Pathogenesis and epidemiology of multiple sclerosis

INTRODUCTION — Diseases that affect central nervous system myelin can be categorized as demyelinating (acquired, usually inflammatory) and dysmyelinating (abnormal formation of myelin, usually due to a genetic disease) (table 1). The most common immune-mediated inflammatory demyelinating disease of the central nervous system is multiple sclerosis (MS).

The pathogenesis, pathology, and epidemiology of MS will be reviewed here. Other aspects of MS are discussed separately:

●(See "Clinical presentation, course, and prognosis of multiple sclerosis in adults".)

●(See "Manifestations of multiple sclerosis in adults".)

●(See "Evaluation and diagnosis of multiple sclerosis in adults".)

●(See "Symptom management of multiple sclerosis in adults".)

●(See "Treatment of acute exacerbations of multiple sclerosis in adults".)

●(See "Initial disease-modifying therapy for relapsing-remitting multiple sclerosis in adults".)

●(See "Treatment of secondary progressive multiple sclerosis in adults".)

PATHOGENESIS — MS is a heterogeneous disorder with variable clinical and pathologic features reflecting different pathways to tissue injury [1]. Inflammation, demyelination, and axonal degeneration are the major pathologic mechanisms that cause the clinical manifestations [2,3]. However, the cause of MS remains unknown [4,5]. The most widely accepted theory is that MS begins as an inflammatory immune-mediated disorder characterized by autoreactive lymphocytes [1,6]. Later, the disease is dominated by microglial activation and chronic neurodegeneration [2].

Immunopathology — Several lines of evidence support an important and possibly defining role for the immune system in the development of MS. The cellular immunology of MS involves altered interactions between T cells, B cells, myeloid cells, and additional immune cell populations [7-10]. As an example, pathology studies have demonstrated that inflammatory T cells, B cells, and macrophages are typically seen on histopathologic examination of MS lesions either in biopsies or at autopsy [11]. Inflammatory activity is also associated with breakdown of the blood-brain barrier [12]. Magnetic resonance imaging (MRI) studies have demonstrated disruption of the blood-brain-barrier, as defined by leakage of gadolinium-based contrast agents, at early points during the development of MS lesions in patients with relapsing-remitting disease. This is at a time associated in neuropathologic studies with infiltration by inflammatory cells. T helper 17-type (Th17) cells that are involved in inflammatory and tissue destruction in many immune-mediated systemic diseases are also associated with active MS lesions [13-15].

Additionally, the risk of developing MS is associated with certain class I and class II alleles of the major histocompatibility complex loci that are involved in T cell activation and regulation. In that regard, myelin-reactive T cells are found in MS plaques, the cerebrospinal fluid (CSF), and the peripheral circulation of patients with MS [16,17]. In addition, antibodies against one myelin protein (myelin oligodendrocyte glycoprotein [MOG]) are associated with an MS-like demyelinating disease. (See "Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD): Clinical features and diagnosis".)

It has long been suspected that a foreign antigen, such as a virus or bacteria, provides an antigenic trigger for MS autoimmunity through molecular mimicry [18]. Attention has centered on the Epstein-Barr virus [19,20], as discussed below (see 'Viral infections' below). Antigen presenting cells, including B cells, may activate CD4+ T cells in response to foreign and endogenous antigens, leading to inflammatory responses and tissue damage [21,22]. Furthermore, the CSF of patients with MS contains immunoglobulin G (IgG) and IgM oligoclonal bands that are not present in the serum of these patients; this indicates production of antibodies by plasma cells specific to the neuraxis. (See "Evaluation and diagnosis of multiple sclerosis in adults".)

Many of the medications and therapies that are used to ameliorate MS were identified from studies of experimental allergic encephalomyelitis (EAE), an animal model of MS. This animal model is induced by myelin antigens [23], including myelin basic protein, proteolipid protein, myelin associated glycoprotein, and MOG [24]. Concomitantly, immunomodulatory drugs that reduce the Th1 immune response (eg, interferon beta), increase the Th2 and the T regulatory cell Th3 responses (eg, glatiramer acetate), block T cell movement from the blood into the central nervous system (eg, natalizumab), or deplete B cells (eg, ocrelizumab) are effective for decreasing MS disease activity. (See "Initial disease-modifying therapy for relapsing-remitting multiple sclerosis in adults".)

