Acta Neurol Scand 2015: 131: 69–79 DOI: 10.1111/ane.12294

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA

Review Article

Targeting Epstein-Barr virus infection as an intervention against multiple sclerosis Jons D, Sundstr€ om P, Andersen O. Targeting Epstein-Barr virus infection as an intervention against multiple sclerosis. Acta Neurol Scand 2015: 131: 69–79. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. We here review contemporary data on genetic and environmental risk factors, particularly Epstein-Barr virus infection, for multiple sclerosis. There is an important immunogenetic etiological factor for multiple sclerosis. However, a general assumption is that immune defense genes are activated by the environment, basically by infections. We contend that the relationship between infectious mononucleosis and multiple sclerosis cannot be completely explained by genetics and inverse causality. Epstein-Barr infection as indicated by positive serology is an obligatory precondition for multiple sclerosis, which is a stronger attribute than a risk factor only. Data on events in the early pathogenesis of multiple sclerosis are cumulating from bio-banks with presymptomatic specimens, but there is only little information from the critical age when Epstein-Barr infection including infectious mononucleosis is acquired, nor on the detailed immunological consequences of this infection in individuals with and without multiple sclerosis. We discuss how focused bio-banking may elaborate a rationale for the development of treatment or vaccination against Epstein-Barr virus infection. A cohort in which intervention against Epstein-Barr infections was performed should be the object of neurological follow-up.

Introduction

Several ties between Epstein-Barr virus infection during infancy or adolescence and subsequent multiple sclerosis were reported and repeatedly reviewed (1–3). A relationship on an individual basis was strongly supported by i) similarities in the geographical distribution between infectious mononucleosis (IM) and MS on a local (4) and global (5) level, where IM is consistent with the age-related risk factor suggested by the classical migration studies (6). ii) a higher titer than in controls of anti-EBNA-1 antibodies, a marker of latent EBV infections, for several years before the onset (7), with seropositivity approaching 100% in patients with manifest MS as compared to approximately 95–96% in the general population (8, 9).

D. Jons1, P. Sundstr€om2, O. Andersen1 1 Section of Clinical Neuroscience and Rehabilitation, Institution of Neuroscience and Physiology, the Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; 2Section of Neurology, Department of Pharmacology and Clinical Neuroscience, University of Ume a, Ume a, Sweden

Key words: Multiple sclerosis; Epstein-Barr virus; neuroepidemiology; bio-banks; vaccination; risk factors O. Andersen, Gr€ona Str aket 11, Sahlgrenska University Hospital, 413 45 G€oteborg, Sweden Tel: + 46-31-342 9066 Fax: +46-31-826227 e-mail: [email protected] Accepted for publication July 22, 2014

iii) epidemiological evidence demonstrating a more than twofold increased risk of MS in persons with a history of infectious mononucleosis (IM) (10, 11). iv) statistical interaction of a history of IM or immunoreactivity to EBV antigens, markers of EBV infection, with the strongest genetic risk factor for MS, HLA DRB1*15 (12). The association indicates that EBV infection is a ‘risk factor’ or even a prerequisite for MS. It was proposed that large-scale vaccination against EBV infection, if feasible and successful, would prevent future MS (10, 13). However, a major concern is that it is largely unknown how EBV may influence the risk. We are facing the dilemma whether EBV based primary prevention should be started and evaluated based on the present incomplete evidence (14, 15) or whether 69

Jons et al. available resources are better devoted to further investigation of the pathophysiological role of multiple MS risk factors and interactions (16). Presymptomatic timing of the increased anti-EBNA-1 titers

