What causes multiple sclerosis? Getting closer to the answers

MS risk is becoming less of a mystery as we gain a better understanding of its causes

Multiple sclerosis (MS) is a complex neuroinflammatory/neurodegenerative disease of the central nervous system. Recent work1,2 has significantly advanced our understanding of the aetiology of MS, emphasising the importance of infection with Epstein–Barr virus as a significant driver of MS risk. MS manifests clinically as neurological dysfunction affecting any area of the central nervous system, in particular the optic nerves, spinal cord, and brainstem. Most people with MS (90%) present with relapse onset MS where episodes of neurological dysfunction are followed by partial or full recovery but over time, disability almost invariably accumulates (secondary progressive MS).3 In contrast, 10% of people present with progression from disease onset, termed progressive onset MS.3 In the developed world, MS is one of the leading causes of neurological disability in young adults, and as the median age of onset in Australia is between 35 and 40years,4 it affects people in their most productive years. Consequently, the economic and social costs to individuals and the wider community are high. In 2017, we estimated the cost of MS to Australia was $1.7 billion annually and increasing significantly each year.5 Currently, there is no cure for MS, but significant gains have been made in its treatment, particularly the use of highly effective disease‐modifying therapies and the development of comprehensive care through specialised MS clinics. However, no current treatment can stop or reverse the neurodegenerative component of the disease.3

MS is considered a complex disease due to its significant heterogeneity in progression and clinical course, its protean clinical manifestations, and its significant intra‐ and inter‐individual variability in clinical course.3 It also has a complex aetiology and there have been multiple theories postulated as to its cause, mainly related to infections and toxin exposures.6 What is clear is that there is no single cause of MS; rather, there are many steps along a causal pathway, starting with genetics and sex, and including environmental, modifiable lifestyle, and infectious factors. Some of these factors may be obligate, such as infection with Epstein–Barr virus,1 while others may be person‐specific and also time‐dependent with periods of vulnerability, often in adolescence, where risk factors such as obesity and infectious mononucleosis may have their greatest effects.7

A useful concept to understand complex disease aetiology is Rothman’s causal pie.8 To develop a disease, an individual must fill a pie dish with slices that represent risk factors to that individual. Only when the pie dish is full (causal sufficiency) will the individual be at greatest risk of developing the disease, although disease manifestation is not inevitable. Each individual will have different slices and different sizes of the slices in their personal pie dish. Epstein–Barr virus infection will likely be a slice in every pie,1 but of different sizes. The timing of exposure will also affect the size of each slice. Finally, factors can interact to either increase or decrease the effect size.7

MS has a well established sex difference in prevalence, with females three times more likely to develop MS than males3 and the difference possibly increasing with time.9 The mechanism of this difference is not known but may reflect genetics or the differential effects of other risk factors by sex. Significant research has looked at the effect of age at menarche, age at first pregnancy and fecundity as risk factors for MS and to date no clear patterns have emerged.10

Genetics play a clear role in the aetiology of MS, with significant enrichment of MS cases in families and increasing risk with increasing relatedness. The concordance rates for monozygotic twins are 20–30%, indicating that the risk of developing MS is at least 25% genetic.11

Large scale studies of MS genetics have identified 233 common risk variants12 for MS, with the strongest associations seen for alleles within the human leucocyte antigen region, particularly HLA‐DRB1*15:01 alleles which increase MS risk 2‐3‐fold.12 However, the explained portion of MS heritability by discovered loci is only between 25–40%,12 indicating significant unexplained heritability for which there are many theories.

It has been known for many years that the risk of developing MS is not uniform around the world.13 People of European ancestry14 are significantly more likely to develop MS compared with those of African and East Asian ethnicity.14 Similarly, admixed populations, as in New Zealand, demonstrate that the prevalence among Pacifica peoples is significantly lower than the risk for Europeans in the same environments.14,15

