Importance听 While the typical onset of multiple sclerosis (MS) occurs in early adulthood, 2% to 10% of cases initially present prior to age 18 years, and approximately 5% after age 50 years. Guidance on approaches to differential diagnosis in suspected MS specific to these 2 age groups is needed.

Observations听 There are unique biological factors in children younger than 18 years and in adults older than age 50 years compared to typical adult-onset MS. These biological differences, particularly immunological and hormonal, may influence the clinical presentation of MS, resilience to neuronal injury, and differential diagnosis. While mimics of MS at the typical age at onset have been described, a comprehensive approach focused on the younger and older ends of the age spectrum has not been previously published.

Conclusions and Relevance听 An international committee of MS experts in pediatric and adult MS was formed to provide consensus guidance on diagnostic approaches and key clinical and paraclinical red flags for non-MS diagnosis in children and older adults.

Multiple sclerosis (MS) typically presents in early or middle adulthood (adult-onset MS [AOMS]), while onset before age 18 years (pediatric-onset MS [POMS]) and at or after age 50 years (late-onset MS [LOMS]) is less common (Figure). The proportion of patients with POMS has been estimated to be approximately 2% to 10%^1-3 of all MS cases, with onset before age 10 years at only between 0.2% and 0.7%.^1,2,4 Incidence of POMS from 2003 to 2019 was 0.95 to 0.98 per 100鈥�000 person-years in Canada,⁵ while in Sweden, incidence of POMS from 2006 to 2016 was 1.12 per 100鈥�000 person-years.⁶ Due to different age cutoffs and evolving MS diagnostic criteria, the proportion of LOMS reported in the literature varies between 1% to 21%.^7,8 A meta-analysis⁹ from 2021 estimated that 5% of the MS population has clinical onset after the age of 50 years. Incidence of LOMS from 2000 to 2009 was 4.43 for men and 7.47 for women per 100鈥�000 person-years in Denmark.¹⁰ Contemporary data from the Danish Multiple Sclerosis Registry of 10鈥�120 persons with MS with onset between 2003 and 2022 indicates that 2.89% had onset before age 18 years, and 14.3% had onset at or after age 50 years (unpublished data, Figure). Unique diagnostic challenges distinguish clinical presentations outside of typical age at onset for MS.

In POMS, diagnostic dilemmas may include differentiating MS from monophasic acute disseminated encephalomyelitis (ADEM) and myelin oligodendrocyte glycoprotein antibody associated disease (MOGAD),¹¹ both of which most commonly occur among children and share some clinical and imaging features with MS. In the differential diagnosis, inherited leukodystrophies should also be considered, although the progressive nature of these conditions would be highly atypical for POMS.¹² Genetic or acquired disorders of immune dysregulation, such as hemophagocytic lymphohistiocytosis, can present with relapsing neurological manifestations that can be initially difficult to distinguish from MS inflammatory attacks. As patients with POMS have higher relapse rates¹³ and magnetic resonance imaging (MRI) inflammatory disease activity¹⁴ at onset compared to AOMS, timely accurate diagnosis is important to allow early initiation of disease-modifying therapy as well as for providing appropriate pediatric specific support. Despite the fact that progressive-onset MS is rarely seen in childhood¹⁵ and the transition to the secondary progressive phenotype occurs after a longer interval, patients with POMS still reach ambulatory disability milestones at younger chronological ages than AOMS owing to the earlier onset.^16,17

At the other end of the age spectrum, LOMS can include both those who experience their first inflammatory demyelinating event as well as those who have progressive onset of neurological symptoms after age 50 years. Diagnostic considerations in LOMS differ from those in younger patients and include increased prevalence of cerebrovascular disease associated with brain MRI white matter hyperintensities (WMH). Correct diagnosis is extremely important in older individuals, as the decision to start disease-modifying therapy in LOMS is complicated by evidence suggesting disease-modifying therapies may be less effective or unnecessary with older age while the risk of adverse effects increases.^18,19 On this background, we provide practical approaches to distinguish disorders that most commonly require consideration in the differential diagnosis of MS based on clinical presentation at both ends of the age spectrum, highlighting potential diagnostic red flags and testing algorithms for POMS and LOMS.

