animal models demonstrate that nerve injury alone, or combined with induction of experimental autoimmune encephalomyelitis (EAE), an animal model of MS, leads to infiltration of immune cells in CNS areas affected by the injury. 18 That the CNS environment has an impact on MS risk is also supported by genetic association studies identifying risk alleles implicating function of CNS resident cells. 5 In humans, even non-specific trauma to the CNS, such as following stroke and peripheral nerve damage, leads to expansion of nervous tissue-specific antigens detectable in peripheral blood; and at the same time, affected brain areas become susceptible to infiltration of antigen specific T cells. 19

This longitudinal cohort study used Swedish register data to assess whether there was a difference in the association with MS with onset after age 20 years, if concussion occurred before or after IM between ages 11 and 20 years among a population of people who had all experienced IM (as a proxy measure of primary EBV infection) in the same age-defined period.

Materials and methods

Ethical permission

Ethical permission for this study was obtained from the Swedish Ethical Review Authority (2019-04755, 2023-03585-02 and 2025-03176-01). All data are pseudonymized.

All Swedish residents, born between 2 January 1980 and 30 December 2002, with follow-up to 31 December 2023 and living in Sweden until at least age 21 years, constituted the population from which the analytical sample was drawn, using information from national registers. The Total Population Register provided information on sex, region of residence at birth (24 counties) and dates of birth, death and migration. The Patient Register provided information on inpatient and outpatient hospital diagnoses of MS, other demyelinating diseases of the CNS, IM and concussion using the 9th and 10th revisions of the International Classification of Diseases (ICD). Further information on IM and concussion diagnoses was obtained from primary care data from the county of Västra Götaland (∼10% of the Swedish population), with diagnoses identified by ICD-10 codes. IM was identified using 075 for ICD-9 and B27.0 for ICD-10. MS was recorded with codes 340 and G35 for ICD-9 and ICD-10, respectively. All demyelinating diseases of the CNS were identified using codes 340 and 341 for ICD-9 and G04, G35, G36 and G37 for ICD-10. Concussion was recorded as 850 in ICD-9 and S06 in ICD-10.

The main analysis included only individuals with a diagnosis of IM between ages 11 and 20 years ( n = 37 432), after excluding individuals a demyelinating disease of the CNS diagnosed by age 20 years.

Statistical analysis

Summary statistics of the variables used in our analyses are presented as counts and percentages for categorical variables, medians and interquartile ranges for continuous variables. Statistical significance of differences was assessed using Pearson’s χ ² test for categorical variables and the Wilcoxon rank sum test for medians. For the two main exposures of interest, concussion before IM (no, yes) and concussion after IM (no, yes), we quantified the associations with MS using two separate univariable Cox regression models. We then fitted a multivariable Cox regression model, which included both exposures, as well as the pre-specified adjustment covariates. These were sex, county of birth, year of birth and age at first IM. The latter two variables were modelled as continuous using restricted cubic splines with four knots. We also included a shared frailty term for birth county (24 counties) to account for unobserved heterogeneity across regions. Attained age was the underlying timescale, thus adjusting for age effectively. Follow-up was from age 21 years to first diagnosis of MS, emigration or death, whichever occurred first. The proportional hazards assumption was evaluated using the Grambsch–Therneau test and by graphical inspection of scaled Schoenfeld residuals: no violation was identified. Interaction testing was used to assess effect modification for concussion before and after IM.

Instead of using the two exposure variables, a separate model used a single variable with mutually exclusive categories: concussion before IM only, concussion after IM only and concussion both before and after IM. A further analysis dichotomized the concussion variables to characterize concussion in terms of relative timing before or after IM. The four variables produced were concussion up to 1 year before IM, concussion up to 1 year after IM, concussion more than 1 year before IM and concussion more than 1 year after IM. The adjusted analysis included all four variables in the same model, with adjustment for the potential confounding factors.

Bayes factors were calculated 20 to assist in interpretation of evidence strength. The statistical analysis was performed using R version 4.5.0 particularly using the ‘survival’ and ‘tidyverse’ packages.

Results

Table 1 shows the characteristics of the study population, all of whom had IM between ages 11 and 20 years and did not have a diagnosis of any CNS demyelinating disease before age 21 years. As expected, there is a greater female predominance among those with MS. On average, those with MS were born a few years earlier than the rest of this population, reflecting the typical ages at diagnosis of MS. The median age at IM was less than half a year younger among those with MS, compared with the rest of the study population.

