Description

Definition and epidemiology. Multiple sclerosis is a chronic inflammatory, immune-mediated disorder of the central nervous system that starts in early adulthood in the majority of cases but can occur in childhood, age 16 or younger, in approximately 5% (70; 76). The earliest described age of onset is prior to 24 months of age (72). The etiology is unknown, though a parainfectious immune response at onset may initiate cross-reactive nerve or myelin immunity that becomes self-sustained in susceptible individuals. The HLA-DRB1 genetic locus and environmental factors, such as vitamin D levels, Epstein-Barr virus infection, and smoking, influence the risk and development of both pediatric and adult multiple sclerosis (10; 58; 60; 28; 87). Overall, the incidence is currently, on average, 2 per 100,000 per year based on a study (84). The prevalence has risen significantly since 2013, to 35.9 per 100,000 in 2019 to March 2020, throughout the world. A total of 2.8 million people had multiple sclerosis in 2020, indicating a 30% increase in prevalence since 2013 that was possibly attributed to diagnosis at an earlier onset of disease, better diagnostic capabilities, and, most likely, longer survival among individuals diagnosed with multiple sclerosis, using the Atlas for Multiple Sclerosis (84).

Pathology. Multiple sclerosis is characterized by chronic inflammation with demyelination as well as axonal degeneration and loss of oligodendrocytes; it involves gliosis, remyelination, and synaptic changes (42; 88; 30). Inflammation is often multifocal but can be diffuse and focal at times during the course of disease. Demyelination is not limited to the white matter but also appears in cortical lesions, and inflammation is present in the meninges in early multiple sclerosis (53). In addition, grey matter loss with cortical and thalamic neuronal loss occurs from the onset of disease (81; 80; 19). Therefore, the central nervous system inflammatory process contributes to a slowly progressive disease course that may represent cumulative neuronal and synaptic loss in addition to other factors, including early-onset axonal degeneration (42; 88; 30).

Epidemiology

Genetics. The HLA-DRB1 locus is the most important genetic risk factor for developing multiple sclerosis, conferring increased susceptibility for whites (-DRB1*1501) and African Americans (-DRB1*1503) (67). It may also contribute to an approximately 3-fold increase in the risk of multiple sclerosis susceptibility in childhood (-DRB1*15) (14; 46). Conversely, variations in the HLA-A gene confer decreased risk susceptibility for developing multiple sclerosis (46). Other genetic loci, including IL2RA and IL7R, have been identified in genome-wide association studies as being important for multiple sclerosis susceptibility (44; 46). Furthermore, these genetic risk loci are associated with increased propensity to childhood onset multiple sclerosis (79). Although other genome-wide association studies have shown over 4000 single nucleotide polymorphisms, the two loci found in the most recent study by the International Multiple Sclerosis Genetics Consortium showed evidence of enrichment of CLECL1 in microglial cells suggestive of microglial activation in the central nervous system cortex. In addition, it is enriched in peripheral immune cells, including T, B, NK, and dendritic cells arising from the thymus, which is a tissue considered important for the peripherally mediated arm of autoimmunity in multiple sclerosis and other autoimmune disorders (45). However, among all the single nucleotide polymorphisms identified, the genetic effects of any locus on the risk for developing multiple sclerosis are significantly less potent compared to the major histocompatibility complex (67; 24; 46). Furthermore, there is an increased propensity among patients with multiple sclerosis of developing young adult-onset Hodgkin lymphoma, suggesting a shared genetic risk factor for both diseases and implicating a genetic cause that may be linked to an Epstein-Barr virus etiology (59). The role of epigenetics in multiple sclerosis has not been clearly defined. However, it is possible that epigenetic mechanisms, including DNA methylation and histone modification, may modulate the interaction between genes and the environment and explain some of the differences in multiple sclerosis susceptibility (17). They may also account for some of the differences in disease activity, including the effects of histone modification on remyelination through regulation of oligodendrocyte differentiation, a process that may be impaired in older individuals (74).

Environmental factors. There are differences in the frequency of multiple sclerosis in different parts of the world. The frequency in Europe, North America, and Australia is high compared with Asia and tropical regions, where it is low. The incidence correlates positively with increased distance from the latitude of the equator, and this cannot be fully explained by genetic differences (05; 06). There is also a risk modification for developing multiple sclerosis that is associated with geographic migration, with risk dependent on both age at the time of migration and direction of migration. The younger the age of migration, the more likely it is that the incidence of multiple sclerosis in migrants is closer to the multiple sclerosis incidence of their residing destination (05). Environmental factors in children as well as in adults, especially Epstein-Barr virus infection and vitamin D levels, contribute significantly to multiple sclerosis risk and may explain some of the geographic variation and differences associated with migration (05).