POMS was defined as clinical MS onset prior to the age of 18 years, AOMS as onset between ages 18 and 49 years inclusive, and LOMS as onset at or after age 50 years. The method of consensus-building was modeled on the consensus development panel²⁰ approach but modified to use televideo technology and electronic communication. A committee comprising international experts in pediatric and adult neuroimmunology and MS was constituted, comprising diverse representation of gender, academic rank, geographic region of practice, and clinical focus. Online meetings were completed to develop and reach consensus on the aims and section topics for the present article. Working groups with expertise relevant to specific article elements created and revised drafts of each manuscript article section, table, and figure based on available evidence and expert opinion. Preliminary draft manuscript elements were presented electronically to the entire group of collaborators for additional rounds of feedback and revision until there was unanimous consensus that the content met the goals of the project. A web-based survey, using the Qualtrics platform, was sent to the authors to identify the most frequently encountered MS mimics in their practice for POMS or LOMS. All diagnoses that were identified by more than one-third of the respondents were presented to the authors for additional feedback and revisions to achieve consensus.

A research librarian (M.P.H.) performed a literature search in Ovid Medline from January 1, 2008, to December 31, 2022, using keywords 鈥渕ultiple sclerosis,鈥� 鈥渄iagnostic error,鈥� 鈥渕issed diagnosis,鈥� 鈥渕isdiagnosis,鈥� 鈥渄iagnostic accuracy,鈥� and 鈥渄ifferential diagnosis,鈥� yielding 1430 unique citations. Covidence systematic review management software (Veritas Health Innovation) was used to review each abstract for relevance to the aims of the project. Four hundred seventy-six articles were retained, individually reviewed by author A.J.S., and made available to all authors during manuscript development. Where possible, non-English language abstracts and manuscripts were translated into English by Google Translate and included. Additional relevant literature published after January 1, 2008, was also identified from the authors鈥� files during manuscript development.

Immunological changes that occur throughout the lifespan not only impact the clinical manifestations of MS, such as relapse frequency, severity, and recovery, but also influence differential diagnosis considerations. Children have larger and higher proportions of naive T cells and higher B-cell functional capacities, resulting in more robust immune responses to antigens than adults,^21,22 which may amplify the inflammatory pathology of MS and explain why most POMS cases (>98%) present with a relapsing-remitting course.²³ Compared to AOMS, patients with POMS tend to experience more active disease, distinguished by frequent relapses followed by good recovery.^13,24 In contrast, progressive MS is rarely seen in childhood,^2,16,25 and thus progressive onset in children should lead to careful consideration of alternate diagnoses, such as leukodystrophies or mitochondrial disorders, among other metabolic diseases.^17,26

Immunosenescence of the adaptive immune system is characterized by changes in immune function due to decreases in naive T- and B-cell populations, clonal expansion of memory T cells, and reduction in T-cell receptor repertoire, shift in CD4+/CD8+ ratios, loss of CD28 costimulatory signal expression, reduced antibody diversity and affinity, and increased secretion of proinflammatory cytokines.^27,28 These changes have been associated with increased frequency and severity of infections and reduced vaccine response with age. In patients with MS, with aging, inflammatory focal activity may decrease due to peripheral immunosenescence, accounting for the gradual reduction of both clinical relapse frequency and rate of development of new T2 or gadolinium-enhancing MRI lesions.^29,30 On pathology, LOMS cases had fewer and less active lesions, less leptomeningeal and perivascular inflammation, and greater reduction in neuron density compared to AOMS.³¹ Compared to AOMS, patients with LOMS present less frequently with a relapse-remitting course, ranging from 57%³² to 74% of patients.³³ Older age at clinical onset is associated with an increased risk of experiencing both progressive onset^31,34 and more rapid transition to progressive disease.^{23,25,31,34,35} Age also affects the phenotype of relapses. In a recent study³⁵ using the Italian registry, spinal cord presentations were more common in LOMS compared to AOMS (34.3% vs 26.6%; P鈥�<鈥�.001), and optic nerve presentations were less common (15.7% in LOMS vs 25.6% in AOMS; P鈥�<鈥�.001). Relapses affecting motor symptoms are more common in older patients and are more often followed by incomplete recovery.^36,37 This is likely related to limited neuroplasticity, reduced brain reserve, and remyelination failure in older patients.^38,39 With aging, astrocytes switch to a proinflammatory phenotype,⁴⁰ which can result in loss of neurons and oligodendrocytes.⁴¹ In addition, myelin fragmentation increases with age and can induce microglial senescence and dysfunction.⁴² Immunosenescence may also drive disability progression in MS via increased secretion of proinflammatory cytokines by peripheral leukocytes and microglia.⁴³ Cellular senescence likely contributes to compartmentalized central nervous system (CNS) inflammation, resulting in chronically activated microglia,⁴⁴ which can be found at the rim of slowly expanding lesions. These are a potential counterpart of smoldering plaques with nonresolving inflammation, a pathological phenotype that becomes more evident with older age and in patients with progressive MS.^45,46 These activated microglia contain iron and can be identified on susceptibility-based MRI sequences as paramagnetic rim lesions, which appear to be highly specific for MS and may be a potential biomarker to help differentiate MS from mimics.^47-49 With the reduction in peripheral-mediated inflammation and the increase in compartmentalized CNS inflammation that leads to more progressive phenotypes in LOMS, key MS mimics in LOMS are typically noninflammatory disorders commonly seen among older adults, such as cerebrovascular disease, cervical spondylosis, and CNS neurodegenerative disorders.