Table 2 shows the associations with MS diagnosed after age 20 years for concussion that occurred before or after IM between ages 11 and 20 years. Concussion before IM is statistically significantly associated with a raised risk of subsequent MS, with a Bayes factor for the adjusted model of 9.524 indicating ‘moderate/substantial evidence’ (range 3–10). Concussion after IM is not associated with MS. These associations are observed before and after adjustment for both measures of concussion, as well as the other covariates: the magnitude of the hazard ratio for concussion before IM increases slightly after adjustment but there is no notable change in the null association for concussion after IM. However, there is no statistically significant effect modification for MS with an interaction for concussion before and after IM of 1.60 (0.16–15.60; P = 0.685), indicating the possibility that concussion after IM influences MS risk cannot be excluded. It is notable that adjustment includes age at IM (modelled as a continuous variable), so the association is not explained by differences in age at IM and concussion before or after IM. Table 3 reports a similar analysis, but using a single concussion variable with mutually exclusive categories. Concussion before IM only was statistically significantly associated with MS, and the Bayes factor for the adjusted model indicates ‘moderate/substantial evidence’ (range 3–10). There is also a raised risk for MS associated with concussion before and after IM, but it is not statistically significant. Concussion after IM was not associated with a raised risk of MS.

Table 4 shows the association of concussion with MS risk when the duration between concussion and MS is examined. Concussion less than 1 year before IM has a statistically significant and higher magnitude association with MS than seen in the results of the main analysis ( Table 2 ) and the Bayes factor signals ‘moderate/substantial evidence’ (range 3–10) in the adjusted analysis. Concussion more than 1 year before IM has a borderline statistically significant association with subsequent MS and the Bayes factor below 3 indicates ‘anecdotal evidence’ in the adjusted analysis. Neither measure of concussion after IM was associated with MS.

Discussion

This national register-based cohort study found that concussion before IM during adolescence is a risk factor for subsequent MS, while concussion that occurred after IM is not, although interaction testing suggests the lack of association for concussion after IM should be interpreted with caution. There is growing evidence that EBV is critical in the aetiology of MS, particularly where primary EBV infection is delayed which increases the risk of IM. 21 We hypothesize that sequelae of exposures such as concussion can create conditions in the CNS that further increases the risk that an atypical EBV infection results in MS pathogenesis. Concussion up to 1 year before IM had a more notable association with MS than concussion more than a year before IM. After EBV seroconversion, some subsequent environmental exposures may not have the same consequences for MS risk, and this may vary by type of exposure. 22

Concussion was selected as the exposure that might alter susceptibility to EBV infection as it is practical to identify the precise timing of concussion diagnoses relative to IM. Other known risk factors include smoking, body mass index and vitamin D level, 22 , 23 but it is difficult to define precisely the timing of such exposures as this can occur over an extended period. Infectious diagnoses other than IM are another set of potentially important exposures relevant to MS risk, but these may not be independent of each other as one infection can increase the risk of another 24 , 25 making it harder to assess potential causation associated with the temporal sequence.

It has been argued that the association of IM with MS is not causal 26 as human leukocyte antigen ( HLA ) regions such as HLA-DRB1 and HLA-DQA1 influence both MS risk and anti-EBV nuclear antigen 1 IgG levels 27-29 ; thus, IM may signal MS risk rather than a causal exposure. However, the results from a sibling comparison study are consistent with a causal association between IM and MS. 30 There is compelling evidence that EBV causes MS: a study reported a 32-fold increased risk of MS following EBV infection, but not other infections, as shown convincingly by a study. 2 However, some other infections like HHV6A and Rubella have modest magnitude associations with MS risk. 31 High levels of EBV nuclear antigen-1-specific antibodies and IM are both associated with MS risk independently and synergistically. 32 A study found further evidence that, before EBV seroconversion, there was no sign of neuroaxonal degeneration (based on levels of neurofilament light chain) indicating that the EBV infection must occur before initiation of MS pathogenesis. 2 This is consistent with other studies 8 and our findings, where concussion that occurs before likely EBV seroconversion may influence EBV-related MS risk. The mechanism by which EBV infection causes MS is thought to involve molecular mimicry for EBV transcription factor EBNA 1 and the GlialCAM CNS protein and thus autoimmunity resulting in demyelination, 1 as well as the involvement of CNS proteins alpha-B crystallin (CRYAB), anoctamin-2 (ANO2), and glial cellular adhesion molecule (GlialCAM). 33 The delay between EBV infection and diagnosis of MS is consistent with research showing that EBV seroreactivity was present 10 years before the first sign of neuroaxonal injury prior to clinical MS onset. 8 This damage was associated with anoctamin 2 (a protein that may cross-react) antibodies produced around the time of EBV infection. 8 Other studies have also demonstrated an average duration of 10–20 years between primary EBV infection and clinical MS onset. 34 , 35