The causal mechanism of Epstein-Barr virus playing a critical role in determining the risk for developing multiple sclerosis has been found in a study published in Science. In a cohort of 10 million young adults on active duty in the United States military, 955 were diagnosed with multiple sclerosis after developing an Epstein-Barr virus infection (13). Bjornevik and colleagues reported that there is a 32-fold increased risk of developing multiple sclerosis (hazard ratio for multiple sclerosis with Epstein-Barr virus seroconversion versus seronegativity was 32.4, with a 95% confidence interval of 4.3 to 245.3 and p < 0.002) over a 20-year period. There were no other viruses shown to increase the risk of developing multiple sclerosis in this patient population per Dr. Ascherio and his group. This included cytomegalovirus, which did not appear to be associated with an increased risk of developing multiple sclerosis in this population cohort. Epstein-Barr virus infection in advance of multiple sclerosis onset was found in nearly all of the 801 cases diagnosed with multiple sclerosis matched with their cohorts, with the exception of one individual patient among this cohort.

An age-matched case-control study showed that pediatric patients with multiple sclerosis who were seropositive for Epstein-Barr nuclear antigen-1 had an increased risk for multiple sclerosis (odds ratio 3.78, 95% confidence interval 1.52-9.38, p = 0.004), whereas those with remote infection with cytomegalovirus had a decreased risk of multiple sclerosis (odds ratio 0.27, 95% confidence interval 0.11-0.67, p = 0.004) (87). Infection with herpes simplex virus (HSV) 1 was associated with a decreased multiple sclerosis susceptibility in those who were positive for the DRB1*15 allele (odds ratio 0.07, 95% confidence interval 0.02-0.32, p = 0.001) and increased susceptibility in those who were negative for the allele (odds ratio 4.11, 95% confidence interval 1.17-14.37, p < 0.05). This suggests that although infection with Epstein-Barr virus may increase risk for developing multiple sclerosis, other infections such as cytomegalovirus may actually reduce it. HSV-1 may decrease multiple sclerosis susceptibility in patients with the DRB1*15 allele (87). However, a multinational observational study of age- and region-matched pediatric patients with multiple sclerosis and non-demyelinating neurologic disorder control participants showed that 86% of pediatric patients with multiple sclerosis compared with 64% of controls had positive Epstein-Barr nuclear antigen-1 and viral capsid antigen consistent with remote Epstein-Barr virus infection, regardless of geographical location (p = 0.025, adjusted for multiple comparisons) (10a). The patients with multiple sclerosis did not show any difference from controls in the seroprevalence of IgG antibodies directed against cytomegalovirus, HSV, parvovirus B19, or varicella zoster virus. Seropositivity for remote Epstein-Barr virus infection in children was associated with a nearly 3-fold increase in the likelihood of multiple sclerosis compared with age- and region-matched controls (95% confidence interval 1.4-5.8, p < 0.05 after adjustment for multiple comparisons) (10a).

The association between Epstein-Barr virus and increased multiple sclerosis susceptibility is established, but the association with other viral and bacterial antigens proposed to be important triggers in the etiology of multiple sclerosis, including Chlamydia pneumoniae and human herpes virus (HHV) 6, is debatable. Antibodies to C pneumoniae detected in the CSF of 28% of pediatric patients with multiple sclerosis comprised less than 1% of total intrathecal IgG, suggesting their presence may be part of a polyspecific oligoclonal response (71). The presence of C pneumoniae DNA in the CSF of adult patients with multiple sclerosis has not been confirmed (15). Although HHV-6 are ubiquitous, HHV-6 found in autopsy specimens of multiple sclerosis lesions and evidence of HHV-6 infection in the serum and CSF of patients with multiple sclerosis is unlikely to explain the epidemiology of multiple sclerosis (05).

Vitamin D insufficiency is associated with increased risk for multiple sclerosis. Based on a case control study using the Department of Defense Serum Repository database, there was a 41% reduction in multiple sclerosis risk in whites for each 50 nmol/L increase in 25-hydroxy vitamin D (relative risk 0.59, 95% confidence interval 0.36-0.97, p = 0.04) (64). Low levels of vitamin D are associated with increased disease activity. In a retrospective study, pediatric patients with multiple sclerosis or clinically isolated syndrome with lower serum 25-hydroxy vitamin D3 levels had an increased relapse rate, with each 10 ng/mL increase in the adjusted 25-hydroxy vitamin D3 level associated with a decrease in the rate of future relapses by one third (incidence rate ratio 0.66, 95% confidence interval 0.46-0.95, p = 0.024) (60). However, in a study of patients with multiple sclerosis treated with interferon beta-1b after a clinically isolated syndrome, a 20 ng/mL increase in the average serum 25-hydroxy vitamin D levels during the first 12 months predicted decreased multiple sclerosis disease activity, characterized by a 57% lower rate of new active lesions (p < 0.001), and a 25% lower 12-month increase in the T2-lesion volume (p < 0.001), whereas it predicted a 57% lower relapse rate during a 4-year period, with p < 0.05 (07). On the other hand, 25-hydroxy vitamin D levels greater than or equal to 20 ng/mL at 12 months predicted reduced disability in subsequent months 12 through 60 as measured by the expanded disability status score of -0.17 (p = 0.004) (07). These findings were supportive of decreased serum 25-hydroxy vitamin D levels below 20 ng/mL as a risk factor for disease progression in the early stages of multiple sclerosis in patients treated with interferon beta-1b. On the other hand, an increased level of 25-hydroxy vitamin D by an increment of 20 ng/mL was associated with improved MRI findings, including decreased T2-lesion volume and decreased rate of active lesion formation in patients treated with interferon beta-1b, suggesting that there may be decreased disease activity in this group (07). More studies are needed to look at vitamin D levels and their supportive role in treating vitamin D deficiency to improve outcome in all multiple sclerosis patients.

