Abstract
Objective:
To determine the frequency of myelin oligodendrocyte glycoprotein (MOG)-IgG and aquaporin-4 (AQP4)-IgG among pediatric-onset multiple sclerosis (POMS) patients and healthy controls, determine whether seropositive cases fulfilled their respective diagnostic criteria, compare characteristics and outcomes in children with POMS vs MOG-IgG-associated disease (MOGAD), and identify clinical features associated with final diagnosis.
Methods:
POMS patients and healthy controls were enrolled at 14 US sites through a prospective case-control study on POMS risk factors. Serum AQP4-IgG and MOG-IgG were assessed using live cell-based assays.
Results:
AQP4-IgG was negative among all 1196 participants, 493 POMS and 703 controls. MOG-IgG was positive in 30/493 cases (6%) and zero controls. Twenty-five of 30 MOG-IgG positives (83%) had MOGAD while 5/30 (17%) maintained a diagnosis of MS on re-review of records. MOGAD cases were more commonly female (21/25[84%] vs 301/468[64%]; p=0.044), younger age (mean 8.2±4.2 vs 14.7±2.6 years; p<0.001), more commonly had initial optic nerve symptoms (16/25[64%] vs 129/391[33%]; p=0.002) or acute disseminated encephalomyelitis (ADEM) (8/25[32%] vs 9/468[2%]; p<0.001), and less commonly had initial spinal cord symptoms (3/20[15%] vs 194/381[51%]; p=0.002), serum Epstein-Barr virus (EBV) positivity (11/25[44%] vs 445/468[95%]; p<0.001), or cerebrospinal fluid oligoclonal bands (5/25[20%] vs 243/352[69%]; p<0.001).
Interpretation:
MOG-IgG and AQP4-IgG were not identified among controls confirming their high specificity for pediatric CNS demyelinating disease. Five percent of those with prior POMS diagnoses ultimately had MOGAD; none had AQP4-IgG positivity. Clinical features associated with a final diagnosis of MOGAD in those with suspected MS included initial ADEM phenotype, younger age at disease onset, and lack of EBV exposure.
Introduction
Acquired inflammatory demyelinating disorders of the central nervous system (CNS) include multiple sclerosis (MS) and less well-characterized disorders that can mimic MS. Aquaporin-4-IgG (AQP4-IgG) seropositive neuromyelitis optica spectrum disorder (AQP4-NMOSD) is an established CNS demyelinating disease that rarely presents in childhood and has the cardinal manifestations of transverse myelitis, optic neuritis, and area postrema syndrome.1 AQP4-IgG detection with the latest cell-based assays is highly specific (>99%) in adults, although this has been less well-studied in pediatrics.2–4 Myelin oligodendrocyte glycoprotein (MOG) antibody-associated disease (MOGAD), a distinct entity with a predilection for children, has been identified with the use of cell-based assay methodologies using full-length conformational MOG.5–10 MOG-IgG can occasionally be positive, particularly at low titers, in other neurologic disorders including 0.3 to 2.5% of adults with MS; the frequency of MOG-IgG false-positivity appears to be more common in adults than children.11–13 There are, however, few studies assessing MOG-IgG among large numbers of children with CNS demyelination and data on pediatric healthy controls are limited.14,15 Typical pediatric MOGAD presentations consist of demyelinating syndromes including acute disseminated encephalomyelitis (ADEM), optic neuritis and/or transverse myelitis.16 The disease spectrum has been further expanded to include clinical and radiological phenotypes including brainstem syndromes, unilateral cortical encephalitis, other encephalitis manifestations, leukodystrophy-like, and other non-classifiable presentations.17–20
Distinction between AQP4-NMOSD, MOGAD, and MS is crucial given important differences in treatment and outcome. AQP4-NMOSD is generally a relapsing disease which lacks the secondary progression associated with MS, has more severe attacks than MS, and may be exacerbated by MS treatments.21–24 MOGAD can be monophasic or relapsing and also lacks the secondary progression associated with MS.14,25,26 While early disease-modifying treatment in pediatric-onset MS (POMS) is critical to reduce relapse risk and limit disability accumulation, drugs approved for MS are likely not as effective in MOGAD.6
It is therefore important to have a large prospectively characterized cohort of POMS patients with long-term follow-up to determine the frequency of AQP4-IgG and MOG-IgG in these individuals and identify the clinical characteristics, clinical course, and final diagnosis in seropositive vs seronegative children. The objectives of this study were to (1) determine the frequency of MOG-IgG and AQP4-IgG in healthy controls and POMS cases; (2) determine whether seropositive cases fulfilled their respective diagnostic criteria or consensus guidelines;16,17,27,28 (3) compare demographic, clinical, biological, and MRI characteristics, as well as disease course and outcome in children with MS vs children with MOGAD; and (4) identify clinical features associated with final diagnosis.