Biological sex is associated with differences in immune responses and neurodegeneration, and impacts the risk of developing MS, its clinical presentation, and its rate of progression.⁵⁰ In prepubertal MS, the sex ratio is close to 1:1, while in postpubertal POMS, the sex ratio is similar to that of patients with AOMS (3:1 female to male), supporting hormonal influences on MS suceptibilty.^2,16 Puberty appears to be a risk factor for first presentation, with most girls presenting 2 years after menarche, and for possible increase in relapse activity in both girls and boys. In addition, premenopausal women have more robust peripheral immune responses⁵¹ and blood-brain barrier permeability allowing increased transmigration of autoreactive lymphocytes⁵² likely contributing to higher relapse rates compared to men.^53,54 After menopause, relapse activity becomes less frequent. Men have higher rates of disability progression than women, until the menopausal transition, when both the rate of disability worsening and the onset of progressive MS accelerates in women, and the differences in rates of progression become similar between the groups.^55,56 Additionally, in primary progressive MS, where the mean age at onset is 45 years, the sex ratio is closer to 1:1, which is also attributed to perimenopausal biology.³³ Meanwhile, decreased levels of testosterone in men has been hypothesized to contribute to higher risk of MS.^53,54

In comparison, aquaporin-4 positive (AQP4+) neuromyelitis optica spectrum disorder (NMOSD) has a significantly higher female to male ratio than MS (as high as 3:1 ages <15 years, 23:1 ages 15-40 years, and 5:1 ages >40 years),⁵⁷ but the annualized relapse rate does not appear to be different between typical-onset and late-onset AQP4+ NMOSD.^58,59 There is no known difference in sex ratio for MOGAD. Hormonal changes occurring during the menopausal transition may also influence the risk of noninflammatory MS mimics, such as increases in nonspecific WMH (similar burden in premenopausal female individuals and male individuals, and higher burden of WMH in postmenopausal female individuals compared to similarly aged male individuals).⁶⁰ In contrast, in Leber hereditary optic neuropathy, the male to female ratio is 3:1 between 5 and 45 years, and closer to 1:1 in the younger and older ages, and studies using cybrids have suggested a possible protective effect of estrogen.^61-63

Frequently encountered MS mimics for both POMS and LOMS are presented in eFigures 1 and 2 in the Supplement. The diagnostic approach to POMS and LOMS implies the accurate identification of the clinical presentation to highlight potential atypical features for MS with respect to both younger and older patient populations (see Solomon et al⁶⁴ for a detailed discussion for typical features of AOMS). A comprehensive list of atypical features or red flags with clinical and imaging features that suggest an alternate diagnosis, along with relative importance of these alternate diagnosis for POMS and LOMS are presented, in Tables 1,2,3, and 4,^65-125 categorized by initial anatomical location of symptom onset (optic nerve, supratentorial, brainstem/cerebellar, and spinal cord). More common diagnoses that may still be missed are listed in the eTable in the Supplement.