Although cigarette smoking and exposure to secondhand smoke from tobacco use may not provide an explanation for the geographic variation and altered multiple sclerosis susceptibility associated with migration, it is a significant risk factor for multiple sclerosis. In a prospective Nurses Health study of women in the United States, women who smoked 25 or more packs a year had an increased incidence of multiple sclerosis (relative risk 1.7, 95% confidence interval 1.2-2.4, p < 0.01) (43). In a population-based case-control study in France, children with exposure to parental cigarette smoking had an adjusted relative risk of 2.12 of having multiple sclerosis (relative risk 2.1, 95% confidence interval 1.43-3.15), an effect that was greater in those with multiple sclerosis onset after 10 years of age (relative risk 2.49, 95% confidence interval 1.53-4.08) (58).

Obesity is another environmental risk factor for multiple sclerosis. A large study using cohorts from two separate Nurses’ Health Studies showed that obesity at 18 years of age doubles the risk of multiple sclerosis with a multivariate pooled relative risk of 2.25 (95% CI: 1.50–3.37, p < 0.001) (63). Childhood and adolescent obesity have been identified as a risk factor for childhood multiple sclerosis, which is similar to adults (61; 50). A case cohort study of patients from Kaiser Permanente Southern California showed that obesity was associated with an increased risk of multiple sclerosis and clinically isolated syndrome in girls 11 to 18 years of age (adjusted OR: 1.58, CI: 0.70 and 4.49 in moderately obese female children; adjusted OR: 3.76, CI: 1.54 and 9.16 for extremely obese female children; p = 0.005) (50). Although these were statistically significant results in young female children, there was no significantly increased risk for multiple sclerosis and clinically isolated syndrome in obese male children according to this study (50). A large prospective study of body mass index in children aged 7 to 13 showed that girls with a 95th or greater percentile for BMI had a 1.61-fold versus 1.95-fold increased risk for multiple sclerosis compared with girls below the 85th percentile for BMI (p < 0.01 and p < 0.001, respectively) (61).

Many patients with multiple sclerosis have a metabolic syndrome that is characterized by clinical symptomatology thought to be a risk factor for coronary artery disease, stroke, and type 2 diabetes, including elevated blood pressure, high blood sugar levels, elevated waistline area around the abdominal region, elevated triglyceride level, and low high-density lipoprotein level. Some investigators have looked at the role of treatments like metformin and pioglitazone for multiple sclerosis MRI abnormalities that may be monitored over time while on these treatments for metabolic syndrome. There was a significantly decreased number of new or enlarging T2 lesions on MRI of the brain in patients treated with metformin or pioglitazone compared with those treated with controls (65). No p values were reported for these findings, but the authors noted that significance meant a p value less than 0.05 for a two-tailed test of significance in their results. The greatest reduction in the mean number of new or enlarging T2 lesions occurred at 18 months following treatment with metformin at 0.5, with a standard deviation of 0.7, and following treatment with pioglitazone at 0.6, with a standard deviation of 0.8. However, no significant differences in the annualized relapse rate or disability on the Expanded Disability Status Scale score was found at 24 months during the follow-up period between those who received metformin or pioglitazone and the control group. Nevertheless, treatment with metformin increased serum adiponectin levels, with a mean of 15.4 and a standard deviation of 5.5 micrograms/mL compared with the control group, where adiponectin levels were relatively reduced, with a mean of 4.5 and a standard deviation of 2.4 micrograms/mL, p < 0.001. Serum leptin levels were significantly decreased in those treated with metformin, with a mean of 5.5 and standard deviation of 2.5 nanograms/mL, compared with the control group, where the mean was comparably increased at 10.5 with a standard deviation of 3.4 ng/mL, p < 0.001 (65). These study results suggest that the treatment of patients with multiple sclerosis, who also have a metabolic syndrome, with metformin may influence the serum levels of adipokines, such as adiponectin. This may be important in obesity—a risk factor for multiple sclerosis in adolescents who are typically female in gender. Research on treatment for metabolic syndrome continues, and there are other studies looking at the role of metformin in patients with multiple sclerosis.