It is unlikely that concussion in adolescence is a consequence of prodromal MS, as a previous study demonstrated an association of concussion in adolescence with subsequent MS, but not for non-CNS trauma. 10 Also, reverse causation seems unlikely as only concussion before IM was associated with MS. An interaction between HLA type and concussion for MS risk has been demonstrated, 12 and possible explanations include HLA -related influences on recognition or tolerance of CNS antigens resulting from the trauma. 36 , 37 Mild TBI may establish an environment in the CNS that facilitates MS pathogenesis when infectious mononucleosis occurs through several putative mechanisms. Concussion can result in immune dysregulation 38 that could potentially alter the immune response to EBV and the course of the infection. Some sequelae of concussion can persist for months or even years in a subset of patients, with the possibility that a chronic neuroinflammatory state is established. 13 , 15 , 39 Thus, for an extended period after concussion, some characteristics may remain that cause heightened susceptibility to an atypical course for an EBV infection among genetically susceptible members of the population.

There have been secular trends in the incidence of MS in Nordic countries, 40 so we adjusted for year of birth to improve precision. There is also variation in incidence by region within countries, 40 so we adjusted for county of birth. This can also account for variation in healthcare and differences in timing of diagnosis between regions. Our use of age as the underlying time scale in Cox regression provided effective adjustment for age and follow-up time. We did not adjust for other MS risk factors as our focus was the relative timing of IM and concussion between ages 11 and 20 years: it is unlikely confounding could notably explain such a temporal sequence during this limited age range. We did not adjust for exposures after age 20 years, as we speculate that some environmental exposures after age 20 years, in individuals who already had IM, are less relevant to the initiation of pathogenesis but may only accelerate subclinical disease progression: for example, infections in people with diagnosed MS can cause exacerbations and disease progression, 41 so there is likely a similar influence on subclinical disease before MS is diagnosed.

A large longitudinal cohort with prospectively recorded diagnoses made it possible to identify the timing of likely first EBV exposure, as well as the relative timing of concussion, but there are some potential limitations. Only ICD-10 identifies IM specifically caused by EBV, but there was insufficient follow-up time and statistical power (few outcome events) to limit our study to IM identified using ICD-10, so we also used earlier ICD versions. However, EBV is the most common cause of IM 42 and it seems implausible that findings on the relative timing of concussion and IM could be due to other less common causes of IM. It could be argued that the reason for a null finding for concussion after IM is because it is more likely to occur later in adolescence and the age-specific risk for MS associated with concussion may vary. However, as we adjusted for age at IM and this did not alter the results notably, differences in age at concussion relative to IM during adolescence do not explain the findings. This is consistent with a previous study that demonstrated similar risks for MS if concussion during adolescence was before or after age 16 years. 11 There were somewhat fewer concussion diagnoses after IM and there was no statistically significant effect modification for concussion before or after IM, so it is possible that the lack of association for concussion after IM could be due to limited power. The association of concussion with MS is dose-dependent, 10 , 12 but we did not have sufficient power to examine this, so we coded the exposures as concussion yes or no, rather than the number of exposures. The cohort members all had a diagnosis of IM during adolescence, so other patterns of association may exist among those who did not have IM in this age range. This is the first report of these associations, but there were only a relatively small number of outcome events (MS diagnoses) and thus limited statistical power, as our study population was limited to the subset with IM documented in healthcare records. The results are statistically significant and biologically plausible, but a chance finding is always possible, so replication is important.

In conclusion, concussion-related inflammation may alter the CNS environment, increasing the possibility that a subsequent primary EBV infection triggers an inappropriate immune response against the CNS, leading to development of MS among susceptible individuals.

Supplementary Material

Supplementary material

Supplementary material is available at Brain Communications online.

Funding

This project was supported by Nyckelfonden (OLL-984743 and OLL-1013159).