However, direct proof of an autoimmune cause of MS is lacking, as no specific autoantibody or autoreactive T cell directed against a self-antigen in the central nervous system can passively transfer MS to experimental animals. EAE itself is an imperfect model of MS, as it does not exactly parallel its clinical or pathological features [1,6,25]. In addition, EAE is responsive to many drugs directed against T cells (eg, cyclosporine and monoclonal antibodies directed at CD4 cells). These same drugs have failed to consistently demonstrate effectiveness as therapies for MS.

In addition to loss of myelin and oligodendrocytes, axonal injury is a prominent pathologic feature of the MS plaque (see 'Pathology' below). Disease progression involves a degenerative phase of cerebral atrophy and axonal loss that is not fully attributable to immune mechanisms or inflammation.

Neuropathologic evidence suggests that oligodendrocyte apoptosis, perhaps triggered by viral or glutamate excitotoxicity, may be the primary event preceding inflammation in at least some newly forming lesions in patients with relapsing-remitting MS [26,27]. However, the importance of oligodendrocyte apoptosis in the pathogenesis of MS remains to be established [28].

Alternate theories — Alternate theories of MS pathogenesis include the following [6,25]:

●A possible immune (but not autoimmune) etiology due to a chronic viral infection; however, other than some evidence of differences in infection with Epstein-Barr virus, a common pathogen, no unique virus has been consistently associated with MS

●A nonimmune noninflammatory etiology due to a genetically determined neuroglial degenerative process

Chronic cerebrospinal venous insufficiency (CCSVI), characterized by purported anomalies of cerebrospinal veins that interfere with venous drainage from the brain, is disproven as having a role in the pathogenesis of MS [29,30].

PATHOLOGY — The characteristic neuropathologic feature of MS is the presence of focal demyelinated plaques within the central nervous system, accompanied by variable degrees of inflammation and gliosis, with partial preservation of axons [31,32]. These lesions tend to be located in the optic nerves, spinal cord, brainstem, cerebellum, and the juxtacortical and periventricular white matter [33]. In addition, demyelinated lesions can also be found in the corpus callosum [34] and cortical gray matter [35,36]. Axonal injury can be a prominent pathologic feature of the MS plaque, though not in the acute phase [28,37-39].

Although traditionally considered a disease of focal white matter lesions, the spectrum of MS pathology is now understood to encompass a broader array of abnormalities, including diffuse damage of so-called normal-appearing white matter (NAWM) and normal-appearing gray matter (NAGM) on magnetic resonance imaging (MRI), both of which are associated with a progressive loss of brain volume [40]. Inflammatory cortical demyelination has been found in 38 percent of biopsy-proven cases of early MS [36].

The core MS clinical phenotypes are those of relapsing and progressive disease (see "Clinical presentation, course, and prognosis of multiple sclerosis in adults", section on 'Disease onset and pattern'). However, the pathology of brain injury in relapsing and progressive forms of MS is probably not fundamentally different, though some reports have suggested that progressive forms of MS are marked by reduced or absent inflammation [41] or by an inflammatory process involving the whole brain and meninges [42]. More convincing evidence suggests that primary progressive MS is part of the clinical spectrum of MS and is not pathophysiologically different from relapsing MS that has evolved into a secondary progressive phase [41,43,44].

EPIDEMIOLOGY AND RISK FACTORS — Among central nervous system disorders, MS is the most frequent cause of permanent disability in young adults, aside from trauma [45,46]. MS affects more females than males. A systematic review of 28 epidemiologic studies found that, from 1955 to 2000, the estimated female to male ratio of MS incidence increased from 1.4:1 to 2.3:1 [47]. Subsequent studies have also found that the female-to-male incidence ratio is increasing, mainly due to an increasing incidence of MS in females [48-51]. The reason for this is unknown [50]. A case-control study from Crete noted that an increase in the incidence of MS in females since 1980 was concurrent with a population shift from rural to urban areas, and speculated that environmental factors accompanying urbanization may trigger the development of MS [52].

The incidence and prevalence of MS varies geographically, as discussed below. (See 'Geographic factors' below.)

The mean age of MS onset ranges from 28 to 31 years in various studies; clinical disease usually becomes apparent between the ages of 15 to 45 years, though clinical onset rarely occurs as early as the first years of life or as late as the seventh decade [4]. The mean age of onset is a few years earlier for females than for males [45]. Relapsing-remitting MS has an earlier onset, averaging 25 to 29 years; this may convert to secondary progressive MS at a mean age of 40 to 49 years [4]. Primary progressive MS has a mean age of onset of 39 to 41 years.