Clinical characteristics of multiple sclerosis in children

The majority of pediatric patients with multiple sclerosis have relapsing-remitting disease. A primary progressive course is rare in pediatric multiple sclerosis, unlike adult multiple sclerosis in which 10% to 15% of patients have a primary progressive course (10a). The ratio of females to males is age-dependent. The gender ratio is nearly equal with a female to male ratio of 0.8:1 in children under 6 years of age, then increases to 1:6:1 in children between ages 6 through 10, and then increases again to 2:1 for onset after 10 years of age (10b). A prospective population-based study in South-East Wales showed that the female to male ratio was 4:1 at adolescence, 2.5:1 in adults until age 45 to 49, and then 1:1.5 after age 50 (27). Exacerbations of inflammation may be manifested clinically with focal neurologic deficits. An initial clinical presentation of optic neuritis, brainstem, or cerebellar symptoms may be characteristic of a monofocal attack, whereas monofocal transverse myelitis is infrequent in pediatric multiple sclerosis (10a). Acute transverse myelitis occurs in 20% of children who initially present with an acquired demyelinating syndrome. The majority have longitudinally extensive transverse myelitis spanning three or more vertebral segments on MRI, which is more typically associated with a neuromyelitis optica spectrum disorder (02). The size of the focal inflammatory lesions along with the extent of the acute inflammation appears to be a function of age, with the more severe vascular inflammation in children and progressively less likelihood for vascular inflammation with increasing age (55; 08; 86). The presentation with large solitary or multifocal lesions may be seen at this time in early-onset acquired demyelinating diseases. A monophasic versus recurrent disease process, such as acute disseminated encephalomyelitis, may present in a similar manner. Episodes of focal neurologic deficits in children, including severe deficits, appear to be followed by recovery more often than similar episodes in adults. Earlier age at onset of multiple sclerosis is associated with an increased likelihood of complete recovery from an initial relapse (27). This may represent the potential for improved recovery in the context of resolution of edema. In addition, however, greater neural plasticity and synaptic changes may contribute at least initially to the tendency for improved recovery from acute deficits in children. Chronic ongoing pathological processes in children often have the potential of disrupting the developmental process. Consequently, cognitive dysfunction is frequent even in the absence of physical disability. In a multicenter, cross-sectional study of 187 pediatric patients with multiple sclerosis and 44 patients with clinically isolated syndrome, patients younger than 18 years of age showed that 35% of pediatric patients with multiple sclerosis and 18% of patients with clinically isolated syndrome had neuropsychological testing results consistent with cognitive impairment, highlighting problems with fine motor coordination (54%), difficulty with visual-motor integration (50%), and slowness of information processing (35%) (47). Processing speed, one of the cognitive domains that is often impacted in children with multiple sclerosis, as measured by the Symbol Digit Modalities Test, was found to increase with age in patients with childhood multiple sclerosis, and then it subsequently decreased over time in these patients (03). Furthermore, the prevalence of seizures in patients with multiple sclerosis was 3% to 4% based on six population-based studies (48). It is possible that the prevalence of seizures in pediatric-onset multiple sclerosis may be higher than in the adult population based on some studies in the past decade (29). However, further research may be needed on this topic for now. Pediatric onset of multiple sclerosis at younger than 16 years of age (p = 0.01) and increased mean expanded disability status scale score (p = 0.004) were both associated with an increased likelihood of having seizures in one study (29).

The period of time to the development of onset of progressive disability appears to be longer in patients with disease onset at an early age (70). At the same time, age at onset appears to represent an independent variable that determines the course of disease. Progression in patients with multiple sclerosis with onset as children tends to overlap with that in which onset is in adulthood, although the onset of progression generally occurs 10 years younger in patients with pediatric-onset disease compared with adults (70). This suggests that multiple sclerosis in children represents a disease process that overlaps with that in adults. It is possible that there are unique precipitating factors in children. At the same time, distinctions pertinent to the disease in children may be a function of age. For example, work on the microbiome in children with multiple sclerosis, indicates that the gut microbiota in children with multiple sclerosis is different from age matched controls without multiple sclerosis, allowing some insight into the mechanism of disease close to the onset of disease (78). Different amino acid tryptophan metabolism pathways resulting in changes in the levels of distinct tryptophan metabolites may affect the incidence and the disease activity of pediatric onset multiple sclerosis. Higher levels of tryptophan and indole lactate, a metabolite of tryptophan, produced by intestinal bacteria within the gut decreased the risk of developing pediatric multiple sclerosis. Every 1 mcg/mL increase in serum tryptophan was associated with a 20% reduction in the adjusted odds of developing multiple sclerosis (95% CI: 4%-34%, p = 0.017). Furthermore, after the adjustment for age, sex, race as well as ethnicity and the serum 25-hydroxy vitamin D levels, an increase in the serum tryptophan level of 1 mcg/mL was associated with a 34% reduction in the risk of developing multiple sclerosis (95% CI: 16% to 44%, p < 0.001). On the other hand, increased levels of kynurenine, which is not produced by the gut microbiota but via liver-derived enzymes, such as tryptophan 2,3-dioxygenase or indoleamine 2,3-dioxygenase enzymes, were associated with an increased relapse rate, with incident rate ratio of 5.8 (p =0.003) (35; 66). That the metabolites of the tryptophan pathway, such as kynurenine, may be associated with increased relapses, is in line with the idea that the site of metabolism, such as the liver versus the gut, may determine whether any metabolites may be a cause of relapses in multiple sclerosis.