Methods
Study design and participants
This cross-sectional seroprevalence study of a prospectively collected cohort of patients was carried out as part of a larger investigation of POMS risk factors conducted at 14 pediatric MS centers in the United States (1R01NS071463, PI Waubant).29 Participating centers included: Anne & Robert H. Lurie Children’s Hospital of Chicago, Boston Children’s Hospital, Children’s Hospital Colorado, Children’s Hospital of Philadelphia, Loma Linda University, Massachusetts’s General Hospital for Children, Mayo Clinic Rochester, State University of New York at Buffalo, Stony Brook University Medical Center, Texas Children’s Hospital Baylor, University of Texas Southwestern/Children’s Medical Center Dallas, University of Alabama, University of California San Francisco, and Washington University School of Medicine in St. Louis. The study was approved by the Institutional Review Board at each participating institution and consent was obtained as per individual institutional guidelines.
Patients with a diagnosis of POMS, based on the 2010 McDonald criteria30, or clinically isolated syndrome (CIS)30 with high risk for MS (CIS-MS), defined as having CIS with two or more silent MRI brain or spinal demyelinating lesions with at least one in the brain,29 were enrolled between November 1, 2011, and March 20, 2018. All participants had disease onset before age 18 years and were seen within four years of disease onset. Cases with encephalopathic onset were required to have two or more non-ADEM attacks or one non-ADEM attack with accrual of silent lesions. Cases with known NMOSD or history of organ transplant were excluded. Diagnosis for cases at recruitment was ascertained centrally by a panel of at least two POMS experts. Frequency-matched healthy controls younger than 20 years were recruited from the pediatric general and subspecialty clinics at participating institutions during the same period as the cases. Exclusion criteria for controls included known diagnosis of MS or another demyelinating disease, any autoimmune disease (except asthma or eczema), any chronic neurologic condition with major disability, a biological family member enrolled as a control, an immediate biological family member (sibling/parent) with MS. Controls and cases were matched for race, ethnicity, gender, and windows for age at enrollment (3–10, 11–14 and 15–19 years). Pairwise matching was not used as it would have resulted in a substantial loss of the sample.
Procedures
Clinical features recorded included: age at first demyelination event, first event clinical sign/symptom localizations (optic nerve [i.e. optic neuritis], cerebrum, brainstem/cerebellum, and/or spinal cord [i.e. myelitis]), fulfillment of 2010 McDonald criteria for MS or CIS (yes/no) at the time of first event as well as at recruitment, and working diagnosis at first demyelinating event (ADEM, CIS, MS, radiologically isolated syndrome [RIS], or demyelinating disease not otherwise specified [DDNOS]) as well as at recruitment (CIS-MS or MS). Clinical course was recorded as disease duration and number of clinical events at recruitment, annualized relapse rate (ARR) at recruitment and at last prospective follow-up, duration of follow-up, and final diagnosis at last follow-up. Disability outcome was recorded as Expanded Disability Status Scale (EDSS) score at last follow-up.