A clear latitudinal gradient of MS risk is present in most areas and appears to be strengthening with time.16 That is, MS prevalence incidence and mortality rates increase with increasing distance from the equator. As the latitudinal gradient is seen in most ethnic groups and across most latitudes, it likely represents an environmental risk factor acting at the population level rather than at the individual level.16 The most likely candidates driving the gradient are low sunlight exposure and low vitamin D levels,17 with strong correlations between latitude, MS prevalence and sunlight exposure.18 Recent investigations have also shown that timing is important, with work from NZ showing that the gradient is strongest at or near birth.19 There is strong support for both direct effects of low sunlight (and in particular ultraviolet B radiation) exposure and vitamin D deficiency as risk factors for the development of MS.17 Vitamin D levels may be a proxy for sunlight exposure as up to 95% of circulating vitamin D is produced by sunlight exposure.20

Epstein–Barr virus is a ubiquitous human herpesvirus that infects around 90% of the adult population.1 Infection can occur at any age but is most common in childhood. Adolescent and adult infections can result in infectious mononucleosis or glandular fever. In MS, more than 99% of cases have evidence of prior exposure to Epstein–Barr virus, and there is a clear risk association with prior infectious mononucleosis.1 Recent work from the United States1 has established that Epstein–Barr virus infection is likely to be a requisite factor for the development of MS, with all but one prospectively diagnosed MS case (n = 955) in a multimillion‐strong military cohort either seroconverting before diagnosis or already being Epstein–Barr virus positive, providing strong evidence for the virus having an obligate role in MS causation. Other recent work has demonstrated an aetiological link between Epstein–Barr virus and MS, based on molecular mimicry between Epstein–Barr nuclear antigen 1 and the central nervous system protein glial cell adhesion molecule (GlialCAM).2 GlialCAM is expressed in the central nervous system by astrocytes and oligodendrocytes, providing a clear link to glial cells as an important mediator of MS pathology and affirming the importance of clonal intrathecal Epstein–Barr virus specific B cell populations in MS.2 However, other factors are critical in modifying an individual’s response to Epstein–Barr virus infection and MS risk, as clearly the risk of developing MS with Epstein–Barr virus infection is very low, with 90% seropositivity in the community and an overall risk of developing MS of 1:1000 in Australia.21

The effect of modifiable risk factors in the onset of MS is of considerable importance as interventions aimed at reducing or mitigating these factors may have significant effects in reducing MS incidence. Two such factors have been determined in MS: tobacco smoking and adolescent obesity.22 Both factors increase the risk of developing MS 1.5‐2‐fold and potentially, changes in the prevalence of these factors may have influenced the increase in MS incidence seen in Australia and other countries. The timing of these factors points to a window of opportunity in adolescence and early adulthood where these factors could be mitigated, particularly in those at greater risk. It is likely that the current obesity epidemic associated with decreased physical activity resulting in lower exposure to sunlight may drive increasing MS incidence over the next decades.

Overall risk factors for MS do not operate in isolation and many interactions between risk factors have been described, particularly for smoking, Epstein–Barr virus, and genetics.7 Considering multiple factors in a model explained up to 80% of the risk of developing a first episode of central nervous system demyelination in an Australian cohort.23 Including interactions and better measures of Epstein–Barr virus exposure as well as adolescent obesity in this model may further increase the explained population attributable risk.

MS used to be referred to as a mystery illness but now MS risk is less of a mystery as we gain a better understanding of its causes. It is now clear that there is no single cause of MS and that there is no “smoking gun” risk factor out there yet to be found. We now recognise there is a chain of risk factors stretching back to conception with each factor an important step on the causal pathway. Some risk factors are immutable, such as sex and genetics, but others are potentially modifiable, such as obesity, smoking, vitamin D levels and ultraviolet light exposure, and potentially, Epstein–Barr virus immunisation. The recognition that Epstein–Barr virus infection is likely an obligate factor1 should increase interest in the development of an early life Epstein–Barr virus vaccine. Similarly, awareness of the importance of sunlight exposure and vitamin D levels, particularly in pregnancy and early life19 but also throughout the life course, presents multiple points of intervention.24 However, increasing sunlight exposure and vitamin D levels are not without risk25,26 and no clear equipoise has developed.25,27 Reducing smoking and adolescent obesity could markedly reduce the risk of MS; however, there is no direct trial evidence to support this due to the complexity and long term nature of such a trial. The push to implement improved preventive health measures to reduce the burden of other diseases, if successful, will potentially reduce the incidence of MS. An improved understanding of MS risk factors is important for those in higher risk categories and is invaluable when counselling a person with MS who wants to reduce the risk in their children or other relatives.