The epidemiology, pathology and pathogenesis of MS: Therapeutic implications

Abstract

Multiple sclerosis (MS) is a chronic, and potentially disabling, inflammatory disease of the central nervous system (CNS). MS is generally characterized by recurrent, and self-limited, episodes of neurological dysfunction, which occur unpredictably and often result in multifocal tissue injury within the CNS. Currently, women are affected two to three times as often as men although this may not have been the case during earlier Time-Periods. The pathogenesis of MS is known to involve both critical genetic and environmental mechanisms. Nevertheless, in addition to these two mechanisms, disease-pathogenesis also involves a "truly" random event. Indeed, it is this random mechanism, which is responsible for the currently-observed (and increasing) excess of women among patients with MS. This review summarizes the current state of knowledge regarding the pathogenesis of MS (includong its epidemiology, pathology, and genetics) and considers the therapeutic implications that these pathogenetic mechanisms have both for our currently available therapies as well as for the possible therapeutic approaches to the management of this potentially disabling condition in the future.

Keywords: Genetics; Multiple sclerosis epidemiology; Pathogenesis; Pathology; Therapy.

Fig. 1

The distribution of the age at symptom-onset in a cohort of $1,463$ patients with multiple sclerosis (MS). The age of symptom-onset in Figure is given as (mean ​± ​1 standard deviation) (Data from: Liguori M, Marrosu MG, Pugliatti M, Giuliani F, De Robertis F, Cocco E, Zimatore GB, Livrea P, Trojano M. Age at onset in multiple sclerosis. Neurol Sci. 2000; 21:S825-9). Modified with permission from: Goodin DS [6].

Fig. 2

Characteristic pattern and distribution of MRI lesions in a patient with MS. Panel A is an axial T2-weighted FLAIR image demonstrating the typical periventricular, subcortical, and central white matter lesions of MS. The white arrow points to a juxtacortical lesion. Panel B is an axial T1-weighted, gadolinium (Gd)-enhanced image demonstrating enhancement of the lesion adjacent to the frontal pole of the L lateral ventricle (white arrow). This enhancement represents the acute focal breakdown of the Blood-Brain-Barrier in an active MS lesion. Gd-enhancement generally lasts (<3 months). The other, non-enhancing, lesions are, therefore, presumably more chronic than this. Panel C is a sagittal T2-weighted FLAIR image demonstrating typical lesions of the corpus callosum in patients with MS (white arrow). These were initially described by the Scottish pathologist James Dawson in 1916 and are often referred to as “Dawson's Fingers”. They represent inflammation that surrounds the draining medullary veins. These veins are oriented at right angles to the ventricular surface and, thus, when inflamed, appear, both on MRI and on pathological examination, as vertical spikes (fingers) off the ventricular surface. Panel D is a sagittal T2-weighted image of the cervical spinal cord demonstrating a demyelinating lesion in the posterior cervical spine at the level of C3 (white arrow).

Fig. 3

The Upper Panel represents the distribution of MS-prevalence around the world. Dotted black line indicates the Equator. Note that the MS-prevalence increases with increased latitude both north and south of the equator – data taken from reviews of worldwide epidemiology of MS [5,32]. The Lower Panel represents the global distribution ultraviolet B (UVB) radiation reaching the earth's surface divided into three Exposure Zones. Zone 1 indicates regions where there is a sufficient amount UVB radiation reaching the earth's surface for adequate vitamin D synthesis during $\left(12/12\right)$ months of the year. Zone 3 indicates regions where there is an insufficient amount UVB radiation reaching the earth's surface for adequate vitamin D synthesis during $\left(12/12\right)$ months of the year. In Zone 2 (where there is a variable insufficiency), an increasing number of insufficient months indicated by the coloration changing from reds to oranges to yellows to greens. The red indicated that there is a sufficient amount of UVB radiation reaching the earth's surface for vitamin D synthesis during (11/12) months during the year. The darkest green indicates that there is a sufficient amount of UVB radiation reaching the earth's surface for vitamin D synthesis during only $\left(1/12\right)$ months of the year. The amount of UVB radiation also varies in Zones 1 and 3 but is either always adequate (Zone 1) or always inadequate (Zone 3). Modified with permission from: Goodin DS [6] and from: Jablonski NG and Chaplin G [163].