Pathogenesis and immunobiology

Multiple sclerosis pathogenesis involves immune cell responses, such as T-cell and B-cell autoreactivity, as well as innate immune responses involving macrophages and resident central nervous system cells, including microglia, and central nervous system involvement, including neurons, astrocytes, oligodendrocytes, axons, and their myelin sheaths (88). Differentiation between acute multiple sclerosis and acute disseminated encephalomyelitis does not appear clear-cut based on pathological findings. A retrospective cohort study of perivenous demyelination cases at the Mayo Clinic showed that both confluent demyelination characteristics of acute multiple sclerosis lesions and perivenous demyelination typically associated with acute disseminated encephalomyelitis were found on brain biopsy or autopsy samples in 3 of the 13 perivenous demyelination cases, indicating that these disorders overlap pathologically (90). The genes related to susceptibility to multiple sclerosis are various in nature and correspond to the risk for developing autoimmunity among other genes that are important for the microglial activation (45). Lesions in children tend to be more inflammatory, with more edema in those lesions than those of adults (22). Diffusion tensor imaging indicates early changes in neuronal connectivity in pediatric multiple sclerosis (83).

Nevertheless, there are some important biological differences between pediatric and adult multiple sclerosis that may affect disease severity and response to treatment. Importantly, in recent years, distinction between demyelinating syndromes of the central nervous system has led to five main phenotypes: multiple sclerosis, AQP4-positive neuromyelitis optica spectrum disorders (NMOSD), myelin oligodendrocyte glycoprotein associated disorders (MOGAD), autoimmune GFAP astrocytopathy, and antibody-negative demyelinating syndromes. The proportion of patients positive for MOG-IgG appears to be higher in pediatric cohorts compared to adults (16; 56; 83). Furthermore, myelin oligodendrocyte glycoprotein appears to predict a disease course other than that of multiple sclerosis (40; 33). Presentations of MOGAD vary with age, with older children more likely to present with optic neuritis or transverse myelitis versus acute disseminated encephalomyelitis more commonly in younger children. The proportion of pediatric patients with CNS demyelinating syndrome who had anti-MOG antibodies was about one third. The percentage of these individuals who then became seronegative for the anti-MOG antibody was about 57%, typically within 1 year. This was the median time to conversion to seronegative in a prospective multicenter cohort study conducted between 2004 and 2017 (85). About 53% of pediatric patients with CNS demyelinating syndrome who had anti-MOG antibodies had full resolution of their baseline MRI changes and only 38% who remained seropositive for anti-MOG antibodies subsequently developed a clinical relapse, indicating that the majority of individuals with anti-MOG antibodies in this cohort study had a monophasic course of illness (85).

On a cellular level, there is an increased proportion of memory T cells in the setting of reduced percentages of naïve T cells in patients with pediatric-onset multiple sclerosis compared with healthy controls of the same age, similar to adult multiple sclerosis (09). However, alterations in B cell subsets during acute relapses in pediatric onset multiple sclerosis differ from those in adult multiple sclerosis. Pediatric onset multiple sclerosis is characterized by an increase in the proportions of plasmablasts as well as non-class switched memory B cells in the cerebrospinal fluid compared with the presence of primarily class-switched memory B-cells and plasma cells in adult patients with multiple sclerosis (73). Furthermore, there are some data that suggest that cerebrospinal fluid plasmablasts are associated with increased IgG synthesis and acute central nervous system inflammation as demonstrated by an increase in the number of gadolinium-enhancing lesions on brain MRI in patients with multiple sclerosis (21). The increased percentages of cerebrospinal fluid plasmablasts in patients with pediatric-onset multiple sclerosis are consistent with the finding of increased central nervous system inflammation in childhood multiple sclerosis compared with adult multiple sclerosis.

There is a greater likelihood of finding neutrophils in the CSF of pediatric patients with multiple sclerosis 11 years of age or younger (23). Although the role of neutrophils has not been clearly defined in multiple sclerosis pathogenesis, there is some evidence from animal studies that CXCR2-positive neutrophils may be essential for mediating central nervous system demyelination in the cuprizone model of demyelination and in the animal model of experimental autoimmune encephalomyelitis (52). Furthermore, neutrophil infiltration controlled by opposing effects of IL-17 and interferon-gamma cytokines on CXCL2-induced neutrophil recruitment is necessary for brain but not spinal cord inflammation in the experimental autoimmune encephalomyelitis model (75). It is possible that neutrophils may play an important role in central nervous system demyelination and that cytokine-specific regulation of neutrophil infiltration is required for the localization of inflammation to specific regions of the central nervous system at the biological onset of multiple sclerosis.

Gray matter atrophy has been found to be associated with disability in adult multiple sclerosis (34). A study looking at cortical lesions in pediatric patients with multiple sclerosis showed that there are relatively fewer cortical lesions in pediatric compared to adult patients with multiple sclerosis (01). However, some studies have shown that pediatric patients with multiple sclerosis have significant levels of grey matter loss in the thalamus, which correlates with cognitive impairment but does not appear to be associated with physical disability as measured by the Expanded Disability Status Scale (57; 77). Acute axonal damage is increased by more than 50% in early active demyelinating lesions in pediatric patients compared to adult patients with multiple sclerosis (69).