Acquired demyelinating syndromes include several CNS inflammatory conditions, such as ADEM, MS, MOGAD, and AQP4+ NMOSD (eFigure 3 in the Supplement). MOGAD is more common prepubertally than MS, as such presentations before the age of 11 years constitute a red flag for MS and may be more suggestive of MOGAD. Clinical and neuroimaging features in patients with prepubertal and postpubertal MS are similar to AOMS, except younger patients tend to have a higher inflammatory lesion load at onset.¹²⁶ Advancements in serological testing of MOG鈥搃mmunoglobulin G (IgG) and AQP4-IgG using cell-based assays (CBAs) have improved differentiation of MOGAD and AQP4+ NMOSD, respectively, from MS and each other.^127,128 AQP4-IgG antibodies when present in serum are typically present at onset and remain positive over time. MOG-IgG antibody titers are highest at the time of incident attack in nearly all MOGAD patients, and decline (often rapidly) over time. In both these disorders it is very uncommon to be seronegative at presentation¹²⁸ and in rare cases, antibodies against MOG-IgG are negative in serum but can be detected in CSF.¹²⁹ CSF testing is less specific than serum testing, and thus should only be used when clinical suspicion for MOGAD is high, yet serum testing is negative. Repeat serum testing should be considered when clinical suspicion is strong, or if laboratory testing was performed just after plasma exchange or treatment with intravenous immunoglobulin.

ADEM accounts for 22% to 32% of acquired demyelinating syndromes in children. Its diagnosis requires the presence of encephalopathy with polyfocal neurological deficits in the absence of concurrent active infection or systemic illness, and the presence of diffuse and poorly demarcated hemispheric and often infratentorial brain lesions; while longitudinally extensive spinal lesions and optic nerve lesions can also occur.¹²⁸ ADEM is typically preceded by symptoms of infection (70%-80% of cases) although no specific pathogen has been identified, may be associated with seizures, and is usually monophasic.¹³⁰ Multiphasic ADEM (defined as recurrent attacks, with each attach meeting criteria for ADEM) is extremely rare. In contrast, MS less commonly presents with encephalopathy and seizures. MOG-IgG has been identified in 57% of pediatric patients presenting with ADEM,¹³¹ and such patients are now classified as having MOGAD.⁶⁵ In younger children, ADEM is the most common presenting feature of MOGAD, compared with optic neuritis and/or transverse myelitis in later teenage and adulthood. Prior to routine MOG-IgG testing, different studies have reported 6% to 20% of POMS presenting with an ADEM phenotype as their initial clinical event.^132,133 In addition, atypical MRI findings for MS, such as extensive and poorly defined brain white matter lesions, with marked improvement on follow-up imaging, were previously described in the pre-pubertal MS groups.¹³⁴ Both aspects are now recognized as characteristic radiological features of MOGAD (eFigure 3 in the Supplement).^65,135 Therefore, some pre-pubertal children in previously published pediatric MS series were probably misdiagnosed, and would now be reclassified as MOGAD.¹²⁸ With improved characterization of MOGAD, an initial ADEM presentation in POMS is rare, and would be a red flag for MS. When the initial demyelinating presentation is atypical for MS, MOG-IgG testing, using a validated CBA on serum samples, is necessary, as clear positive titers (鈮�1:100 for fixed assays, or twice the doubling dilution of a positive result for live assays) are highly specific for MOGAD.⁶⁵ On the other hand, low positive titers (<1:100 for fixed assays), particularly in patients with suspected MS or presentation atypical for MOGAD should be interpreted with caution as false positives are more common with lower titers.¹³⁶ In those with low positive titers, additional clinical or MRI features typical for MOGAD should be present.⁶⁵ In the absence of supportive features, consider repeating serum testing, as the titers may rise near a subsequent new clinical event in MOGAD, and/or MRI surveillance, as development of new silent lesions is more typical of POMS, unless an alternative diagnosis is established.^65,136

AQP4+ NMOSD only makes up鈥墌鈥�3% of the pediatric acquired demyelinating syndromes, but is important to recognize given the potential severity of injury associated with acute attacks and the availability of specific effective treatment options.¹³⁷ The 2015 NMOSD criteria are 97% sensitive for pediatric NMOSD.¹³⁸ Compared to MS, AQP4+ NMOSD more typically involves both optic nerves (occasionally simultaneous and mainly the posterior segments) or the optic chiasm, the area postrema, and/or the spinal cord with a longitudinal extensive pattern (eFigure 3 in the Supplement). Serological testing for AQP4-IgG with a CBA helps distinguish AQP4+ NMOSD from MS.¹³⁹