The mean titers of antibodies to the nuclear EBV antigen EBNA1, indicating chronic infection with EBV, were reported to be higher in blood from MS patients prospectively collected before the onset, compared with matched controls (7). In individuals who later developed MS, antiEBNA-1 antibodies were similar to those of controls before the age of 20 years, but 2–3 times higher at age 25 years and older. A within-person analysis showed that a fourfold increase in antiEBNA complex antibody titers (between baseline at entry in the army and subsequent sampling) was associated with a threefold increase in risk of developing MS and that the individual EBV titers increased slowly over time (17). In a material with remarkably late onset of MS, anti-EBNA-1 antibody titers were increased up to 20 years before onset (18). A geographically based study confirmed the increase in anti-EBNA-1 antibody titers, which was highest in the last 5 years preceding onset of MS (9). In another prospective study, 10 of 305 samples tested for anti-EBNA-1 before MS onset were seronegative. However, a further follow-up specimen was available which revealed that these 10 individuals became EBNA1 positive on average 5 years before MS onset, with high titers soon after seroconversion (8). High anti-EBNA-1 titers in patients with the clinically isolated syndrome (CIS), an episode of focal neurology suggestive of MS and frequently retrospectively confirmed to be the onset attack, were predictive of conversion to MS (17, 19). None of these prospective studies provided data on the clinical correlate (IM?) to seroconversion. The risk of remaining EBV seronegative is extremely low in manifest MS. A meta-analysis of three studies selected for stringent analysis of EBV status reported that all MS patients had been infected with EBV. Stated differently, the pooled odds ratio (OR) for MS in EBV-seronegative individuals was 0.0 (95% confidence interval: 0– 0.43) (20). The ‘stringent’ criteria were that two methods gave negative results, enzyme-linked immunosorbent assay (ELISA) and indirect immunoflurescence and/or anticomplement immunofluorescence. Only one of the three studies included samples collected before the onset of MS, as discussed above (9). However, the requirement of two separate negative assays 70

considerably reduced the number of patients in the meta-analysis. Challenging the notion that MS patients are generally seropositive concerning anti-EBNA-1, eight % of pediatric MS patients were found to be seronegative. Related to the lower rate of seropositivity found in normal children, the OR for later MS in seronegative persons was the same in adult and pediatric MS cases (20). However, in another study, using ELISA, only 1/147 children with MS were negative for both EBV viral capsid antigen (VCA) and EBNA1 (21). There may be difficulties in the interpretation of these studies, as children considered to have MS may have a more heterogenous pathology, some cases presenting as an ADEM-like attack (22). To conclude, individuals who are not EBV infected have a very low, if any, risk of MS. Age-related incidence of primary EBV infection

Over 95% of adults worldwide are infected with EBV (23), apart from an unconfirmed report on low prevalence in Greenland inuits (24). Most of the children in developing countries are infected before 1 year of age, with >90% seropositivity in 5- to 9-year-old children, while the age of primary infection is delayed in western countries, with approximately 50% seropositivity in this age group (25, 26). There are two peaks of seroconversion, one between 1 and 6 years and the other at 14–20 years, with infection occurring later in high compared with low socioeconomic groups. According to a report from Hong Kong, 61% of infants in that area were EBV infected during the first 2 years of life (27). In the age group 1– 5 years 70% of Iranians were EBV-antibody positive in contrast to only 56% of Germans at the same age (28). In southern Italy, 82% were seropositive at age 8 to 10 (29). EBV infection occurs as infectious mononucleosis (IM) in a significant proportion of the adolescent population, with a peak incidence at 15–24 years of age (30), while IM is extremely rare in non-industrialized countries (26). IM is a dramatic event with massive proliferation of T cells, mainly cytotoxic CD8+, which are present in the blood and invade other tissues (31). Nevertheless, the subsequent lifelong infection is usually asymptomatic. CD4+ T cells and natural killer cells also proliferate in IM. In healthy people, EBNA1-specific CD4+ T cells are reported to control the EBV-induced B cell proliferation (32, 33). Twenty-five percentage of US college freshmen were seronegative and around 15% seroconverted yearly thereafter. Approximately 25% of adoles-