There is some evidence that there is greater central nervous system remyelination in early versus chronic multiple sclerosis, although the mechanism through which this occurs is not well delineated (38). Abnormal central nervous system remyelination in multiple sclerosis may involve dysregulation of the Wnt pathway. Signaling through the Wnt/beta-Catenin pathway, known to be critical in anterior-posterior patterning in vertebrates, is thought to play an important role in normal developmental myelination in mammals (32). Increased levels of Wnt pathway mRNA and proteins have been found in multiple sclerosis lesions, suggesting that this pathway may also be important in inhibiting remyelination in multiple sclerosis (32). Impaired axonal regeneration in multiple sclerosis may be multifactorial, involving both extrinsic factors, including myelin-associated inhibitors such as NOGO, MAG OMgp, lipid sulfatide, and chondroitin sulfate proteoglycan; signaling pathways that may transduce inhibitory signals for axonal regeneration, and intrinsic factors that may be age-dependent (18; 68).

Clinical applications

Monitoring of clinical disease activity and the use of effective immune-based therapies are important for preventing long-term disability in multiple sclerosis patients. Consequently, delays in treatment with disease-modifying drugs may contribute to the greater number of relapses in pediatric multiple sclerosis compared to adult multiple sclerosis. However, a study demonstrated that pediatric patients with multiple sclerosis have a greater number of relapses than adult patients with multiple sclerosis after controlling for treatment with disease-modifying medications, suggesting that there may be increased central nervous system inflammation in pediatric patients compared with adults with multiple sclerosis (39).

Phase III studies of therapeutic agents in patients with multiple sclerosis have excluded individuals older than 10 years of age and up to 18 years of age in the past, but not recently. However, there is currently considerably less information regarding the efficacy and safety of immune-based agents in children compared to adults with multiple sclerosis. Furthermore, treatment options for acute exacerbations are the same in both pediatric patients and adults with multiple sclerosis, with the use of intravenous corticosteroids as first-line therapy and the addition of plasmapheresis for patients with acute exacerbations that are refractory to intravenous corticosteroids.

Although some have used IVIG for the treatment of acute exacerbations in pediatric patients with multiple sclerosis, the bulk of the literature suggests that adult patients with multiple sclerosis have no benefit from IVIG for an acute relapsing event. There is some evidence that the newer disease-modifying medications, such as fingolimod, teriflunomide, and dimethyl fumarate, may be better as first-line drugs for the initial treatment of multiple sclerosis of pediatric age onset based on clinical trials and an observational study (49). This finding contrasts with the multiple sclerosis literature in the past, which advised the use of first-line agents, interferon beta-1a, interferon beta-1b, and glatiramer acetate. These injectable medications were originally used as first-line therapies for both pediatric-onset and adult multiple sclerosis. Further studies are needed to look at these treatments in a meaningful manner through randomized, prospectively designed clinical trials in order to determine whether use of the newer agents may be truly as safe as use of the original injectable first-line medication treatments in pediatric-onset multiple sclerosis. Typical disease-modifying therapies are not approved by the Food and Drug Administration in the United States for use in patients older than 10 years of age with pediatric multiple sclerosis who present with a clinically isolated syndrome suggestive of multiple sclerosis, with the exception of fingolimod, which was approved in May 2018. One reason may be the possibility that an acute syndrome suggestive of ADEM versus MOGAD or NMOSD is in the differential diagnosis at the initial clinical presentation and may be hard to distinguish in the patients with pediatric-onset multiple sclerosis relative to adult patients with multiple sclerosis. Therefore, treatments are more widely available for adult patients who present with a clinically isolated syndrome suggestive of multiple sclerosis.

PARADIGMS, a phase 3, double-blind study of fingolimod that enrolled 215 pediatric patients with multiple sclerosis, showed that fingolimod treatment for up to 2 years reduced the annualized relapse rate and new or newly enlarged T2-weighted MRI lesions compared to interferon beta-1a, including in treatment-naive and younger patients (25). The adjusted annualized relapse rate was 0.12 with fingolimod and 0.67 with interferon beta-1a, corresponding to an absolute difference of 0.55 relapses and a relative difference of 82%. There were a significant number of mild adverse events, such as leukopenia and elevated transaminases, that occurred in the individuals treated with fingolimod. A few individuals in this clinical trial of fingolimod versus placebo had seizures separate from other adverse events typically seen in adults.

TERIKIDS was a randomized, double-blind, parallel group, phase 3 placebo-controlled trial of over 55 centers that enrolled 166 individuals between the ages of 10 and 17 diagnosed with pediatric-onset multiple sclerosis to study teriflunomide against placebo; the trial lasted nearly 96 weeks in the majority of patients (26). A significant number of individuals switched to the 192-week open-label extension trial period, in which all received teriflunomide, prior to completing the entire duration of the 96-week double-blind study. This switch was possibly due to abnormal MRI imaging findings discovered during the trial and may have led to a reduction of the power of their clinical study results as well as the lack of a statistically significant difference between two groups in the primary endpoint—time to first confirmed clinical relapse. The hazard ratio was 0.66, with a 95% confidence interval of 0.388 to 1.113, p = 0.29. Nevertheless, the secondary endpoint results indicated a reduction in the number of new or expanding T2-weighted lesions on MRI, with a relative risk ratio of 0.45 and 95% confidence interval of 0.29 to 0.71, p = 0.00061. This secondary endpoint finding led to the approval of teriflunomide for use in pediatric multiple sclerosis in European countries but not in the United States due to safety concerns; specifically, the trial showed various adverse events, some serious, including an increase in acute pancreatitis requiring at least two to three individuals to stop their study treatment.