Other frequently encountered diagnoses that may mimic POMS should be considered based on atypical clinical or imaging features of initial anatomical presentations (ie, primarily affecting the optic nerves, brainstem/cerebellar, spinal cord and/or supratentorial regions) (eFigure 1 in the Supplement). For instance, painless sequential bilateral vision loss over weeks to months with pseudo-edema of the optic disc, predominant involvement of the papillomacular bundles, and vascular tortuosity suggests Leber hereditary optic neuropathy, while vision loss with normal fundoscopy and pupillary light response may lead to the consideration of psychogenic visual disturbance. Ataxia in the setting of infections could suggest acute cerebellitis, Bickerstaff brainstem encephalitis, or the start of opsoclonus myoclonus ataxia syndrome. Acute spinal cord symptoms can be related to infectious acute flaccid myelopathy, particularly if there is a known outbreak in the region.¹⁴⁰ Acute encephalopathy with seizures (or delta brush) and orofacial dyskinesias suggests N-methyl-d-aspartate receptor antibody associated encephalitis. As discussed earlier, progressive onset in children would be a red flag. For example, progressive encephalopathy with symmetric confluent white matter involvement would be concerning for inherited leukodystrophy or mitochondrial disorder. Additionally, a strong family history of genetic disorders in the setting of white matter abnormalities would also be a red flag for MS. Genetic and acquired hemophagocytic lymphohistiocytosis is usually associated with pancytopenia, hepatosplenomegaly, and multiorgan involvement, however, primary CNS hemophagocytic lymphohistiocytosis can present with multifocal supratentorial and brainstem lesions without systemic features, which relapse and remit and are steroid sensitive.

When evaluating a patient with suspected LOMS, several clinical and imaging features can help distinguish MS from other conditions (Tables 1,2,3, and 4). The presenting pattern of symptoms can guide the differential diagnosis, as MS-related events typically develop over hours or days, in contrast to acute vascular disease, which usually presents suddenly within seconds or minutes. In addition to clinical features, the location of lesions on imaging can help distinguish infarct from typical MS demyelination. Multiple areas of simultaneous restricted diffusion involving different vascular territories and associated with cortical/subcortical infarcts and leptomeningeal enhancement can suggest primary CNS vasculitis. Vasculitis may be associated with very high protein levels in the CSF, higher than expected for MS, and diagnosis can be aided with MR or conventional angiogram imaging. Vasculitis also can be associated with systemic autoimmune disorders, and systemic symptoms that are unusual for MS can lead to identification of alternate diagnosis such as sicca in Sjogren鈥檚 syndrome, or arthralgias in rheumatoid arthritis and/or systemic lupus erythematosus.

With older age, in addition to increased stroke risk due to large vessel occlusion, patients are more likely to experience subclinical small vessel disease (SVD),¹⁴¹ which can cause WMH that can be misinterpreted as MS lesions. Appearance and anatomical location can help to discriminate demyelinating lesions from chronic vascular abnormalities, as MS MRI lesions are ovoid, are often directly adjacent to the ventricles and surround a central vein, and characteristically involve the corpus callosum and the spinal cord. While vascular WMH are often punctate, mainly involve the anterior subcortical WM of brain hemispheres, may rarely be associated with microbleeds, and do not involve the spinal cord, the corpus callosum, or the juxtacortical white matter (eFigure 4 in the Supplement).¹⁴² Brainstem lesions from SVD typically involve the central pons including the medial lemniscus, while MS lesions are more typically located peripherally.^143,144 Vascular WMH may, however, occasionally appear in typical MS lesion locations (periventricular or brainstem) and both typical vascular WMH and MS lesions may coexist in AOMS and LOMS. Thus, when applying the McDonald MRI criteria for dissemination in space, it may be prudent to consider SVD and more stringently assess lesion features and require more than one periventricular lesion in older individuals.¹⁴⁵ Advanced imaging features like the central vein sign will likely help in distinguishing between MS and non-MS lesions where available.¹⁴⁶ Spinal fluid analysis is also helpful as presence of CSF specific oligoclonal bands would indicate abnormal intrathecal immunoglobulin synthesis supporting a diagnosis of MS over SVD. Kappa free light chains also reflect intrathecal inflammation, with similar sensitivity and specificity to oligoclonal bands, and have the advantages be being more cost-effective and rater-independent.¹⁴⁷

