Множественный склероз после века 50 Годы: Сравнительный анализ положения с поздним наступлением и взрослым наступлением.

INTRODUCTION

Multiple sclerosis (MS) is a chronic demyelinating disease of the central nervous system associated with neurodegeneration and autoimmunity. The development of new treatment options for immunomodulation and improvements in healthcare in general have resulted in the global mortality of MS decreasing 11.5% since 1990, while the number of older patients is increasing in Europe and North America. 1 MS typically presents as adult-onset MS (AOMS) between the ages of 20 and 40 years. However, disease onset can occur in all age groups, and some patients develop symptoms only after the age of 50 years, which is defined as late-onset MS (LOMS). The proportion of people with LOMS (pwLOMS) has been increasing, with estimates of up to 12% in some cohorts. 1 2

The literature indicates that around 65% of pwLOMS are female, which is lower than the female-to-male ratio in people with AOMS (pwAOMS). LOMS presents more frequently as a relapsing-remitting phenotype, but the rate of progressive forms of disease is increasing. 3 LOMS also seems to progress more rapidly, since the time to reach an Expanded Disability Status Scale (EDSS) score of 6.0 is significantly shorter for pwLOMS than for patients with a lower age at onset. 4 Regarding magnetic resonance imaging (MRI) characteristics, pwLOMS and pwAOMS were found to show no difference in T1-weighted and T2-weighted lesion volumes. 5 Regarding lesion location, the proportion of spinal cord and supratentorial lesions has been found to be higher in pwLOMS. 6

Aging is a physiological process associated with functional decline in various systems. Older people display reduced cerebral plasticity and larger numbers of inflammatory cells traversing the blood–brain barrier, which contributes to incomplete recovery after demyelination and therefore an impaired recovery after disease relapse. Older patients are more likely to exhibit inappropriate immune system activation leading to a chronic inflammatory state that is characterized by axonal loss and neurodegeneration. 7 These pathophysiological changes may explain why progressive forms of disease are more common in older MS patients and also why pwLOMS have a shorter relapsing-remitting phase and fewer relapses. 8

LOMS has not been thoroughly characterized in Portugal. A previous study involving our center included 17 patients. 9 In addition, a more-recent study comparing pwLOMS and pwAOMS at another Portuguese center produced findings consistent with what has been described in the literature. 10 Like for most cohorts related to this subject, that previous study compared pwAOMS and pwLOMS while focusing on their disease durations rather than their actual ages, for which there is a discrepancy. This could result in biased interpretations of clinical and radiological characteristics due to age-related factors and comorbidities. 11 Therefore, given the longevity of current cohorts of people with MS (pwMS), we aimed to 1) describe the sociodemographic, clinical, radiological, and treatment characteristics of MS patients older than 50 years, and 2) perform a comparative analysis between pwLOMS and pwAOMS.

METHODS

Data were collected retrospectively from the clinical records of pwMS under follow-up at University Hospital of São João, which is a tertiary referral center in Portugal. We included patients between the ages of 50 and 75 years. LOMS was defined as disease onset after the age of 50 years and AOMS was defined as disease onset between the ages of 18 and 50 years. Patients without active follow-up between January 2021 and April 2023, whose diagnosis was still uncertain, or whose clinical records lacked sufficient information were excluded. All of the included pwMS were diagnosed according to the 2017 revised McDonald Criteria. 12

This study was conducted after receiving approval from the Ethics Committee of Unidade Local de Saúde São João (approval number: 124/2023). The need for informed consent was waived by the Ethics Committee following national and local regulations regarding retrospective data collection from medical records.

Clinical, sociodemographic, radiological, and treatment data were collected for all included patients. Disease onset was considered as the date of first symptoms or, if that was unavailable, the date of diagnosis. EDSS scores were obtained directly from medical records or calculated from a neurological examination. 13 The EDSS score at the diagnosis was considered as the EDSS score recorded at the first patient encounter that led to the diagnosis (either inpatient or outpatient). Subsequent EDSS scores were obtained from scheduled outpatient appointments. The current EDSS score was defined as the EDSS score at the last visit. EDSS score data from first admissions or relapse visits were not considered for these variables. The first clinical syndrome was categorized according to its likely location within the central nervous system. If a valid location could not be determined, symptoms were categorized as being motor, sensory, or both motor and sensory. When such a classification failed, the presenting features were designated as other unknown symptoms. Patients for whom there was insufficient clinical information about the presenting features were classified as unspecified. Lesion location was obtained from the reports or directly from MRI images when they were available. Information regarding disease-modifying therapies (DMTs) was also obtained, including all DMTs ever received by each patient, as well as the duration receiving each DMT.