FOCUS, a phase II trial, evaluated the safety, efficacy, and pharmacokinetics of dimethyl fumarate in 22 pediatric patients aged 10 to 17 years with relapsing-remitting multiple sclerosis (04). The primary endpoint was the change in T2 hyperintense lesion incidence from the baseline period to the final 8 weeks of treatment. Secondary endpoints were pharmacokinetic parameters and adverse event incidence. Dimethyl fumarate treatment was associated with a reduction in magnetic resonance imaging activity in pediatric patients with pharmacokinetic and safety profiles consistent with those in adults.

The CONNECT study, an open-label, randomized, rater-blinded, active-controlled, phase 3 clinical trial that enrolled 150 patients with pediatric-onset multiple sclerosis conducted in various centers throughout the world, studied the effect of dimethyl fumarate compared with interferon beta-1a during a 96-week period in individuals ranging between 10 and 17 years of age diagnosed with pediatric-onset multiple sclerosis (82). The trial demonstrated that the annualized rate of relapse was 0.24 at 96 weeks, with a 95% confidence interval of 0.15 to 0.39 for patients treated with dimethyl fumarate, compared with an annualized relapse rate of 0.53, with a 95% confidence interval of 0.33 to 0.84, for interferon beta. This resulted in a rate ratio of 0.46, with a 95% confidence interval of 0.26 to 0.80, p = 0.006, for dimethyl fumarate compared with interferon beta-1a. However, the hazard ratio for risk of relapse for dimethyl fumarate compared with interferon beta-1a was not statistically significant, p > 0.05.

Other secondary endpoints included the number of T2 lesions and proportion of patients who were relapse free. Approximately 66.2% of those treated with dimethyl fumarate were relapse free at 96 weeks compared with 52.3% in those treated with interferon beta-1a. Nevertheless, the primary endpoint was the proportion of patients who did not have any new or any enlarging T2 hyperintense lesions. Here, the study showed that 16.1% (95% CI: 8% to 27.7%) of those treated with dimethyl fumarate who completed the trial had no newly enlarging or new T2 hyperintense lesions, whereas only 4.9% (95% CI: 0.6% to 16.5%) of those treated with interferon beta-1b had none of these types of MRI lesions. The study did not have sufficient power to detect statistically significant differences between these two groups for the primary endpoint. No major differences were noted for the number of treatment-emergent adverse events between the two groups in the CONNECT study since it started in August 2014 to its completion in November 2020, per the authors of the article regarding this clinical trial.

Natalizumab has also been used in the past to treat pediatric multiple sclerosis refractory to first-line disease-modifying therapies. Based on the experience of 24 pediatric patients with multiple sclerosis in a network of pediatric multiple sclerosis centers in the United States, the majority of patients showed good response to natalizumab after 18 months of treatment based on MRI and clinical measures of disease activity (89). A prospective cohort study of 55 pediatric patients with multiple sclerosis in Italy receiving a median of 26 monthly infusions of natalizumab showed that the mean expanded disability status scale score decreased from 2.7 to 1.9 (p < 0.0001); the majority of patients developed no new T2 or gadolinium enhancing lesions on MRI; and clinical adverse effects were relatively mild (37). Although 21 out of 50 patients tested serologically were positive for the JCV antibody, there were no confirmed cases of progressive multifocal leukoencephalopathy. Although it is possible that pediatric patients with multiple sclerosis have a lower risk of developing progressive multifocal leukoencephalopathy than adults, further studies are necessary to determine the risk of progressive multifocal leukoencephalopathy in children with multiple sclerosis.

A clinical study is currently looking at not just the safety and pharmacokinetics, but also the efficacy, of ocrelizumab, a humanized monoclonal antibody directed against CD20, a marker for B cells, via intravenous infusion given every 24 weeks compared to fingolimod 0.5 mg taken daily via oral route in patients aged 10 to 17 with pediatric multiple sclerosis. A previous small, retrospective study evaluating rituximab, a chimeric monoclonal antibody targeting CD20 on B cells, for treatment in eight patients with NMOSD and three patients with pediatric-onset multiple sclerosis showed major improvement in reducing relapses in 9 out of 11 patients, including two patients with pediatric-onset multiple sclerosis who remained relapse free on follow-up, retrospectively suggesting that targeting B cells may be efficacious in preventing relapses (12; 36). Rituximab is the only B-cell depletion therapy that has been approved by the FDA (December 2021) for the treatment of B-cell lymphomas and leukemia in pediatric patients between 6 months and 18 years of age, placing this drug in the treatment armamentarium for pediatric-onset multiple sclerosis.

In contrast to B-cell depletion therapies, cyclophosphamide has been reviewed in patients with aggressive disease courses in pediatric-onset multiple sclerosis and has been shown to reduce the relapse rate and prevent further disability only 1 year after starting treatment in the majority of pediatric multiple sclerosis patients with severe disease refractory to first-line disease-modifying treatments (54). However, there are long-term risks with cyclophosphamide, a drug with a toxicity profile that is severe in adult patients.