AQP4-IgG and MOG-IgG serology analysis
Sera from all cases and controls were collected at time of study enrollment and stored at the University of California San Francisco in −80° F freezers. All specimens were batch analyzed for AQP4-IgG and MOG-IgG at the Mayo Clinic neuroimmunology laboratory (Rochester, MN) in early 2020. The laboratory was blinded to clinical details and the status of samples as cases or controls. AQP4-IgG was assessed using a live cell-based fluorescence-activated-cell-sorting (FACS) assay technique using human embryonic kidney (HEK293) cells transfected with the M1 AQP4 isoform as previously described.4 Sera were screened at 1:5 dilution and the IgG-binding index was calculated by assessing the ratio of median fluorescence intensities of AQP4 transfected cells divided by non-transfected cells. Those with an IgG-binding index ≥2 were repeated at 1:5 dilution and titrated at doubling dilutions with the farthest dilution an IgG-binding index ≥2 recorded as the endpoint of positivity (reference value, <1:5).
MOG-IgG1 testing was performed with a live cell-based FACS assay as previously described.24 Sera were screened at 1:20 dilution and the IgG-binding index was calculated by assessing the ratio of median fluorescence intensities of MOG transfected cells divided by non-transfected cells. Those with an IgG binding index of 2.5 or greater were repeated at 1:20 dilution and titrated at doubling dilutions with the farthest dilution an IgG-binding index ≥2.5 recorded as the endpoint of positivity (reference value, <1:20).
Samples for AQP4-IgG or MOG-IgG that were positive on the initial screen but in which the same sample was negative during the titration step were repeated a third time and those negative on the third assessment were considered a negative result while if positive on the third assessment, the end titer was recorded as their final positive result.
Other studies
For all cases, Epstein Barr virus (EBV) antibody status using the viral capsid antigen32 was obtained. The presence of unique (≥2) CSF oligoclonal bands (OCBs) was recorded, when available. Brain MRI information, including presence of new T2 lesions and new gadolinium-enhancing lesions at around 12- and 24-month (±3 months) follow-up was also recorded, when available. All imaging was performed at the respective recruitment site as either standard radiological follow-up or because of clinic symptoms.
Confirmation of final diagnosis of seropositive cases
All principal investigators (PIs) from participating sites were notified of seropositive cases. The PIs, all of whom are POMS experts, re-reviewed the clinical records and MRIs to confirm final diagnosis. For seropositive cases determined to have MS or unclear final diagnosis, clinical information and MRIs were again reviewed by a panel of four expert investigators (SM, EW, EPF, site PI) and final consensus was reached using current accepted criteria for AQP4-NMOSD,27 proposed consensus criteria/typical clinical and imaging findings for MOGAD16,17,28,33, and expert opinion. Seropositive patients with a final diagnosis of MS were classified as false antibody-positive.
Statistical analysis
Demographic features, family history, and antibody screen results were compared between cases and controls. Next, demographic features and family history as well as clinical, laboratory, and MRI features were compared between MOGAD and all MS cases as well as MOGAD and false antibody-positive MS cases. Continuous variables were assessed using the mean and standard deviation (SD) or median and range. Feature comparisons between groups were performed using chi-squared test or Fisher’s exact test for categorial variables and t-test with unpooled variance estimates for continuous variables. The EDSS and MOG values were assessed by Wilcoxon rank-sum test and the ARR was assessed by logistic regression test from negative binomial regression.
Logistic regression models were used to estimate associations between sex, age at disease onset, first demyelinating disease diagnosis, optic nerve localization at first event, spinal cord localization at first event, EBV status, presence of CSF OCBs, MOG titer and the risk of MOG-IgG true positivity. Unadjusted odds ratios (OR) were calculated with univariable logistic regression. Using a forward-step process, variables with p-values less than 0.1 in the unadjusted model were included in an adjusted multivariable logistic regression. P-values of less than 0.05 were considered statistically significant. OR were calculated with 95% confidence intervals. Analyses were conducted using SAS 9.4 (SAS Institute, Inc., Cary, NC).
Results
Baseline characteristics of cases and controls
We included 1196 participants: 493 with a prior working diagnosis of MS and 703 controls. The mean age at enrollment was 15.4 years (SD 3.0) in cases and 14.9 years (SD 3.9) in controls (table 1). Female subjects were the predominant sex among cases and controls, at 322 of 493 (65%) and 412 of 700 (59%), respectively.