In support of a possible autoimmune basis for MS (see 'Pathogenesis' above), some [53,54] but not all [55] studies have observed that patients with MS are more likely than controls to have other autoimmune disorders. However, the majority of such studies are limited in quality, as noted in a systematic review of comorbid autoimmune disease in MS; based upon a meta-analysis of population-based studies, the most prevalent autoimmune conditions identified by the systematic review were psoriasis and thyroid disease (7.7 and 6.4 percent, respectively) [56]. The results also suggested that MS is associated with a possible increased risk of inflammatory bowel disease, uveitis, and pemphigoid.

In addition, patients with other autoimmune disorders are more likely to have MS. A large Danish study found that patients with type 1 diabetes mellitus had an increased risk for developing MS compared with the general population [57]. Another large, well-designed cohort study found that patients with inflammatory bowel disease have an increased risk for demyelinating diseases, including MS [58].

Genetic susceptibility — There are over 200 polymorphisms associated with MS in various studies [59,60]. Among the strongest risk associations are those of certain class I and class II alleles of the major histocompatibility complex (MHC), particularly the HLA-DRB1 locus [61-67]. Mounting evidence suggests that the risk of MS is associated with multiple non-MHC susceptibility genes of modest effect (eg, CD6, CLEC16A, IL2RA, IL7R, IRF8, and TNFRSF1A) [64,65,68-70]. In addition, polymorphisms in the IL7R gene may slightly increase the risk of MS [62,71,72]. (See 'Pathogenesis' above.)

The presence of a vitamin D response element (VDRE) located in the promoter region of many but not all HLA-DRB1 alleles suggests that environmental differences in vitamin D might interact with HLA-DRB1 to influence the risk of MS [73]. However, other factors related to HLA variation may have more impact on MS risk than vitamin D regulation of HLA-DR expression. In one study of White Australians that compared 466 MS cases and 498 controls, the risk of developing MS varied more than 10-fold according to HLA-DRB1 allele type and associated sequence variation in the promoter region, with odds ratios ranging from 0.28 to 3.06 [74]. A protective effect was associated with HLA-DRB1*04, *07, and *09 (DR53 group) alleles, while DRB1*15 and *16 (DR51 group) and *08 (DR8 group) were associated with a higher risk. However, VDRE sequence variation itself was not independently associated with MS risk. Most of the HLA-DRB1 alleles expressed a functional VDRE sequence, including alleles that had no apparent effect on MS risk. Further, in a study from Sardinia, where the prevalence of MS is high, the VDREs associated with several of the HLA-DRB1 variants linked to MS risk in Sardinians were often mutated and nonfunctional [75]. These results suggest that, at least in Sardinia, the effect of vitamin D on HLA-DRB1 expression as mediated by VDRE is quite limited. Instead, among Sardinians, polymorphisms in B-cell activating factor (BAFF) regulatory elements has a very high relationship to the development of MS, as discussed below.

In twin studies, the risk of developing MS for dizygotic twin pairs is the same as that for siblings (3 to 5 percent); however, the risk for monozygotic twins is at least 20 percent and may reach close to 39 percent [76]. For purposes of genetic counseling, the sibling risk of MS is 3 to 5 percent. Studies of unaffected family members that have noted abnormalities on magnetic resonance imaging (MRI) scanning suggest that the risk may be even higher.

The frequency of familial MS varies from 3 percent to 23 percent in different studies. One well-designed population study of 8205 Danish patients with MS found that the relative lifetime risk of MS was increased sevenfold (95% CI 5.8-8.8) among first-degree relatives (n = 19,615) [77]. The excess familial lifetime risk for first-degree relatives was 2.5 percent (95% CI 2.0-3.2) in addition to the sporadic absolute risk of MS in Danish women and men of 0.5 and 0.3 percent, respectively. These sporadic rates from the Danish population are among the highest in the world.

Accumulating data suggest that susceptibility of MS is influenced by the sex of the affected parent [78-84]. Most but not all studies have found a maternal parent-of-origin effect, with an excess of maternal inheritance of MS susceptibility [78,82,83]. By contrast, studies of parent-child pairs with MS have found that paternal transmission is equal to or greater than maternal transmission [79-81]. The explanation for this discrepancy is unclear, but epigenetic mechanisms (eg, DNA modifications such as histone acetylation and DNA methylation that do not modify the DNA sequence) transmitted through cell division may be involved in direct transmission from an affected parent [84].