As in POMS, the other frequently encountered differential diagnoses for LOMS can be approached based on symptom localization to the optic nerves, brainstem/cerebellar, spinal cord and/or supratentorial regions, with specific attention to atypical 鈥渞ed flag鈥� presentations and associated clinical and imaging features that suggest an alternate diagnosis in older adults (eFigure 2 in the Supplement). For example, painless acute vision loss with altitudinal field deficits in older adults is more commonly due to anterior ischemic optic neuropathy than MS. Multiple cranial neuropathies can suggest granulomatous disease/neurosarcoidosis. Although progressive myelopathy is a characteristic pattern in LOMS, other diagnosis should be considered particularly if there is a longitudinally extensive spinal cord lesion (LESCL). For example, compression from cervical spondylosis can present with a chronic myelopathy with a LESCL on imaging typically caudal to the compression and associated single flat area of enhancement (鈥減ancake鈥� sign) at the level immediately below the greatest stenosis.¹⁴⁸ Other causes of chronic progressive myelopathy that increase with age and are associated with LESCL include B12 and copper deficiency, homocysteinemia, and dural arteriovenous fistulas. Acute myelopathies that have LESCL include neurosarcoidosis, MOGAD, and APQ4+ NMOSD. Neurosarcoidosis may show a nodular enhancement pattern and have persistent enhancement greater than 3 months. Levels of disease activity can distinguish MS from its demyelinating mimics. In MOGAD and AQP4+ NMOSD, relapses continue into older age, while progressive onset and progression independent of relapses are unlikely to occur.^58,65,135 As in POMS, testing for MOG-IgG or AQP4-IgG using CBA is important. In older patients, more extensive and confluent and relatively symmetric supratentorial white matter lesions, and in some cases with persistent restricted diffusivity, may indicate adult-onset leukodystrophy.¹⁴⁹ Of note, contrast enhancement in MRI can also be seen in the adult-onset X-linked adrenoleukodystrophy, Alexander disease, and leukoencephalopathy with calcifications and cysts.¹⁴⁹ Lesions that persistently enhance in the brain for more than 3 months could suggest an alternate diagnosis, such as primary or secondary CNS tumors or granulomatous disease/neurosarcoidosis.

Aging is associated with increasing risk of comorbidities that can confound the diagnosis of MS, impact its course, or be responsible for clinical worsening misinterpreted as MS-related progression.¹⁵⁰ Certain comorbidities are more prevalent in people with MS, including depression, anxiety, and vascular diseases compared to the general population.¹⁵¹ While the prevalence of comorbid depression and anxiety remains the same across different ages, the prevalence of vascular disease increases with age.¹⁵² Careful clinical assessment is important to differentiate between a typical MS relapse and sudden onset of symptoms characteristic of cerebrovascular diseases. Age is also an important factor when differentiating between paroxysmal MS symptoms¹⁵³ and transient ischemic attacks.¹⁵⁴ Vascular disease can cause silent MRI WMH that may confuse MS diagnosis but also, in people with established MS, may be mistaken for new inflammatory lesions, leading to inappropriate and unnecessary disease-modifying therapy changes. Similarly, there can be an increase in WMH over time in those who have migraines,¹⁵⁵ which can also be misinterpreted as new MS lesions. The risk of polypharmacy increases with age in the general population as well as in patients with MS. Polypharmacy can obscure the clinical presentation, as it can result in various symptoms, such as increased fatigue, cognitive impairment, dizziness, and falls.¹⁵⁶ Discontinuation of medications can at times improve such symptoms. Similarly, conditions such as cervical spondylosis, spinal canal stenosis, and osteoarthritis of the hips and knees can complicate the clinical assessment of persons suspected to have MS. In such cases, surgical interventions can lead to improved functional outcomes.^157,158

While diagnostic criteria for MS perform well across the age span in those with presentations typical of MS, differential diagnoses in POMS and LOMS鈥攖he exclusion of no better explanation鈥攔equire specific considerations in these patients. In young children with incident CNS demyelination, evolving data surrounding MOGAD,¹¹ including advanced imaging techniques, has the potential to further our understanding of key differences in MS vs MOGAD pathophysiology. In older adults, prevalent comorbid vascular disease that confounds diagnostic MRI interpretation may be ameliorated by emerging imaging biomarkers that have demonstrated high accuracy for MS, such as the central vein sign. Higher proportions of central vein sign are found in patients with MS compared to those with mimics.^159,160 MRI paramagnetic rim lesions similarly appear specific for MS in adult populations (although with limited sensitivity), and in preliminary data differentiate pediatric MS from other inflammatory brain disorders.^48,49,161 Both central vein sign and paramagnetic rim lesions can be obtained on a baseline scan to help with diagnosis. Increased recognition of and improved testing for genetic disorders with CNS manifestations^149,162,163 also carry the potential to influence diagnostic approaches to MS in both pediatric and older adult populations. Finally, diagnosis in pediatric and adult populations with incidental MRI findings suggestive of MS will require a cohesive strategy for longitudinal observation until sensitive and specific biomarkers are proven to confirm the diagnosis of MS prior to its clinical manifestations. Inclusion of patients of diverse ages must inform the design of future research evaluating these and other approaches to MS differential diagnosis.