Table 1:
Demographics of controls and cases
| Controls (n=703) | Cases (n=493) | p value | |
|---|---|---|---|
| MOG-IgG screening at enrollment: Positive | 0/703 (0%) | 30/493 (6%) | <0.0011 |
| AQP4-IgG screening at enrollment: Positive | 0/703 (0%) | 0/493 (0%) | ∙∙ |
| Subject age at enrollment, years: Mean (SD) | 14.9 (3.9) | 15.4 (3.0) | 0.0232 |
| Disease duration at enrollment, years: Mean (SD) | ∙∙ | 1.0 (1.0) | |
| Sex: Female | 412/700 (59%) | 322/493 (65%) | 0.0241 |
| Race | 0.2291 | ||
| White | 431/653 (66%) | 316/458 (69%) | |
| Black | 109/653 (17%) | 82/458 (18%) | |
| Asian | 49/653 (8%) | 22/458 (5%) | |
| Other | 64/653 (10%) | 38/458 (8%) | |
| Ethnicity: Hispanic or Latino | 158/667 (24%) | 154/471 (33%) | <0.0011 |
| First-degree family history of autoimmune disease: Yes | 157/703 (22%) | 149/493 (30%) | 0.0021 |
Additional missing values: the variable Age had 3 missing values.
Chi-squared test;
T-test with unpooled variance estimates.
Abbreviations: MOG, myelin oligodendrocyte glycoprotein; AQP4, aquaporin-4.
The mean disease duration at time of enrollment for all cases was 1.0 years (SD 1.0). All 703 controls were negative for both AQP4-IgG and MOG-IgG. AQP4-IgG was negative in all 493 cases at time of enrollment; of these 493 cases, 222 (45%) had also tested negative for AQP4-IgG prior to enrollment. MOG-IgG was positive in 30 of 493 (6%) cases at time of enrollment. Of these 30 cases, 25 (83%) were ultimately assigned a diagnosis of MOGAD and their median titer was 1:80 (range 1:20, 1:1280). These cases were phenotypically similar to other MOGAD patients with relapsing disease previously seen in the United States Network of Pediatric Multiple Sclerosis Centers.34 The remaining five of these 30 cases (17%) were ultimately diagnosed with MS and their median titer was 1:20 (range 1:20, 1:160); these individuals were clinically determined to have MOG-IgG false-positivity.
Demographic and clinical characteristics of final diagnosis of MOGAD and MS cases
Among all cases, 25 had a final diagnosis of MOGAD, 448 MS, and 20 CIS-MS (table 2). The mean age at first demyelinating event was 8.2 years (SD 4.2) for MOGAD and 14.7 years (SD 2.6) for MS or CIS-MS (p<0.001). Female subjects were the predominant sex at 21 of 25 (84%) for MOGAD vs 301 of 468 (64%) for MS or CIS-MS (p=0.044). Presenting symptoms localized to the optic nerve were more common in MOGAD (16/25 [64%] vs 129/391 [33%]; p=0.002). MOGAD cases with optic nerve involvement were more likely to have bilateral simultaneous optic nerve involvement (10/15 [67%] vs 32/125 [26%]), while MS or CIS-MS cases were more likely to have unilateral optic nerve involvement (92/125 [74%] vs 5/15 [33%]) (p=0.005). Presenting symptoms localized to the spinal cord were more common in MS or CIS-MS (194/381 [51%] vs 3/20 [15%]; p=0.002). Patients with MOGAD were more likely to present with an ADEM phenotype (8/25 [32%] vs 9/468 [2%]; p<0.001) at their first demyelination attack. The disease characteristics of the 9 MS/CIS-MS cases with ADEM as first demyelinating attack are summarized in table 3.CSF OCBs were present in five of 25 (20%) MOGAD cases and 243 of 352 (69%) MS or CIS-MS cases (p<0.001). EBV serostatus was positive in 11 of 25 (44%) MOGAD cases and 445 of 468 (95%) MS or CIS-MS cases (p<0.001).