As genome-wide association studies increase in size, the ability to detect risk alleles conferring even very small increases in MS susceptibility increases. In a 2019 report, the number of genetic variants linked to MS risk was >200 [59]. Although the precise functional effects of these variants are mostly unknown, they are over-represented in regulatory as opposed to coding regions of genes associated with immunologic function, and many of the variants are associated with other autoimmune conditions as well [85]. As an example, a genome-wide association study focused on a population from Sardinia, where there is a high prevalence of MS and systemic lupus erythematosus (SLE) [86]. A variant in the TNFSF13B gene, encoding B-cell activating factor (BAFF), was associated with both MS and SLE. The proposed mechanism is that the TNFSF13B variant causes higher production of soluble BAFF, leading to enhanced humoral immunity and an increased risk of autoimmunity. The authors of this study suggested that among this group, there is an inverse relationship between susceptibility to malarial diseases and autoimmunity. This relationship clearly would not be as important in Northern Europe.

Environmental factors — Environmental factors appear to play a major role in determining the risk of MS. These include viral infections, geographic latitude and place of birth, sunlight exposure and vitamin D levels, and others.

Viral infections — The literature supports a possible infectious stimulus of the immune system as a trigger for MS more than it supports any effect from vaccination [87-89]. Although many viruses have been associated with MS [90], attention has centered on the Epstein-Barr virus (EBV).

●Epstein-Barr virus – EBV has long been suspected as a possible cause or trigger of MS [91-96]. In a 2010 meta-analysis of 18 case-control and cohort studies, the risk of MS was increased after infectious mononucleosis (relative risk 2.17, 95% CI 1.97-2.39) [97,98]. A subsequent study of a cohort comprising over 10 million active-duty young adults in the US military identified 955 diagnosed with MS during their time of service, and determined EBV status using longitudinal serum samples available for 801 cases of MS and 1566 matched controls [99]. Among 35 incident cases of MS who were EBV seronegative at study entry, all but one seroconverted before MS diagnosis. Compared with persistent EBV seronegativity, the risk of MS after EBV seroconversion was elevated by over 30-fold (hazard ratio [HR] 32.4, 95% CI 4.3-245.3); the wide confidence intervals reflect the small number of seronegative cases. Among all incident MS cases, the proportion who were EBV positive by study end was also elevated (800 out of 801) compared with that in controls, corresponding to a similar risk estimate (HR 26.5, 95% CI 3.7-191.6). Although not proof of causation, this finding is strong evidence implicating EBV in the pathogenesis of MS.

The presence of antibodies to EBV has also been associated with MS. A prospective nested case-control study of women found significant elevations in anti-EBV antibody titers before the onset of MS, particularly antibody to the EBV nuclear antigen 2 (EBNA2) [100]. Another nested case-control study found that higher antibody titers to EBNA complex and EBV viral capsid antigen were associated with an increased risk of MS [101]. In addition, there is evidence of molecular mimicry between the EBV nuclear antigen 1 (EBNA1) and the glial cell adhesion molecule (GlialCAM), with one study demonstrating clonal B cells can produce antibodies that cross-react against the EBNA1 and GlialCAM [19].

One difficulty in proving a link between EBV and MS is that serological evidence of EBV can be found in 83 to 90 percent of adults in the Western hemisphere [100,102,103]. On the other hand, EBV seropositivity among adult MS patients is near 100 percent, significantly higher than healthy controls [92,103-107]. Additionally, children with MS are significantly more likely than healthy peers to have serological evidence for prior EBV infection, at an age when EBV seropositivity is much less common than in adults [108]. However, there is conflicting evidence concerning the presence of EBV in brain tissue of patients with MS [109-112].

●Other viruses – In the cohort study of over 10 million military recruits showing an increased risk of MS after EBV, there was no increased risk associated with other viruses, including cytomegalovirus (CMV) [99]. CMV infection was associated with protection from MS in two case-control studies, one of which focused on pediatric-onset MS [113] and the other of which focused on a broad range of onset [114]. The latter study found the negative association was persistent when pediatric-onset MS cases were removed from the analysis, suggesting that cytomegalovirus infection is also protective for adult-onset MS [114]. It is not clear if this association is causal or spurious, nor is it known precisely how such an infection would be protective.

Varicella zoster virus (VZV) was linked to MS in some studies [115-117]. A case-control study found viral particles identical to VZV, and DNA from VZV, in cerebrospinal fluid (CSF) samples from patients with acute relapses of MS [116]. Viral particles were not seen in CSF samples from cases of MS in remission or in samples from control subjects, and DNA from VZV was not seen in most patients in remission. These findings suggest that VZV may participate in, or be activated at the same time as, MS exacerbations. However, they require confirmation in additional studies.