Accepted for Publication: June 13, 2024.

Published Online: September 16, 2024. doi:10.1001/jamaneurol.2024.3062

Corresponding Author: Le H. Hua, MD, Lou Ruvo Center for Brain Health, Cleveland Clinic, 888 W Bonneville Ave, Las Vegas, NV 89106 (hual@ccf.org).

Conflict of Interest Disclosures: Dr Hua reported personal fees from Genentech, Novartis, EMD Serono, TG Therapeutics, Horizon, Alexion, and Genzyme and grants from Biogen outside the submitted work. Dr Solomon reported personal fees from Octave Bioscience, Kiniksa Pharmaceuticals, TG Therapeutics, Horizon Therapeutics, Bristol Meyers Squibb, and EMD Serono and grants from Bristol Meyers Squibb outside the submitted work. Dr Tenembaum reported personal fees from F. Hoffmann-La Roche (member of the Myelin Oligodendrocyte Glycoprotein Antibody-Associated Disease Scientific Committee and teaching webinar on pediatric neuromyelitis optica spectrum disorder) and Alexion Pharmaceuticals (chair of the Neuromyelitis Optica Relapse Adjudication Committee) outside the submitted work. Dr Scalfari reported personal fees from Sanofi, Biogen, Novartis, and Roche outside the submitted work. Dr Rovira reported speaker honoraria and advisory board fees from Novartis , Sanofi, Roche, Biogen, Bristol Myers, and Bayer and serving as chief marketing officer and cofounder of TensorMedical outside the submitted work. Dr Newsome reported grants from Biogen, Genentech, Roche, Lundbeck, National Multiple Sclerosis Society, Department of Defense, Patient Centered Outcomes Research Institute, and Stiff Person Syndrome Research Foundation and personal fees from Biogen, Genentech, Roche, Bristol Meyers Squibb, EMD Serono, TG Therapeutics, and Novartis outside the submitted work. Dr Marrie reported grants from Biogen Idec (coinvestigator; no funds to me or my institution) and Roche (coinvestigator; no funds to me or my institution) outside the submitted work. Dr Magyari reported grants from Merck, Roche, Biogen, and Sanofi and personal fees from Merck, Sanofi, Biogen, Roche, Novartis, and Moderna outside the submitted work. Dr Hemmer received funding from the MultipleMS and WISDOM EU consortium, the Clinspect-M consortium funded by the Bundesministerium f眉r Bildung und Forschung and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany鈥檚 Excellence Strategy within the framework of the Munich Cluster for Systems Neurology (EXC 2145 SyNergy 鈥� ID 390857198). He served on scientific advisory boards for Novartis, AllergyCare, Polpharma, Sandoz and TG Therapeutics; his institution received research grants from Regeneron and Hoffmann LaRoche for multiple sclerosis research. He holds part of 2 patents; 1 for the detection of antibodies against KIR4.1 in a subpopulation of patients with multiple sclerosis and 1 for genetic determinants of neutralizing antibodies to interferon. All conflicts are outside of the submitted work. Dr Hemingway reported grants from the Medical Research Council (academic support), an honorarium into research fund from Roche, and an educational travel grant from UCB outside the submitted work. Dr Graves reported personal fees from Horizon and TG Therapeutics and served on a steering committee with Novartis (fees paid to her institution) outside the submitted work. Dr Bove reported grants from Biogen, Novartis, Roche Genentech, and Eli Lilly and personal fees from Alexion, Amgen, EMD Serono, Sanofi, and TG Therapeutics outside the submitted work. Dr Banwell reported support from Novartis, Sanofi, and Roche outside the submitted work. Dr Corboy reported grants from EMD Serono, Clene Nanomedicine (consultation), and Bristol Meyers Squib (consultation) outside the submitted work. Dr Waubant reported research support from the Department of Defense, National Multiple Sclerosis Society, Patient Centered Outcomes Research Institute, Consortium of Multiple Sclerosis Centers, and Race to Erase MS outside the submitted work. No other disclosures were reported.