Fig. 4

Gross pathological and histopathological appearance of MS lesions. Panel A is a section of the ‘naked eye’ appearance of a coronal cut section from an unfixed brain of a patient with MS. Black arrows indicate the chronic lesions (plaques) of MS in the subcortical, periventricular and central white matter. Panel B represents a coronal section from the brain of an MS patient (luxol fast blue stain) and demonstrates the areas of chronic demyelination in the central and periventricular white matter. The solid black arrows indicate areas of complete demyelination. The rounded periventricular lesions are chronic; the star-like lesion is active. The dashed black arrows indicate the appearance of so-called “shadow” plaques. These shadow plaques may represent early active lesion formation, aborted lesions, or areas where partial, but incomplete, re-myelination has taken place. Panels C and D (hematoxylin and eosin stain) demonstrates the perivenular infiltrate of lymphocytes and plasma cells in an active MS lesion. Panel C is cut along the venule's course whereas Panel D is cross sectional. Modified with permission from: Adams CWM. (1989) Color Atlas of Multiple Sclerosis and Other Myelin Disorders. Wolfe Medical Publications Ltd. Printed by Cowell WS, Ipswich, Suffolk, United Kingdom.

Fig. 5

Using the Canadian MS data (Table 4), response-curves are depicted for developing MS to an increasing odds that a proband, randomly selected from the susceptible subset $\left(G\right)$, experiences the event $\left(E\right)$ – i.e., they experience a set of environmental conditions that are “sufficient”, by themselves, to cause MS in them [64]. {NB: exposure here is being measured as the “odds” that the proband will experience a “sufficient” exposure [64]. This measure is related to the probability of a“sufficient” exposure, not to any specific exposure conditions}. Response-curves representing susceptiblewomen (black lines) and men (red lines) are depicted separately. The constant $\{\mathit{c}=P\left(MS|M,G,E\right)\}$ represents the limiting probability of developing MS in susceptible men whereas the constant $\{\mathit{d}=P\left(MS|F,G,E\right)\}$ represents this limiting probability in susceptible women. The probability that a proband, randomly selected from the susceptible subset $\left(G\right)$ is a female is: $\left\{P\left(F|G\right)=p\right\}$, whereas the probability that this proband is a male is: $\left\{P\left(M|G\right)=1-p\right\}$ The hazard proportionality constant, ($R)$, if the hazards are proportional, is defined elsewhere [64]. The blue lines represent the change in the (F:M) sex-ratio with increasing exposure, with the scale provided in the upper left. The thin grey vertical lines represent the portion of the response-curves that corresponds to the observed change in the (F:M) sex-ratio – $\left(2.2\to 3.2\right)$– which took place in Canada between Time-Periods #1 & #2 (see Table 4). In constructing these depicted curves, three assumptions have been made. These are: (1) at every exposure level, the randomly selected proband (who experiences a “sufficient” exposure”) has some non-zero probability of being either a man or a women. Nevertheless, this probability may not be the same (or even similar) for the two sexes – i.e., we are assuming only that the threshold is the same for males and females or: ($\lambda =0)$ [64]; (2) with any increase in exposure – i.e., any increase in $P\left(E|G\right)$ – the probability of developing MS is increased for both susceptible men and susceptible women although, again, possibly with different degrees of increase for each sex; and: (3) “male-MS” and “female-MS” represent the same underlying disease process [64]. From these three assumptions and from the observations, both in Canada and around the world, that the MS-prevalence is increasing, especially among women [7,[59], [60], [61], [62], [63], [64]] – see alsoTable 4 – four conclusions directly, and necessarily, follow [64]. First, the hazards are proportional. Second, at every exposure level, susceptible men are more likely than susceptible women to experience a “sufficient” exposure [64]. Third, at every exposure level a susceptible man who experiences a “sufficient exposure” less likely to actually develop MS than a susceptible woman in the same circumstances. This conclusion reflects a “truly” random mechanism in MS-pathogenesis [64]. And fourth, the difference in disease-expression between the sexes under the stated conditions is explained entirely by this “truly” random mechanism [64]. Thus, this “truly” random mechanisms is similar to a biased coin, which favors disease-development in suscveptible women over susceptiblemen. Notably, however, as noted previously [64], the fact that a coin is biased only indicates that, when the coin is flipped, the two possible outcomes are not equally likely. It does not imply that the coin-flip is a non-random process. The consequences of dropping or modifying any or all of these assumptions are considered elsewhere [64].