We divided approved DMTs according to their efficacy in the treatment of MS. Low-efficacy DMTs included all interferon subtypes, glatiramer acetate, dimethyl fumarate, and teriflunomide. Moderate-to-high-efficacy DMTs included fingolimod, siponimod, ozanimod, cladribine, rituximab, ofatumumab, ocrelizumab, natalizumab, and mitoxantrone. Other DMTs were considered separately. Symptomatic therapies were also considered, namely antidepressants of any kind, antispastics (baclofen and tizanidine), neurogenic bladder treatment (oxybutynin, solifenacin, trospium chloride, and tamsulosin), bowel dysfunction treatment (laxatives, enemas, and antidiarrheal agents), dietary supplements (vitamin D, vitamin B12, and folic acid), pain relievers of any kind (opioids, nonsteroidal anti-inflammatory drugs, and acetaminophen), and fampridine.

Statistical analyses were performed using Stata/MP ® software (version 14, https://www.stata.com/stata14/ ). For the descriptive analysis of categorical variables, absolute and relative frequencies were calculated. After assessing the normality of the distribution, each numerical variable was analyzed using mean±standard deviation or median [interquartile range] values. Inferential analyses were performed using Pearson’s chi-squared test for qualitative categorical variables and using the Wilcoxon rank-sum test for continuous quantitative variables with a skewed distribution.

Ordered logistic regression and linear regression were performed using the current EDSS score and the time to an EDSS score of 6.0 as outcome variables, respectively. The analysis of the time to an EDSS score of 6.0 only considered patients who reached that outcome. Survival analysis for the time to an EDSS score of 6.0 was also performed using Cox regression, including all patients. The covariates are detailed in the Results section. For the ordered logistic regression, to ensure proportional odds, the EDSS-score variable was collapsed using a round-down approach for the steps with a small number of individuals as follows: EDSS scores ≥8 were recoded as EDSS score=8, EDSS score=2.5 was recoded as EDSS score=2, and EDSS score=5.5 was recoded as EDSS score=5. The Stata ® omodel command was used for testing the proportional-odds assumption. Schoenfeld residuals were evaluated for testing the proportional-hazards assumption. For linear regression, the Breusch–Pagan/Cook–Weisberg test for heteroskedasticity was used, and the variance inflation factor was also assessed to test for collinearity.

Propensity-score matching was performed using the Stata ® psmatch2 package, 14 with the current EDSS score as an outcome variable and age and sex as matching variables. Matching was performed without replacement.

RESULTS

Out of 326 initially enrolled pwMS aged between 50 and 75 years, 95 were excluded due to having no follow-up between January 2021 and April 2023 or due to symptom onset occurring before the age of 18 years. One patient was excluded due to insufficient data regarding the date of diagnosis and the date of symptom onset. Therefore, we finally included 230 pwMS in the study analyses.

Demographic data, comorbidities, and disease-related complications are reported in Table 1 . Females predominated in both groups, comprising 70.7% of pwLOMS and 75.1% of pwAOMS. The pwLOMS and pwAOMS were aged 67 [61–70] years and 58 [54–63] years, respectively. The length of education and the employment status did not differ significantly between the two groups. The most-prevalent comorbidities in the LOMS and AOMS groups were vascular risk factors (68.3% and 66.7%, respectively) and psychiatric disorders (39.0% and 39.7%, respectively). The prevalence of comorbidities was not significantly different in the two groups, except for autoimmune diseases, which were more common in pwLOMS (7.3% vs. 1.1%, p =0.01), as further presented in Table 1 . The most-prevalent autoimmune disorder was autoimmune thyroid disease (AITD), which is included under thyroid conditions in Table 1 . Urinary symptoms were the most-common disease-related complication in both pwLOMS (19.5%) and pwAOMS (24.9%).