Another hypothesis proposes that early life infections may attenuate the response that leads to autoimmune disorders such as MS [118]. In support of this theory, a population-based case-control study found that higher exposure to infant siblings during the first six years of life was inversely associated with the risk of MS [119]. The proposed explanation was that greater infant sibling exposure leads to increased early life infection exposure or reexposure; this in turn confers protection against autoimmunity later in life.

Geographic factors — The incidence and prevalence of MS varies geographically [120,121]. High frequency areas of the world (prevalence of 60 per 100,000 or more) include all of Europe, southern Canada, northern United States, New Zealand, and southeast Australia. In many of these areas, the prevalence is more than 100 per 100,000; the highest reported rate (300 per 100,000) is in the Orkney Islands. In the United States, the estimated prevalence is 100 to 150 per 100,000, for a total of 300,000 to 400,000 persons with MS [122,123]. Of note, confidence in these prevalence estimates is limited by inconsistent registration, tracking, and reporting of MS cases [124].

The geographic variance in MS prevalence was previously thought to be explained in part by racial differences; White populations, especially those from Northern Europe, appeared to be most susceptible, while people of Asian, African, or Native American origin appeared to have the lowest risk, with other groups intermediate. However, subsequent studies in the United States demonstrated an increased incidence of MS in Black adults [125,126] and children [127], suggesting that this racial susceptibility may be changing.

There is well-documented association between latitude and MS prevalence, with the prevalence of MS increasing from south to north [47,121]. In an analysis from the Nurses' Health Study, for example, the adjusted rate ratios were 3.5 for the northern United States and 2.7 for the middle tiers relative to the southern tier [128]. Persons migrating from a high- to low-risk area after the age of puberty are thought to carry their former high risk with them, while those that migrate during childhood seem to have the risk associated with the new area to which they migrated.

The universal association between latitude and prevalence of MS was challenged by findings from a 2010 systematic review and meta-analysis of epidemiologic studies of MS [49]. The study found that the prevalence of MS increased with geographic latitude in Western Europe, North America, and Australia/New Zealand, but the incidence of MS increased with latitude only in Australia/New Zealand, and not in Western Europe or North America. Higher prevalence could be explained by other factors, such as longer survival, but the clear gradient of incidence in Australia and New Zealand is very important. The latitudinal differences in prevalence were reviewed a 2019 analysis, which concluded that they are indeed present and reflect the epidemiology of MS [121].

Sunlight and vitamin D — One proposed explanation for the possible association of MS with latitude is that exposure to sunlight may be protective, either because of an effect of ultraviolet radiation or of vitamin D [129-131]. A number of studies have found an inverse relationship between sun exposure, ultraviolet radiation exposure, or serum vitamin D levels, and the risk or prevalence of MS [130,132-138], while others have shown that these factors are inversely related to MS disease activity in established MS [139-142]. As examples, an analysis of data from the Nurses' Health Study and Nurses' Health Study II observed that the risk of developing MS was significantly reduced for women taking ≥400 international units/day of vitamin D (relative risk 0.59, 95% CI 0.38-0.91) [137], and a prospective report of over 450 patients with a clinically isolated syndrome suggestive of MS showed that serum 25-hydroxyvitamin D levels, measured in the first 12 months, were inversely associated over the subsequent four years with the risks of conversion to clinically definite MS, the presence of new active MS brain lesions on MRI, and MS progression [140]. Note that these studies enrolled primarily Caucasians; the results may not apply to other racial or ethnic groups [143].

Mendelian randomization studies have also found a link between genetically low serum 25-hydroxyvitamin D and the later risk of multiple sclerosis. (See "Vitamin D and extraskeletal health", section on 'Multiple sclerosis'.)

Others — Environmental triggers unrelated to geography may be involved in the development of MS [144]. A number of studies have suggested an association between tobacco smoking and MS [144]. As examples, a cross-sectional study of 22,312 people in Norway found a higher risk of MS in ever-smokers than in never-smokers (relative risk 1.81, 95% CI 1.13-2.92) [145], and a case-control study in the United Kingdom found similar results [146]. Smoking may also be a risk factor for disease progression [129,146-151]. By contrast, a population-based case-control study from Sweden reported that use of oral snuff (smokeless tobacco) was not associated with MS risk among patients who did not smoke cigarettes, while greater use of oral snuff appeared to be protective against MS among those who did smoke cigarettes [152]. This finding suggests that nicotine is not responsible for the increased risk of MS among tobacco smokers, but more data are needed.