Table 2:
Demographics and disease characteristics of MS and CIS-MS vs MOGAD cases
| MS or CIS-MS (n=468) | MOGAD (n=25) | p value | |
|---|---|---|---|
| Sex: Female | 301/468 (64%) | 21/25 (84%) | 0.0441 |
| Age at first demyelination attack, years: Mean (SD) | 14.7 (2.6) | 8.2 (4.2) | <0.0013 |
| Diagnosis at first demyelination attack | <0.0012 | ||
| ADEM | 9/468 (2%) | 8/25 (32%) | |
| CIS | 169/468 (36%) | 9/25 (36%) | |
| MS | 276/468 (59%) | 7/25 (28%) | |
| RIS | 9/468 (2%) | 0/25 (0%) | |
| DDNOS | 5/468 (1%) | 1/25 (4%) | |
| Clinical localization of first event | |||
| Brainstem/cerebellar | 210/400 (53%) | 9/22 (41%) | 0.2891 |
| Cerebrum | 242/384 (63%) | 13/23 (57%) | 0.5311 |
| Spinal cord | 194/381 (51%) | 3/20 (15%) | 0.0021 |
| Optic nerve | 129/391 (33%) | 16/25 (64%) | 0.0021 |
| Unilateral | 92/124 (74%) | 5/15 (33%) | 0.0022 |
| Bilateral | 32/124 (26%) | 10/15 (67%) | 0.0022 |
| Diagnosis at recruitment | 0.2922 | ||
| CIS-MS | 44/466 (9%) | 4/25 (16%) | |
| MS | 422/466 (91%) | 21/25 (84%) | |
| McDonald criteria fulfilled at recruitment: Yes | 422/466 (91%) | 20/25 (80%) | 0.0922 |
| Disease duration at recruitment: mean (SD) | 1.0 (1.0) | 1.5 (1.3) | 0.0563 |
| CSF OCBs: Present | 243/352 (69%) | 5/25 (20%) | <0.0011 |
| EBV-VCA: Positive | 445/468 (95%) | 11/25 (44%) | <0.0012 |
| MOG titer value: Median (min, max) | 20 (20, 160) | 80 (20, 1280) | 0.0334 |
| MOG titer level: ≥160 | 1/468 (0%) | 11/25 (44%) | <0.0012 |
| Steroid use in 1 month before MOG test: Yes | 145/468 (31%) | 11/25 (44%) | 0.1731 |
| DMT use in 3 months before MOG test: Yes | 237/468 (51%) | 10/25 (40%) | 0.3001 |
| Ever new T2 brain lesions at 12-month follow-up: Yes | 144/243 (59%) | 2/10 (20%) | 0.0141 |
| Ever new enhancing brain lesions at 12-month follow-up: Yes | 83/243 (34%) | 1/10 (10%) | 0.1722 |
| Ever new T2 brain lesions at 24-month follow-up: Yes | 80/199 (40%) | 5/15 (33%) | 0.7862 |
| Ever new enhancing brain lesions at 24-month follow-up: Yes | 48/199 (24%) | 3/15 (20%) | 1.0002 |
| Disease duration at last follow-up, years: Mean (SD) | 3.8 (2.3) | 5.0 (2.7) | 0.0363 |
| ARR at last follow-up: Estimate (CI) | 0.4 (0.35, 0.44) | 0.5 (0.36, 0.83) | 0.1375 |
| EDSS at last follow-up: Median (min, max) | 1.0 (0.0, 7.5) | 1.5 (0.0, 3.5) | 0.0384 |
Additional missing values: Disease Duration at Recruitment 2; Disease Duration at Last Follow-up 2; ARR 8; EDSS 11.
Chi-squared test;
Fisher’s exact test;
T-test with unpooled variance estimates;
Wilcoxon rank-sum test;
LR test from negative binomial regression.
Abbreviations: MS, multiple sclerosis; CIS, clinically isolated syndrome; CIS-MS, clinically isolated syndrome with high risk for MS; MOG, myelin oligodendrocyte glycoprotein; MOGAD, myelin oligodendrocyte glycoprotein antibody-associated disease; ADEM, acute demyelinating encephalomyelitis; RIS, radiologically isolated syndrome; DDNOS, demyelinating disease not otherwise specified; OCB, oligoclonal band; EBV-VCA, Epstein-Barr virus-viral capsid antigen; ARR, annualized relapse rate; SDMT, symbol digit modalities test; EDSS, expanded disability status score; SD, standard deviation; CI, confidence interval.