Table 2 presents disease-specific information. The age at onset was 53 [51–57] years for pwLOMS and 34 [27–41] years for pwAOMS, while the disease durations were 8.25 [2.54–13.52] years and 23.12 [17.54–27.54] years, respectively ( p <0.001). The most-prevalent clinical phenotype was relapsing-remitting MS (RRMS), representing 53.7% of pwLOMS and 59.0% of pwAOMS. The proportion of patients with primary progressive MS (PPMS) was higher for pwLOMS than for pwAOMS (19.5% vs. 8.0%, p =0.03), while there was no significant intergroup difference in secondary progressive MS (SPMS) (22.0% vs. 30.9%, p =0.26). The EDSS score at the diagnosis was higher in pwLOMS (2 [1–4]) than in pwAOMS (1.5 [0–2.5], p =0.03), but there was no significant intergroup difference at the last follow-up (3.5 [1.75–6] vs. 3 [1.5–6], p =0.86). A total of 78 (33.9%) patients achieved an EDSS score of 6.0. The time from diagnosis to achieving an EDSS score of 6.0 was 15.7 [9.0–21.6] years in pwAOMS and 7.7 [5.3–10.8] years in pwLOMS ( p <0.01), which occurred at ages of 53.3 [48.3–60.2] years and 66.0 [58.0–70.8] years, respectively ( p <0.001).

Regarding presenting symptoms, myelitis and motor symptoms at an unknown location were the most-common presenting feature in pwLOMS (19.5%). On the other hand, there were more patients with unspecified symptoms among pwAOMS (36.5%), followed by myelitis (12.7%). The number of relapses throughout the disease course was significantly lower in pwLOMS (1 [0–2]) than pwAOMS (3 [1–6], p <0.001). There was no statistically significant difference in the annualized relapse rate (ARR). Furthermore, the prevalence of cognitive impairment was not significantly different in the two groups ( p =0.47). Lumbar-puncture results were available in the clinical records in significantly more pwLOMS than pwAOMS (90.2% vs. 57.0%, p <0.001). There was no significant difference regarding the presence of oligoclonal bands (OCB) (87.9% and 82.6% in pwLOMS and pwAOMS, respectively) or of an increased IgG index (69.8% and 62.7%, respectively).

Most patients were currently receiving a DMT, with the proportion of current DMTs being higher for pwLOMS than pwAOMS (97.6% vs. 84.1%, p =0.02). pwLOMS and pwAOMS showed equivalent percentages of ever having received a DMT. The number of DMTs prescribed was lower for pwLOMS (1 [1–2]) than for pwAOMS (2 [2–4], p <0.001), while a higher proportion of pwAOMS had a history of using low-efficacy DMTs (94.7% vs. 73.2%, p <0.001), particularly interferon subtypes (84.7% vs. 46.3%, p <0.001). Among patients receiving low-efficacy DMTs, the median duration was also longer in pwAOMS than pwLOMS (15.3 years vs. 6.4 years, p <0.001), despite the percentage of time receiving such DMTs since disease onset being lower for the former (74.0% vs. 93.0%, p =0.03). Regarding moderate-to-high-efficacy DMTs, ocrelizumab and ofatumumab were more frequently used in pwLOMS than pwAOMS (22.0% and 4.9% vs. 7.4% and 0%, respectively). Among patients receiving moderate-to-high-efficacy DMTs, the median duration was longer in pwAOMS than pwLOMS (5.2 years vs. 2.0 years, p =0.01), while the median percentage of time receiving such DMTs since disease onset was higher for the latter (23.3% vs 56.3%, p =0.03). The durations of receiving different categories of DMTs among the entire cohort are presented in Table 3 .

The most-used symptomatic treatment and concomitant therapy in the LOMS and AOMS groups was dietary supplements (53.7% and 59.8%, respectively), followed by antidepressants (34.1% and 50.8%, respectively) and pain relievers (31.7% and 32.3%, respectively). Additional results for the applied treatments are presented in Table 3 .

Table 4 presents the locations of MRI lesions at the time of diagnosis, after 2 years of follow-up, after 5 years of follow-up, and at the time of the most-recent examination. At the time of diagnosis, MRI lesions were mostly in periventricular locations in the LOMS (85.3%) and AOMS (85.7%) groups, followed by subcortical and spinal cord lesions. At that time the proportion of juxtacortical lesions was higher in pwLOMS than in pwAOMS (26.5% vs. 9.1%, p =0.02). After 2 years of follow-up, MRI demyelinating lesions were mostly in periventricular, subcortical, and spinal cord locations in both groups. After 5 years of follow-up, the spinal cord was replaced by the infratentorial region as the third-most-prevalent location of demyelinating lesions in both groups, with the proportion being significantly higher in pwLOMS than in pwAOMS (86.7% vs. 51.9%, p =0.01). At the time of the most-recent examination, most of the MRI lesions were found in periventricular locations in both the LOMS (87.8%) and AOMS (93.8%) groups, followed by subcortical lesions (85.4% and 86.0%, respectively) and infratentorial lesions (73.2% and 71.3%, respectively).