Although treatment monitoring with biomarkers such as neurofilament may be used in the future for various disorders, including neurodegenerative disorders such as Alzheimer disease, there is a need for a new biomarker for predicting progression to secondary progressive disease or primary progressive disease in pediatric-onset multiple sclerosis, such as slowly expanding lesions, which determine the progression of multiple sclerosis in patients with relapsing-remitting multiple sclerosis (31; 20). Slowly expanding lesions on MRI may be a biomarker for prognostication in relapsing-remitting pediatric patients in the future.

The distinction between the different CNS demyelinating phenotypes is important as it affects the management of these patients. Although there are currently no evidence-based guidelines for the treatment of patients with MOG-IgG associated disorders, typical immunomodulatory treatments used in multiple sclerosis appear to be ineffective in MOG-IgG-associated disorders, whereas azathioprine, mycophenolate mofetil, rituximab, and, in particular, IVIG appear to be associated with a reduction in relapse frequency (41).

The relapsing-remitting course of multiple sclerosis in childhood may be slower compared to adults. A natural history study of pediatric multiple sclerosis showed that the time to secondary progression is longer in pediatric compared to adult multiple sclerosis, although age at onset of secondary progression was earlier in pediatric multiple sclerosis (70). The slower course of pediatric multiple sclerosis may, in part, be due to greater potential for central nervous system remyelination and axonal regeneration in pediatric compared to adult multiple sclerosis. However, it is unclear whether repair and growth are the main causes for the slow progressive nature of pediatric-onset multiple sclerosis. The presence of edema, which may resolve quickly in the central nervous system in childhood-onset multiple sclerosis, may correlate with the disappearance of MRI lesions in this group of patients (22). Furthermore, the relative lack of physical disability in pediatric multiple sclerosis compared with adult multiple sclerosis may be associated with the relative lack of cortical lesions in pediatric multiple sclerosis (01).

Multiple sclerosis in children may represent a form of the disease close to its onset and, in this context, may allow insights into the pathogeneses. It is likely that the initial inflammatory CNS event occurs as a parainfectious disorder, possibly involving various infectious agents in some patients. In this context, the initial episode of CNS inflammation may occur as an episode of molecular mimicry. Subsequent episodes of inflammation arising from this parainfectious repertoire occur in individuals rendered susceptible to multiple sclerosis. There is evidence of molecular mimicry of the Epstein-Barr virus EBNA1 and glial cell adhesion molecule. A study by a group at Stanford University, including Tobias Lanz, showed that there is evidence of molecular mimicry, suggesting that Epstein-Barr virus may play a causal role in the etiology of some adult-onset multiple sclerosis subsets (51). Samples from the CSF of nine patients with multiple sclerosis in age groups 16 to 20, 21 to 25, 26 to 30, 36 to 40, and 41 to 45 included antibodies directed against viruses from the largest B-cell clonal expansions associated with multiple sclerosis. They were used to create recombinant antibodies. Using protein microarray-based studies, a cross-reactive antibody recognized EBNA1 and glial cell adhesion molecule. Proteolipid protein amino acid 139-151 experimental autoimmune encephalomyelitis mice immunized with EBNA1 amino acid 386-405 versus scrambled control peptide showed that EBNA1 amino acid 386-405 increased the severity of experimental autoimmune encephalomyelitis with worsening weakness, greater infiltration of immune cells, and worse areas of demyelination in the spinal cord. Thus, molecular mimicry is a potential cause for certain forms of experimental autoimmune encephalomyelitis as well as certain types of human multiple sclerosis. Additional studies may provide further insight into the development of other directed treatments in patients with multiple sclerosis by using targeted therapies against EBNA1, which is just a transcription factor in Epstein-Barr virus DNA, as well as therapies directed against glial cell adhesion molecule found in oligodendrocytes as there was evidence of cross reactivity between the two proteins. Epstein-Barr virus could play a very important role in the early stages of multiple sclerosis as EBNA1 antibodies were found in CSF B-cell plasmablasts of nearly 25% of patients with multiple sclerosis.

Multiple sclerosis is considered rare in children, in part, because it is underdiagnosed and often mistaken for other diseases mimicking multiple sclerosis. Furthermore, age-dependent changes in the immune responses to environmental triggers and greater plasticity of the developing target central nervous system in children may also contribute to the decreased incidence of pediatric compared to adult multiple sclerosis. Modifying risk factors in childhood through the supplementation of vitamin D, immunization by vaccination against Epstein-Barr virus, treatment with anti-Epstein-Barr virus therapies, as well as B-cell depletion therapies that are effective in controlling Epstein-Barr virus infection, in addition to promoting tobacco cessation, weight reduction to prevent obesity, and dietary changes to modify the gut microbiome to decrease central nervous system inflammation, are strategies that can be developed to prevent multiple sclerosis, augmenting current treatment strategies of reducing central nervous system inflammation to stop further disease activity and progression.

Disclaimer. The views expressed in this article do not necessarily reflect the views of the Department of Veterans Affairs or the United States Government.