Table 3:
Demographics and Disease Characteristics of MS/CIS-MS cases with ADEM as first demyelinating attack
| MS/CIS-MS with ADEM at first attack (n=9) | |
|---|---|
| Sex: Female | 3/9 (33%) |
| Age at first demyelination attack, years: Mean (SD) | 10.5 (2.6) |
| Clinical localization of first event | |
| Brainstem/cerebellar | 6/8 (75%) |
| Cerebrum | 7/8 (88%) |
| Spinal cord | 2/5 (40%) |
| Optic Nerve | 1/7 (14%) |
| Unilateral | 0/1 (0%) |
| Bilateral | 1/1 (100%) |
| Diagnosis at recruitment | |
| CIS-MS | 0/9 (0%) |
| MS | 9/9 (100%) |
| McDonald criteria fulfilled at recruitment: Yes | 9/9 (100%) |
| Disease duration at recruitment: Mean (SD) | 2.1 (2.2) |
| CSF OCBs: Present | 5/8 (63%) |
| EBV-VCA: Positive | 8/9 (89%) |
| MOG titer level: ≥160 | 0/9 (0%) |
| Steroid use in 1 month before MOG test: Yes | 4/9 (49%) |
| DMT use in 3 months before MOG test: Yes | 7/9 (78%) |
| Ever new T2 brain lesions at 12-month follow-up: Yes | 2/6 (33%) |
| Ever new enhancing brain lesions at 12-month follow-up: Yes | 2/6 (33%) |
| Ever new T2 brain lesions at 24-month follow-up: Yes | 2/6 (33%) |
| Ever new enhancing brain lesions at 24-month follow-up: Yes | 1/6 (17%) |
| Disease duration at last follow-up, years: Mean (SD) | 5.8 (3.4) |
| ARR at last follow-up: Estimate (CI) | 0.48 (0.27, 0.86) |
| EDSS at last follow-up: Median (min, max) | 1.0 (0.0, 2.0) |
Abbreviations: MOG, myelin oligodendrocyte glycoprotein; MS, multiple sclerosis; CIS-MS, clinically isolated syndrome with high risk for MS; ADEM, acute demyelinating encephalomyelitis;; DDNOS, demyelinating disease not otherwise specified; OCB, oligoclonal band; EBV-VCA, Epstein-Barr virus-viral capsid antigen; ARR, annualized relapse rate; EDSS, expanded disability status score; SD, standard deviation; CI, confidence interval.
Clinical and MRI disease course of final diagnosis of MOGAD and MS cases
The mean disease duration at time of recruitment was 1.5 years (SD 1.3) for MOGAD and 1.0 years (SD 1.0) for MS or CIS-MS. Thirty-one percent of MS or CIS-MS cases and 44% of MOGAD cases received steroids within one month before MOG-IgG testing (p=0.173). Fifty-one percent of MS or CIS-MS cases and 40% of MOGAD cases received a disease modifying therapy (cyclophosphamide, dimethyl fumarate, fingolimod, glatiramer acetate, interferon beta, methotrexate, mycophenolate, natalizumab, peginterferon beta, or rituximab) within 3 months of MOG-IgG testing (p=0.300). At around 12-months follow-up (from disease onset), two of 10 (20%) MOGAD cases had new T2 brain lesions, compared to 144 of 243 (59%) MS or CIS-MS cases (p=0.014), and one of 10 (10%) MOGAD cases had new gadolinium enhancing lesions, compared to 83 of 243 (34%) MS or CIS-MS cases (p=0.172). At around 24-months follow-up (from disease onset), five of 15 (33%) MOGAD cases had new T2 brain lesions, compared to 80 of 199 (40%) MS or CIS-MS cases (p=0.786), and three of 15 (20%) MOGAD cases had new gadolinium enhancing lesions, compared to 48 of 199 (24%) MS or CIS-MS cases (p=1.000). Illustrative examples of MRI lesions of study participants with MOGAD, seronegative MS, and false MOG-IgG-positive MS are shown in figure 1.
Figure 1.
Orbital and brain MRI examples in patients with myelin oligodendrocyte glycoprotein antibody-associated disorder (MOGAD) and multiple sclerosis (MS).