Fig. 6

The frequecy-distrubutions of 5-allele conserved extended haplotypes (CEHs) $A\sim C\sim B\sim DRB1\sim DQB1$ – from different regions around the world. The data is from the study of Gragert and colleagues [91]. For simplicity of its presentation, the populations displayed in the Figure have been restricted to African, Mexican, South Asian Indian, Native North American, European, Chinese, Aleut, and Japanese. Nevertheless, these displayed $CEH$ frequency distributions are fully representative of every world region reported by Gragert and colleagues [87], including those not presented in the Figure – i.e., Korean, Southeast Asian, Vietnamese, Hawaiian, Filipino, Middle Eastern, and Native South American – seeTable 7. This graph plots the cumulative number of unique $CEHs$ in each population (beginning with the highest haplotype-frequency $CEH$) against the percentage of the total number of $CEHs$ for each of these populations. In each population, the majority of the total haplotypes are accounted for by only a very small number of very high-frequency unique $CEHs$. Nevertheless, the composition of these very high-frequency $CEHs$ varies markedly between populations (see Table 7). Moreover, $CEH$-diversity also varies between populations. The most diversity is found among Africans, where the top $10CEHs$ account of 6.5 ​ % of the total number of $CEHs$ present compared to Japan where the top $10CEHs$ account of 24.4 ​ % of the total number of $CEHs$ present (see Table 7). Notably, these haplotypes span $\sim 3mb$ of DNA and involve both Class I and Class II MHC alleles. Nevertheless, there are 451 loci within the extended $MHC$, of which 252 are expressed [86]. Almost, certainly, some (or many) of these expressed genes will also be linked to these $CEHs$ so that the “nominal” identity of any 5-allele CEH cannot be assumed to be an “actual” identity of the extended haplotype.

Fig. 7

An example illustrating the difficulties that arise when trying to make inferences regarding causation from association studies. Illustrated are different causal pathways that might explain a possible association between epilepsy and MS. Approximately $0.5-1\%$ of the general population suffers from epilepsy whereas, among MS patients, several studies suggest that this percentage may be as high as $\left(2to4-fold\right)$ greater [5]. However, regardless of whether this association is genuine, the schemes and connections presented in this Figure are hypothetical and speculative. In Pathway ($1)$, it is envisioned that there is some antecedent event (Event A), which is common to both condition – e.g., a shred genetic pre-disposition, a traumatic event, a toxic exposure, etc. This creates the association. In Pathway $\left(2\right)$, both arms produce the same “side effect (Event I), which, potentially, can create an association. Otherwise, these causal pathways involve completely unrelated and non-overlapping events. Prevention of MS will not prevent epilepsy and vice versa. These relationships are considered “spurious associations”. Also, even if these pathways do create the association, there may be other, more important, pathways leading to MS and epilepsy, which don't include any common events (e.g., the dashed part of Pathway 5), and that weaken the observed association. In Pathway ($5)$, the antecedent event (Event M) only predisposes to epilepsy; MS follows automatically (if other necessary intervening Events also occur). However, because epilepsy generally occurs late in the MS disease course (i.e., long after the symptom-onset), the solid arrow pathway is excluded because it violates Bradford Hill's criteria of temporality that effect must follow cause [81]. Notably, however, most individuals with epilepsy don't have MS, so that following an Event N (e.g. a developmental anomaly, trauma, genetic factors, etc), an alternative dashed arrow path to the epilepsy must exit, which leads to epilepsy without MS occurring. As a result, if epilepsy occurs, depending upon circumstances, it can precede or follow MS. In Pathways $\left(3\&4\right)$, the antecedent event (Event K) only predisposes to MS; epilepsy follows automatically (again, presuming that certain necessary intervening Events occur). However, these two pathways represent alterative explanations for the connection between MS and epilepsy. In MS, lesions are sometimes located in the cortex or in the white matter adjacent to the cortex [5,[68], [69], [70], [71], [72]]. In Pathway $\left(3\right)$ some of these lesions are presumed to occasionally be epileptogenic and cause epilepsy. This is the pathway invoked by most authorities to explain the possible association [5]. However, it is well-established that MS can cause incoordination, imbalance, and falls [5]. In Pathway $\left(4\right)$, it is presumed that these symptoms lead to more frequent head strikes and brain injuries and, thus, more epilepsy. However, potentially, one can enter this second causal pathway to epilepsy at the brain injury stage following numerous antecedent traumatic events (e.g., motor vehicle and bike accidents, blunt trauma, serious falls, work-related injuries, blast injuries, etc), designated here as (Event L). In this manner, the antecedent event of MS can be bypassed entirely. Indeed, in any large case series of post-traumatic epilepsy, the likelihood of finding any MS-patients among the cohort is vanishingly small so that, undoubtedly, the most common paths to epilepsy for Pathways $\left(3\right)$, $\left(4\right)$ and $\left(5\right)$ are the alternative paths where MS and epilepsy occur separately from each other. Thus, the prevention of MS will have no impact on the occurrence of post-traumatic epilepsy and vice versa, despite MS and epilepsy both being components of one possible causal pathway.