Epstein-Barr virus infection in multiple sclerosis cents who contracted the virus were reported to develop IM (34), although some investigators found a higher percentage of IM in seroconverters in this age group (45–65%)(35, 36). EBV type 1 was significantly overrepresented in individuals who developed IM compared with silent seroconversion (30). However, recombination events may confound classifications based on polymorphisms in the EBNA-1 gene (37). EBV infection may still occur in the middle age, although with modified symptoms, dominated by fever (38). The age at EBV seroconversion was reported to increase in Japan. The proportion of individuals in the population who seroconverted in the 5– 7 year group diminished from >80% to 59% during the 1990s. This was predicted to influence the incidence of EBV-associated disorders (25). If confirmed in other areas, it may contribute to the suspected contemporary increase in MS incidence. In a study based on hospital admissions in the UK, females were admitted for IM more frequently in the age group 10–19 years. The authors speculated that this female excess might play a role in determining the increased risk of MS in females (39). Increased risk of MS after IM

Originally shown in case–control (40) and cohort (41) studies, and proven beyond doubt in a historical prospective study, a history of verified IM increases the risk of later MS 2.3–2.8 times. In this study, the Danish MS registry was matched with centralized anti-EBV heterophile (Paul-Bunnel) antibody results from analyses requested from Danish clinicians upon suspicion of IM (11). The surplus risk appeared less than 5 years after IM, and remained elevated for three decades (42). In a smaller study based on diagnosis registers, an increased risk was confirmed 10 years after IM (43). Meta-analyses were confirmative (44, 45). Two large questionnaire studies, from a longitudinal Canadian database and a Dutch material confirmed the relationship with EBV but found no differences between cases and controls concerning a history of several common childhood infections or vaccinations (46, 47). Seroconversion in early childhood was not associated with the production of heterophile antibodies (48). At least 25% of specific seroconversions during college were associated with IM (49). As mentioned, sampling from IM cases in the pivotal study (11) was prompted by clinical suspicion, while the status of controls regarding acute seroconversions was unknown. Therefore, the risk of MS after seroconversion in

adolescence may have been underestimated. That MS may be related to an event which is hematologically identical to IM but sometimes clinically unnoticed seems to be a viable hypothesis (50). Questionnaire studies on the relationship between IM and MS are unreliable, as the clinical manifestations of IM have a low specificity, compared with other causes of sore throat (51). Bio-bank materials with a sufficient number of seroconversion events during a wide span of age in preclinical stages of MS compared with healthy controls are needed to evaluate the relative risk. The association with EBV infection is more fundamental for MS than other viral diseases

The hygiene hypothesis implicates that early childhood infections are protective clearly for allergic disorders and probably for some autoimmune diseases (3, 52), which might explain the increased EBNA1 titer if it resulted from a late EBV infection, often expressed as IM, known to be associated with MS. However, the ultimate consequence of this hypothesis would be a high MS risk in EBV-seronegative individuals – where the opposite is true. Yet, when seroconversion does occur in these individuals, the risk of MS is higher compared with seroconversion at a younger age (8). No other environmental factor is associated with such strong and abrupt change in risk of MS. A relationship between higher age at infection of measles and other common exanthemata and the risk of subsequent MS was reported (53). However, an authoritative study, based on a national MS register, did not reproduce these results concerning the childhood exanthemata (54). Although the age of IM did not influence the risk of MS (42, 47), EBV is the only agent for which the immunopathological details (specific vs polyclonal antibody production) determine the risk of subsequent MS. Interaction between EBV infection and genetics

At variance with the effects of remote infections, it has been suggested that MS may be a predominantly genetic disease. In a series of studies, the MS risk in adopted children, half-siblings, conjugal pairs and stepchildren was not different from that of the general population (55). The conclusion was that the action of the environment in influencing MS risk is operative at a population level. In a recent study based on national MS and twin registers (56), the heritability was estimated to 0.64, while the shared environmental component was 0 (0-0), or, including siblings and 71