Table 5 presents data obtained by applying different models in ordered logistic regression analyses using the current EDSS score as an outcome. Current age was an independent predictor of the current EDSS score, with older patients presenting with a higher current EDSS score regardless of the LOMS status, sex, disease duration, or time being without treatment or receiving low-efficacy DMTs. A longer disease duration also reflected a higher current EDSS score (odd ratio [OR]=1.04, 95% confidence interval [CI]=1.00–1.07, p =0.025). In addition, a longer time without treatment or receiving low-efficacy DMTs was correlated with a lower current EDSS score (OR=0.98, 95% CI=0.97–0.99, p =0.001). A longer time to start treatment did not appear to add significantly to the model. However, when adjusting for treatment variables, pwLOMS had significantly lower odds of a higher EDSS score. The broadly similar findings of an alternative analysis using a linear regression model are presented in Supplementary Table 1 (in the online-only Data Supplement).

The findings of survival analyses assessing the time to an EDSS score of 6.0 are presented in Table 6 . This analyses indicates that the hazard of an EDSS score of 6.0 was higher for pwLOMS, even when adjusting for age, sex, and treatment variables. The findings of analyses of complementary linear regression models assessing the time to an EDSS score of 6.0 are presented in Supplementary Table 2 (in the online-only Data Supplement). pwLOMS reached an EDSS score of 6.0 faster than pwAOMS regardless of current age, sex, disease duration, or time being without treatment or receiving low-efficacy DMTs. This analysis also revealed a positive correlation between current age and time to an EDSS score of 6.0 in pwMS who reached an EDSS score of 6.0, which may reflect an immortal time bias.

The propensity-score matched analysis, whose findings are presented in Table 7 , revealed that when matching for current age and sex, the current EDSS score remained not significantly different in the two groups.

We also analyzed whether there were any significant changes in the results reported above if the analysis was stratified by MS subtype; that is, RRMS vs. progressive MS (encompassing SPMS and PPMS). There were no significant changes in any of the analyzed variables (data not shown) other than the time to reach an EDSS score of 6.0 (RRMS: n =7, mean difference=-10.47, 95% CI=-31.79 to 10.85, p =0.26; progressive MS: n =65, mean difference=-8.5, 95% CI=-14.45 to -2.57, p =0.006).

DISCUSSION

MS has been traditionally categorized according to different phenotypes and different ages at onset (pediatric, adult, and late), with the disease characteristics differing among these groups. However, the disease concepts of MS have been changing recently, with this being further recognized as one disease along a spectrum of clinical manifestations from acute inflammation to chronic neurodegeneration. 15 pwLOMS may indeed either differ fundamentally from pwAOMS or merely reflect either later overt clinical manifestations or the later diagnosis of a chronic disease. If the latter is true, it would be expected for the clinical characteristics at similar ages to be similar. In our study, we found only minor overall differences among pwLOMS and pwAOMS older than 50 years.

Females constituted 70.7% of pwLOMS and 75.1% of pwAOMS, which are similar to the proportions found in a recent study from Norway. 2 While other studies have found lower female-to-male ratios among pwLOMS, 3 16 17 this ratio did not differ significantly between the groups in the present study.

Since the present cohort included patients at an advanced age, it is expected that comorbidities would have accumulated over time. MS is associated with an increased risk of developing comorbidities, 18 and the most common in both of the present groups were vascular risk factors and psychiatric disorders, which have been described in older pwMS. 19 20 21 Both of these comorbidities are associated with faster disease progression and a higher mortality rate. The prevalence of comorbidities was not statistically different in the two groups, except for autoimmune diseases, which although uncommon, were more common in pwLOMS. The co-occurrence of other inflammatory diseases in pwMS has been widely reported in the literature, with the most common being type 1 diabetes and AITD in most reports. 22 This co-occurrence could be explained by overlapping pathological mechanisms such as genetic, environmental, and immunological factors favoring the concept of polyautoimmunity in MS and other autoimmune diseases. Regarding disease-related complications, urinary symptoms were the most prevalent in both groups, which has also been reported previously in older patients. 20