Competing interests

S.M. has received multiple sclerosis–related research grants and/or honoraria for advisory boards/lectures from Roche, Novartis, AstraZeneca, Merck, Teva and IQVIA. L.A. has received lecture/honoraria from Merck, Novartis and Teva. F.P. has received research grants from Denka, Janssen, Merck KGaA, Novartis, Pfizer and UCB, fees for serving on data monitoring committees in clinical trials with Lundbeck and Roche and preparation of expert witness statement for Novartis. T.O. has received lecture/advisory board honoraria from Merck, Novartis and Sanofi and unrestricted MS research grants from the same companies. He has received academic funding from the Swedish Research Council, the Swedish Brain Foundation, the Knut and Alice Wallenberg foundations and Margaretha af Ugglas Foundation. The other authors report no competing interests.

Data availability

The data used in this study cannot be shared publicly due to regulations under Swedish law. Inquiries about the data used in the present study can be addressed to the corresponding author. Researchers who are interested in Swedish register data can refer to https://dataguiden.se/english . The syntax used for the statistical analysis is available as Supplementary material .

References

1 Lanz TV , Brewer RC , Ho PP , et al Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM . Nature . 2022 ; 603 : 321 – 327 . 35073561 10.1038/s41586-022-04432-7 PMC9382663

2 Bjornevik K , Cortese M , Healy BC , et al Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis . Science . 2022 ; 375 ( 6578 ): 296 – 301 . 35025605 10.1126/science.abj8222

3 Karachaliou M , Vidal M , Pembrey L , et al Seroepidemiology of maternal and childhood pathogen exposure in three European mother-child cohorts . Int J Infect Dis . 2025 ; 159 : 107994 . 40701399 10.1016/j.ijid.2025.107994

4 Rahman S , Wathington D , Waterboer T , et al Seroprevalence of Chlamydia trachomatis, herpes simplex 2, Epstein-Barr virus, hepatitis C and associated factors among a cohort of men ages 18–70 years from three countries . PLoS One . 2021 ; 16 ( 6 ): e0253005 . 34157055 10.1371/journal.pone.0253005 PMC8219163

5 Consortium IMSG . Multiple sclerosis genomic map implicates peripheral immune cells and microglia in susceptibility . Science . 2019 ; 365 ( 6460 ): eaav7188 . 31604244 10.1126/science.aav7188 PMC7241648

6 Tengvall K , Huang J , Hellström C , et al Molecular mimicry between Anoctamin 2 and Epstein-Barr virus nuclear antigen 1 associates with multiple sclerosis risk . Proc Natl Acad Sci U S A . 2019 ; 116 ( 34 ): 16955 – 16960 . 31375628 10.1073/pnas.1902623116 PMC6708327

7 Thomas OG , Bronge M , Tengvall K , et al Cross-reactive EBNA1 immunity targets alpha-crystallin B and is associated with multiple sclerosis . Sci Adv . 2023 ; 9 ( 20 ): eadg3032 . 37196088 10.1126/sciadv.adg3032 PMC10191428

8 Jons D , Grut V , Bergström T , et al Seroreactivity against lytic, latent and possible cross-reactive EBV antigens appears on average 10 years before MS induced preclinical neuroaxonal damage . J Neurol Neurosurg Psychiatry . 2024 ; 95 ( 4 ): 325 – 332 . 37802637 10.1136/jnnp-2023-331868 PMC10958269

9 Dunmire SK , Hogquist KA , Balfour HH . Infectious mononucleosis . Curr Top Microbiol Immunol . 2015 ; 390 ( Pt 1 ): 211 – 240 . 26424648 10.1007/978-3-319-22822-8_9 PMC4670567

10 Montgomery S , Hiyoshi A , Burkill S , Alfredsson L , Bahmanyar S , Olsson T . Concussion in adolescence and risk of multiple sclerosis . Ann Neurol . 2017 ; 82 ( 4 ): 554 – 561 . 28869671 10.1002/ana.25036

11 Montgomery S , Hiyoshi A , Burkill S , Alfredsson L , Bahmanyar S , Olsson T . Reply to “concussion may not cause multiple sclerosis” . Ann Neurol . 2017 ; 82 ( 4 ): 652 – 653 . 28976595 10.1002/ana.25062

12 Johansson E , Alfredsson L , Strid P , Kockum I , Olsson T , Hedström AK . Head trauma results in manyfold increased risk of multiple sclerosis in genetically susceptible individuals . J Neurol Neurosurg Psychiatry . 2024 ; 95 ( 6 ): 554 – 560 . 38212058 10.1136/jnnp-2023-332643 PMC11103305