Jons et al. half-siblings 0.01 (0-0.18). It was concluded that the shared (between patients) environmental factors was small, although there seemed to be a considerable non-shared environmental factor. However, as 95–97% of the adult population has experienced EBV infection, these data do not fit into a heritability analysis, which is based on variation in a population. One study in identical twins, using an immunodominant epitope as proxy for IM, suggested a higher IM frequency in the affected twin (57). The risk of MS was 10-fold increased in DRB1*15-positive individuals with a history of IM (58). It was confirmed that the risk of MS was higher in individuals with both the serological and the genetic risk factors anti-EBNA-1 antibodies and HLA DRB1*15, compared with the sum of each risk factor alone, a phenomenon that suggests a biological interaction. In other words, the increased risk for MS associated with HLA DRB1*15 relates to an augmented risk increase for symptomatic EBV infection (or antibody response against EBV) in persons with this HLA allele (12). Presymptomatic phase of MS

According to the prevailing view, pathological events start several years before the clinical onset of MS, probably during adolescence or even during childhood. When the first neurological symptoms occur (at approximately 30 years of age, on average), magnetic resonance imaging (MRI) usually shows several lesions, some of which have a chronic appearance. Fortuitous MRI findings suggestive of MS in asymptomatic individuals (the Radiologically Isolated Syndrome) were followed by onset of MS in one-third of the individuals after 5 years, and two-thirds showed radiological progression (59). Asymptomatic visual field disturbances have been detected at the time of MS onset. These disturbances showed subsequent slow progression over several years (60), which is suggestive of an extended presymptomatic phase during

which subtle functional disturbances accumulate. There are also case reports indicating that cerebrospinal fluid-enriched, oligoclonal immunoreactivity may appear up to 15 years before the clinical onset of MS (61). The average onset age of MS is 29 years, and the duration of preclinical pathology in MS may be years or decades. Classical migration studies have indicated varying windows of risk for etiological factors between 5 and 25 years of age (6). This large window in time is not inconsistent with the age span in which EBV infection tends to be associated with IM (8). Analysis of representative bio-bank material with population controls is required to obtain data on the sequence of events. Bio-banks of specimens obtained prior to MS onset lack samples from younger persons

Here, we review MS studies on presymptomatic samples (Table 1) from large bio-banks in the United States and Sweden. These studies all show increased antibody reactivity against EBNA1 in blood samples drawn prior to the onset of MS. The collected materials show a bias toward a higher onset age of MS than in population-based cohorts. A shortcoming of these studies is that samples were drawn more than a decade after adolescence, that is, probably more than 10 years after preclinical MS pathogenesis started in most cases. Therefore, more instructive data may be gained from studies of samples drawn closer to EBV seroconversion. An ongoing, large-scale survey of microbiological bio-banks in Sweden (DEMOS: Vitamin D and Epstein-Barr virus – Infectious Mononucleosis – in Multiple Sclerosis etiology) represents an attempt to obtain such samples; a large proportion of these samples were collected during childhood and adolescence, the period when EBV and other MS risk factors are believed to initiate the MS disease process. Risk estimates on environmental factors, such as smoke exposure and low

Table 1 Presymptomatic bio-bank studies on Epstein-Barr virus in multiple sclerosis Publication year, our comment 2001, 2004, 2005, 2006, 2011, 2013, a

First prospective MS bio-bank study (7) Several virus serologies analyzed (9) Age dependent EBV ab levels in MS (17) Elevated EBV abs >15 years before onset (18) EBV abs independent from 25(OH)-vitamin D levels (119) Inverse relation between EBV abs and 25(OH)-vitamin D levels in young subgroup (120)

N 18 73 83 42 222 192

Agea 50 28 24 32 23 26

(17–59) (17–39) (19–49) (16–41) (16–60)

Time before onsetb 2 6 4 15 5 9

(

Targeting Epstein-Barr virus infection as an intervention against multiple sclerosis.

We here review contemporary data on genetic and environmental risk factors, particularly Epstein-Barr virus infection, for multiple sclerosis. There i...
131KB Sizes 0 Downloads 8 Views