Row 1. A to B3: two patients with MOGAD
Row 2. C to E2: three patients with MS negative for AQP4-IgG and MOG-IgG
Row 3 & 4. F to J: five patients with MS and false positive MOG-IgG and negative AQP4-IgG
(A) Orbital MRI with axial T1-weighted fat-suppression images post-gadolinium reveals bilateral optic nerve enhancement involving the anterior optic pathway (A, arrows) with the left side extending ≥50% of the length of the nerve in a patient with MOGAD. (B) Axial fluid attenuated inversion recovery (FLAIR) images reveal faint left-sided deep grey matter T2-hyperintensities (B, arrowheads) along with some additional white matter T2-hyperintensities (B1, arrows); subsequent Axial FLAIR images in the same MOGAD patient during a separate attack reveal a large, poorly demarcated, expansile T2-hyperintense lesion in the right middle cerebellar peduncle and cerebellum (B2, arrow) that resolved completely in follow-up 9 months later (B3). (C) Orbital MRI with axial T1-weigthed fat-suppression images post-gadolinium reveals unilateral left optic nerve enhancement of the intracranial segment of the optic nerve extending <50% of the nerve (C, arrows) in a patient with MS seronegative for MOG-IgG and AQP4-IgG. (D) Axial FLAIR images reveal multiple periventricular lesions with characteristic morphology for MS (D, arrows) in a patient with MS seronegative for MOG-IgG and AQP4-IgG. (E) Axial FLAIR images reveal a well circumscribed T2-hyperintense lesion in the left lateral pons and middle cerebellar peduncle (E1, arrow) in a patient with MS seronegative for MOG-IgG and AQP4-IgG that persisted 7 months after initial MRI (E2, arrow). (F) Orbital MRI with gadolinium reveals a short segment of enhancement in the left proximal optic nerve (F1, arrow) in a patient with MS with an accompanying low titer MOG-IgG felt to represent a false positive; the MRI brain in this same patient (F) reveals multiple focal white matter lesions with a morphology consistent with MS (F2, arrows). (G) Axial FLAIR reveals innumerable periventricular T2-lesions (G, arrows) some of which are ovoid or perpendicular to the ventricles highly consistent with MS in a patient with low titer MOG-IgG that was felt to represent a false positive result. (H) Axial FLAIR reveals two small foci of T2-hyperintensity (arrows) in the pons and cerebellum in a patient with MS and low titer MOG-IgG felt to be a false positive. (I) Axial FLAIR reveals multiple ovoid T2-hyperintensities, some of which are periventricular in a patient with MS and low titer MOG-IgG felt to be a false positive. (J) Axial FLAIR reveals multiple periventricular T2-hyperintense lesions (J1, arrows) with corresponding T1-hypointensity (J2, arrow) in a patient with MS and MOG-IgG positivity, felt to be a false positive.
Abbreviations: AQP4-IgG, Aquaporin-4-IgG; FLAIR, fluid attenuated inversion recovery; MOG-IgG, myelin-oligodendrocyte-glycoprotein-IgG; MOGAD, myelin oligodendrocyte glycoprotein antibody associated disease; MRI, magnetic resonance imaging; MS, multiple sclerosis.
Multivariate logistic regression model of final diagnosis of MOGAD
Logistic regression analysis demonstrated that the following features were associated with a final MOGAD diagnosis in multivariable (AOR) models (table 4): first demyelinating disease diagnosis of ADEM with AOR of 17.50 (CI 3.20, 95.59; p=0.001), younger age (≤12 years) at disease onset with AOR of 8.32 (CI 2.38, 29.08; p<0.001), female sex with AOR of 4.11 (CI 1.08, 15.68; p=0.038), and presenting symptoms localized to the optic nerve with AOR of 3.37 (CI 0.98, 11.64; p=0.055)). EBV positivity was found to be associated with a decreased risk of MOGAD with AOR of 0.17 (CI 0.05, 0.58; p=0.004).