13 Richter S , Winzeck S , Correia MM , et al Predicting recovery in patients with mild traumatic brain injury and a normal CT using serum biomarkers and diffusion tensor imaging (CENTER-TBI): An observational cohort study . EClinicalMedicine . 2024 ; 75 : 102751 . 39720677 10.1016/j.eclinm.2024.102751 PMC11667275

14 Silverberg ND , Iverson GL , Cogan A , et al The American congress of rehabilitation medicine diagnostic criteria for mild traumatic brain injury . Arch Phys Med Rehabil . 2023 ; 104 ( 8 ): 1343 – 1355 . 37211140 10.1016/j.apmr.2023.03.036

15 Friberg S , Lindblad C , Zeiler FA , et al Fluid biomarkers of chronic traumatic brain injury . Nat Rev Neurol . 2024 ; 20 ( 11 ): 671 – 684 . 39363129 10.1038/s41582-024-01024-z

16 Korn T , Kallies A . T cell responses in the central nervous system . Nat Rev Immunol . 2017 ; 17 ( 3 ): 179 – 194 . 28138136 10.1038/nri.2016.144

17 Needham EJ , Stoevesandt O , Thelin EP , et al Complex autoantibody responses occur following moderate to severe traumatic brain injury . J Immunol . 2021 ; 207 ( 1 ): 90 – 100 . 34145056 10.4049/jimmunol.2001309 PMC7616895

18 Hammarberg H , Lidman O , Lundberg C , et al Neuroprotection by encephalomyelitis: Rescue of mechanically injured neurons and neurotrophin production by CNS-infiltrating T and natural killer cells . J Neurosci . 2000 ; 20 ( 14 ): 5283 – 5291 . 10884312 10.1523/JNEUROSCI.20-14-05283.2000 PMC6772346

19 Wang WZ , Olsson T , Kostulas V , Höjeberg B , Ekre HP , Link H . Myelin antigen reactive T cells in cerebrovascular diseases . Clin Exp Immunol . 1992 ; 88 ( 1 ): 157 – 162 . 1373351 10.1111/j.1365-2249.1992.tb03056.x PMC1554378

20 Rostgaard K . Simple nested Bayesian hypothesis testing for meta-analysis, Cox, Poisson and logistic regression models . Sci Rep . 2023 ; 13 ( 1 ): 4731 . 36959371 10.1038/s41598-023-31838-8 PMC10036629

21 Rostgaard K , Nielsen NM , Melbye M , Frisch M , Hjalgrim H . Siblings reduce multiple sclerosis risk by preventing delayed primary Epstein-Barr virus infection . Brain . 2023 ; 146 ( 5 ): 1993 – 2002 . 36317463 10.1093/brain/awac401

22 Alfredsson L , Olsson T . Lifestyle and environmental factors in multiple sclerosis . Cold Spring Harb Perspect Med . 2019 ; 9 ( 4 ): a028944 . 29735578 10.1101/cshperspect.a028944 PMC6444694

23 Vitturi BK , Cellerino M , Boccia D , et al Environmental risk factors for multiple sclerosis: A comprehensive systematic review and meta-analysis . J Neurol . 2025 ; 272 ( 8 ): 513 . 40664781 10.1007/s00415-025-13248-0 PMC12263771

24 Al-Aly Z , Xie Y , Bowe B . High-dimensional characterization of post-acute sequelae of COVID-19 . Nature . 2021 ; 594 ( 7862 ): 259 – 264 . 33887749 10.1038/s41586-021-03553-9

25 Moon S , Han S , Jang IH , et al Airway epithelial CD47 plays a critical role in inducing influenza virus-mediated bacterial super-infection . Nat Commun . 2024 ; 15 ( 1 ): 3666 . 38693120 10.1038/s41467-024-47963-5 PMC11063069

26 Burnard S , Lechner-Scott J , Scott RJ . EBV and MS: Major cause, minor contribution or red-herring? Mult Scler Relat Disord . 2017 ; 16 : 24 – 30 . 28755681 10.1016/j.msard.2017.06.002

27 Kreft KL , Van Nierop GP , Scherbeijn SMJ , Janssen M , Verjans GMGM , Hintzen RQ . Elevated EBNA-1 IgG in MS is associated with genetic MS risk variants . Neurol Neuroimmunol Neuroinflamm . 2017 ; 4 ( 6 ): e406 . 29379819 10.1212/NXI.0000000000000406 PMC5778394