LOMS was associated with an increased proportion of PPMS, a finding that has been reported previously. 3 16 23 It was interesting that the proportion of people with SPMS was numerically higher in the AOMS group. Combining the SPMS and PPMS categories resulted in no numerical difference between AOMS and LOMS for the progressive forms of the disease. This corroborates previous authors concluding that age—rather than the age at onset—determines the onset of the progressive phase. 24

Our pwLOMS exhibited 1 [0–2] relapses, which is lower than those found in a Turkish cohort (2.24±1.46) and in a Kuwaiti cohort (2 [1–4]). 4 25 This might indicate the presence of under- or overreporting bias in different cohorts. The ARR did not differ between the two groups, a finding that was also found in another study. 26 We did not find differences in cognitive impairment between the two groups. This finding corroborates recent literature supporting that while both age and MS independently affect cognition, this is not significant impacted by the disease duration. 27 28 This contrasts with a recent study finding that LOMS patients performed worse on some cognitive testing measures than did AOMS patients who were matched for either age or disease duration. 29 However, the prevalence of overall cognitive impairment did not differ significantly between the age-matched LOMS and AOMS groups. 29

The proportion of patients with OCB in cerebrospinal fluid (CSF) was not significantly different in the two groups in the study, which is consistent with various other studies that have compared pwLOMS and pwAOMS. 30 31 32 33 However, one LOMS cohort showed a lower OCB positivity. 34 The rates of OCB positivity differed markedly between these studies, which could reflect different laboratory techniques or different timing of lumbar puncture. The rate has varied from 43.8% to 92% for pwAOMS 30 31 and from 46.85% to 98% for pwLOMS. 3 The rates in our study are at the higher end of these ranges for both AOMS (82.6%) and LOMS (87.9%). We could not find studies that compared the proportions of LOMS and AOMS patients with elevated IgG index. However, rates of IgG-index elevation of 66% and 45.84% have been reported, and this was 69.8% in our LOMS patients. 35 36 In summary, there were no differences in CSF parameters between the groups in our study, and the LOMS patients seemed to be at the upper limits of reported proportions regarding both OCB positivity and IgG-index elevation, which is probably due to differences in laboratory and clinical practices between centers.

Among those patients for whom the location of the presenting symptoms was able to be determined, the most-common initial clinical syndrome was myelitis in both groups, which is in agreement with the findings for another Portuguese cohort. 10 However, the rates of myelitis were lower in our study than in that previous cohort: 19.5% vs. 48.7% in pwLOMS and 12.7% vs. 47.4% in pwAOMS. 10 This was probably attributable to a higher proportion of unspecified symptoms and clinical syndromes where a location could not be determined accurately due to the symptoms being localizable to multiple regions or due to incomplete clinical data.

Several studies have found pwLOMS to exhibit faster progression and higher rates of reaching higher EDSS scores relative to pwAOMS. 4 17 23 34 37 38 39 However, the lifetime risk of reaching an EDSS score of 6.0 has been reported to be higher among patients with an earlier onset. 23 40 Some studies have even suggested that a higher age at onset does not confer a worse prognosis, 33 34 39 40 while others have suggested that the disease severity is affected more by the current age than by the age at onset. 41 42

Despite pwLOMS exhibiting higher EDSS scores at the diagnosis, we found that the current EDSS score was not significantly different in the two groups regardless of age, sex, disease duration, and time being without treatment or receiving low-efficacy DMTs. We also found that our pwLOMS reached an EDSS score of 6.0 faster than the pwAOMS following a diagnosis. This suggests that the age at disease onset influences the rates of disability accumulation and progression to significant disability milestones, while the current age appears to determine the overall disability severity that has accumulated. It is also interesting to note that the EDSS score in the pwLOMS in this study at the latest follow-up was lower than that found in a previous study that included patients of the same center. 9 This may reflect better therapeutic options over time and/or diagnoses of milder forms of the disease.