Birth month has been implicated as a possible risk factor for MS, though the literature is conflicting. A 2013 meta-analysis and systematic review found that the risk of MS was increased for those born in April and May and decreased for those born in October and November [153], suggesting that the gestational or neonatal environment influences the risk of MS later in life. However, it is possible that studies finding a month of birth effect are actually false positive results that result from confounding caused by seasonal variation in birth rates, with data from Europe and North America showing excess births in March, April, or May, and reduced births in November, December, and January [154].

Obesity in childhood or adolescence may also be a risk factor for MS, as suggested by several case-control studies and Mendelian randomization analyses [131,155-160].

Eighty percent of the immune system cells reside in the gastrointestinal tract, and there is some evidence that the microbiome of patients with MS differs from healthy controls. Further research is needed to clarify the role of gut bacteria and its relationship to the immune system in patients with MS [161,162].

Vaccinations — Because the pathogenesis of MS is thought to involve the immune system, it has been hypothesized that a stimulus of the immune system (eg, a vaccine) may trigger the disease. However, several studies have failed to show any positive association between vaccines and MS [163-168].

●Despite an epidemiological study suggesting increased MS risk following the hepatitis B vaccine [169], several well-designed studies seemingly have refuted the possible link: one finding no association between hepatitis B vaccination and the development of MS [163], another other finding no association between several different vaccines and disease relapse in patients with MS [164], and a third study finding that hepatitis B vaccination was associated with a reduced risk of MS [170].

●A 2006 systematic review of nine case-control studies found a negative association between tetanus vaccination and the risk of MS (odds ratio 0.67; 95% CI 0.55-0.81) [165].

●A summary of published evidence (through January 2001) supported the safety of vaccination in patients with MS [166], and a subsequent case-control study found no association between several different vaccines and the development of MS and/or optic neuritis [167].

Although a later, well-designed case-control study found an increased risk of MS in patients who had received hepatitis B vaccination [171], the indisputable large benefit of this vaccine far outweighs the possible and still unproven risk of developing MS that the vaccine may carry [171,172]. This is especially the case considering the vaccine is now given the first days of life in the United States and studies suggesting any link were conducted in adults.

Vaccination against the human papillomavirus has been shown not to increase the risk of MS [173]. (See "Human papillomavirus vaccination", section on 'Vaccine safety'.)

SUMMARY

●Pathogenesis and immunology – Multiple sclerosis (MS) is a heterogeneous disorder with variable clinical and pathologic features reflecting different pathways to tissue injury. Inflammation, demyelination, and axon degeneration are the major pathologic mechanisms that cause the clinical manifestations. The most widely accepted theory is that MS begins as an inflammatory autoimmune disorder mediated by autoreactive lymphocytes. Later, the disease is dominated by microglial activation and chronic neurodegeneration. (See 'Pathogenesis' above and 'Immunopathology' above.)

Alternate theories of MS pathogenesis include an immune (but not autoimmune) etiology due to a chronic viral infection, a nonimmune noninflammatory etiology due to a genetically determined neuroglial degenerative process, and chronic cerebrospinal venous insufficiency (CCSVI). In particular, the CCSVI theory has been disproven. (See 'Alternate theories' above.)

●Epidemiology – The estimated female-to-male ratio of MS incidence is approximately 2:1, with some data suggesting the ratio is even higher. The median and mean ages of MS onset are 23.5 and 30 years of age, respectively. The peak age of onset is approximately five years earlier for females than for males. Onset of MS can rarely occur as late as the seventh decade. (See 'Epidemiology and risk factors' above.)

●Genetics – Genetic factors appear to contribute to the risk of MS, especially variation involving the HLA-DRB1 locus. (See 'Genetic susceptibility' above.)

●Epstein-Barr virus – There is mounting evidence of an association between the Epstein-Barr and MS, but proof of causality is not yet established. (See 'Viral infections' above.)

●Risk factors

•The incidence and prevalence of MS vary geographically. (See 'Geographic factors' above.)

•There is an inverse relationship between sun exposure, ultraviolet radiation exposure, or serum vitamin D levels, and the risk or prevalence of MS. (See 'Sunlight and vitamin D' above.)

•Tobacco smoking and childhood or adolescent obesity may be risk factors for MS. (See 'Others' above.)

•There is no association between vaccines and the risk of MS. (See 'Vaccinations' above.)