28 Zhou Y , Zhu G , Charlesworth JC , et al Genetic loci for Epstein-Barr virus nuclear antigen-1 are associated with risk of multiple sclerosis . Mult Scler . 2016 ; 22 ( 13 ): 1655 – 1664 . 26819262 10.1177/1352458515626598

29 Rubicz R , Yolken R , Drigalenko E , et al A genome-wide integrative genomic study localizes genetic factors influencing antibodies against Epstein-Barr virus nuclear antigen 1 (EBNA-1) . PLoS Genet . 2013 ; 9 ( 1 ): e1003147 . 23326239 10.1371/journal.pgen.1003147 PMC3542101

30 Xu Y , Hiyoshi A , Smith KA , et al Association of infectious mononucleosis in childhood and adolescence with risk for a subsequent multiple sclerosis diagnosis among siblings . JAMA Netw Open . 2021 ; 4 ( 10 ): e2124932 . 34633426 10.1001/jamanetworkopen.2021.24932 PMC8506233

31 Ingvarsson J , Grut V , Biström M , et al Rubella virus seropositivity after infection or vaccination as a risk factor for multiple sclerosis . Eur J Neurol . 2024 ; 31 ( 10 ): e16387 . 39023088 10.1111/ene.16387 PMC11414808

32 Hedström AK , Huang J , Michel A , et al High levels of epstein-barr virus nuclear antigen-1-specific antibodies and infectious mononucleosis act both independently and synergistically to increase multiple sclerosis risk . Front Neurol . 2019 ; 10 : 1368 . 32038456 10.3389/fneur.2019.01368 PMC6992610

33 Sattarnezhad N , Kockum I , Thomas OG , et al Antibody reactivity against EBNA1 and GlialCAM differentiates multiple sclerosis patients from healthy controls . Proc Natl Acad Sci U S A . 2025 ; 122 ( 11 ): e2424986122 . 40063790 10.1073/pnas.2424986122 PMC11929495

34 Levin LI , Munger KL , Rubertone MV , et al Temporal relationship between elevation of epstein-barr virus antibody titers and initial onset of neurological symptoms in multiple sclerosis . JAMA . 2005 ; 293 ( 20 ): 2496 – 2500 . 15914750 10.1001/jama.293.20.2496

35 Ascherio A , Munger KL , Lennette ET , et al Epstein-Barr virus antibodies and risk of multiple sclerosis: A prospective study . JAMA . 2001 ; 286 ( 24 ): 3083 – 3088 . 11754673 10.1001/jama.286.24.3083

36 Wang J , Jelcic I , Mühlenbruch L , et al HLA-DR15 molecules jointly shape an autoreactive T cell repertoire in multiple sclerosis . Cell . 2020 ; 183 ( 5 ): 1264 – 1281.e20 . 33091337 10.1016/j.cell.2020.09.054 PMC7707104

37 Friese MA , Jakobsen KB , Friis L , et al Opposing effects of HLA class I molecules in tuning autoreactive CD8+ T cells in multiple sclerosis . Nat Med . 2008 ; 14 ( 11 ): 1227 – 1235 . 18953350 10.1038/nm.1881

38 Roth S , Cao J , Singh V , et al Post-injury immunosuppression and secondary infections are caused by an AIM2 inflammasome-driven signaling cascade . Immunity . 2021 ; 54 ( 4 ): 648 – 659.e8 . 33667383 10.1016/j.immuni.2021.02.004

39 Hadanny A , Efrati S . Persistent post-concussion syndrome: Pathophysiology, diagnosis, current and evolving treatment strategies . Expert Rev Neurother . 2025 ; 25 ( 8 ): 959 – 971 . 40481820 10.1080/14737175.2025.2515061

40 Sumelahti ML , Tienari PJ , Hakama M , Wikström J . Multiple sclerosis in Finland: Incidence trends and differences in relapsing remitting and primary progressive disease courses . J Neurol Neurosurg Psychiatry . 2003 ; 74 ( 1 ): 25 – 28 . 12486261 10.1136/jnnp.74.1.25 PMC1738192

41 Loebermann M , Winkelmann A , Hartung HP , Hengel H , Reisinger EC , Zettl UK . Vaccination against infection in patients with multiple sclerosis . Nat Rev Neurol . 2012 ; 8 ( 3 ): 143 – 151 . 22270022 10.1038/nrneurol.2012.8

42 Dunmire SK , Verghese PS , Balfour HH . Primary epstein-barr virus infection . J Clin Virol . 2018 ; 102 : 84 – 92 . 29525635 10.1016/j.jcv.2018.03.001