Almost all of the patients in this cohort were receiving a DMT at some point during their disease course (98.3%), which contrasts with 36.4% of patients in an Austrian cohort of pwMS over 55 years of age having never received any DMT. 43 Our pwLOMS comprised a higher proportion of patients currently receiving DMTs (97.6% vs. 84.1%, p =0.02), which is also not commonly described in the literature 17 but is consistent with another LOMS Portuguese cohort. 44 These findings could reflect differences in center-specific practices, as well as the regulations regarding access to DMTs by pwLOMS differing between countries. Furthermore, during the later, less-inflammatory stages, pwAOMS may develop treatment fatigue that leads to DMT cessation in a shared decision-making process. In terms of the different DMTs applied, a higher proportion of pwAOMS had a history of receiving low-efficacy DMTs, particularly the interferon subtypes, with a longer time receiving these DMTs relative to pwLOMS. This might reflect older pwAOMS having had less exposure to high-efficacy DMTs, particularly earlier during their disease course, when therapeutic options were restricted to interferons or glatiramer acetate.

Regarding symptomatic treatment, the most-used medications other than dietary supplements were antidepressants, pain relievers, benzodiazepines, and antispastics in both groups, which is expected for older pwMS. 43

The rate of juxtacortical MRI lesions at the time of diagnosis was higher in pwLOMS than pwAOMS, which has also been reported previously. 6 This finding might be due to the MRI protocols and field strengths available when diagnosing AOMS patients not being as sensitive in detecting juxtacortical lesions. 45 For MRI performed after 5 years of follow-up, infratentorial lesions were more prevalent in pwLOMS. Early infratentorial involvement has been described as a predictor of disability in MS patients. 46 However, and in contrast to our study, there are reports that infratentorial lesions are not frequent in pwLOMS. 6 Spinal cord lesions are also associated with the risk of higher disability severity, 6 but pwLOMS only showed a tendency to have a higher proportion of spinal cord involvement relative to pwAOMS.

We decided to only analyze patients aged between 50 and 75 years. As described above, aging is associated with changes in physiological processes, and it has been demonstrated that disability is predicted more by age than by disease duration. 7 41 Most cohort analyses comparing LOMS and AOMS have not taken our approach, which we believe allowed us to explore the true effect of age at onset while considering the known effects of aging. Comorbidities are also associated with faster disease progression and increased mortality. 19 21 Probably because we compared patients of similar ages, our two groups had similar comorbidity profiles and therefore the comorbidities did not act as a confounding factor in our study.

One limitation of this study was that some variables had missing data, which could have reduced the accuracy of our results. Another limitation is the retrospective design, which means that the risk of bias must be considered. pwAOMS had reduced access to moderate- and high-efficacy DMTs, since these were only introduced in the latter part of the 2000s, while pwLOMS were mainly diagnosed after high-efficacy DMTs became available. Confounding by indication might also have been present, as shown by the counterintuitive finding that the outcome was better for patients who were receiving lower-efficacy DMTs for longer. However, it is likely that these patients had lower disease activity or lower disability, which reduced the likelihood of their treating physicians escalating their therapy.

Notwithstanding the possible presence of biases, our results demonstrate that pwAOMS and pwLOMS within the same age strata exhibit statistically equivalent outcomes. We further showed that, upon diagnosis, LOMS patients progress more rapidly to an EDSS score of 6.0. This probably reflects a reduced neurological reserve in pwLOMS that is attributable to age, longer undiagnosed disease duration, or both of these factors combined.

This study has contributed to improving the understanding of pwLOMS and has reproduced findings in similar cohorts, but it has also contrasted with some of them. 3 9 10 A higher proportion of our pwLOMS had PPMS, and myelitis was the most-common initial clinical syndrome. The EDSS score at the diagnosis was higher in pwLOMS, but the current EDSS score was similar in the two groups, emphasizing the role of age in disability accumulation. 41 There was a high percentage of pwLOMS receiving DMTs in our study, which is uncommon in the literature. In contrast, symptomatic treatment was prevalent, as expected for older pwMS. 43 The MRI lesion location data also replicated findings in the literature regarding juxtacortical involvement at the time of the diagnosis, but diverged regarding early infratentorial involvement. 6

In conclusion, our results show that age is a major driver of disability in pwMS, irrespective of AOMS or LOMS. This further supports the notion that MS exists along a continuum of relapsing-remitting and progressive phases, with the latter ensuing when the neurological reserve is exhausted. 46 47 Therefore, LOMS patients—along with AOMS patients—should be managed with the aim of promoting their neurological reserve through controlling any ongoing inflammatory activity with appropriate DMTs, as well as addressing factors that deplete the neurological reserve such as comorbidities and a lack of physical and cognitive stimulation. 47 48