Table 4:
Odds of MOG-IgG true positivity in cases
| Unadjusted Odds-Ratios (95% CI) | p value | Adjusted Odds-Ratios (95% CI) | p value | |
|---|---|---|---|---|
| Sex | 0.054 | 0.038 | ||
| Female | 2.91 (0.98, 8.62) | 4.11 (1.08, 15.68) | ||
| Male | reference | reference | ||
| Age at disease onset | <0.001 | <0.001 | ||
| Age ≤ 12 years | 26.69 (9.66, 73.74) | 8.32 (2.38, 29.08) | ||
| Age > 12 years | reference | reference | ||
| Diagnosis at first demyelinating attack | <0.001 | 0.001 | ||
| ADEM | 24.00 (8.25, 69.85) | 17.50 (3.20, 95.59) | ||
| Not ADEM | reference | reference | ||
| Optic nerve (localization at first event) | 0.003 | 0.055 | ||
| Present | 3.61 (1.55, 8.39) | 3.37 (0.98, 11.64) | ||
| Absent | reference | reference | ||
| Spinal cord (localization at first event) | 0.005 | ∙∙ | ||
| Present | 0.17 (0.05, 0.59) | ∙∙ | ||
| Absent | reference | |||
| EBV-VCA interpretation | <0.001 | 0.004 | ||
| Positive | 0.04 (0.02, 0.10) | 0.17 (0.05, 0.58) | ||
| Negative | reference | reference | ||
| Ever a CSF record with OCBs | <0.001 | ∙∙ | ||
| Yes | 0.11 (0.04, 0.31) | ∙∙ | ||
| No | reference | |||
| MOG titer | <0.001 | ∙∙ | ||
| ≥ 160 | 366.93 (44.26, >999.99) | ∙∙ | ||
| < 160 | reference |
Abbreviations: MOG, myelin oligodendrocyte glycoprotein; ADEM, acute disseminated encephalomyelitis; EBV-VCA, Epstein-Barr virus-viral capsid antigen; OCB, oligoclonal band.
Characteristics and disease course of final diagnosis of false MOG-IgG-positive MS cases
The characteristics of cases with a final diagnosis of false MOG-IgG-positive MS (n=5) are summarized in tables 5 and 6. OCBs were present in all false MOG-IgG-positive MS cases compared to five of 25 (20%) MOGAD cases. The median MOG-IgG titer was 1:20 (range 1:20, 1:160) in false MOG-IgG-positive MS compared to 1:80 (range 1:20, 1:1280) in MOGAD.
Table 5:
Demographics and disease characteristics of MS with false MOG-IgG-positive
| MS with false positive MOG-IgG (n=5) | |
|---|---|
| Sex: Female | 4/5 (80%) |
| Age at first demyelination attack, years: Mean (SD) | 13.4 (5.0) |
| Diagnosis at first demyelination attack | |
| ADEM | 0/5 (0%) |
| CIS | 1/5 (20%) |
| MS | 4/5 (80%) |
| DDNOS | 0/5 (0%) |
| Clinical localization of first event | |
| Brainstem/cerebellar | 3/4 (75%) |
| Cerebrum | 4/4 (100%) |
| Spinal cord | 2/3 (67%) |
| Optic Nerve | 4/4 (100%) |
| Unilateral | 1/4 (25%) |
| Bilateral | 3/4 (75%) |
| Diagnosis at recruitment | |
| CIS-MS | 0/5 (0%) |
| MS | 5/5 (100%) |
| McDonald criteria fulfilled at recruitment: Yes | 5/5 (100%) |
| Disease duration at recruitment: Mean (SD) | 0.6 (0.5) |
| CSF OCBs: Present | 5/5 (100%) |
| EBV-VCA: Positive | 3/5 (60%) |
| MOG titer value: Median (min, max) | 20 (20, 160) |
| MOG titer level: ≥160 | 1/5 (20%) |
| Steroid use in 1 month before MOG test: Yes | 1/5 (20%) |
| DMT use in 3 months before MOG test: Yes | 4/5 (80%) |
| Ever new T2 brain lesions at 12-month follow-up: Yes | 1/3 (33%) |
| Ever new enhancing brain lesions at 12-month follow-up: Yes | 1/3 (33%) |
| Ever new T2 brain lesions at 24-month follow-up: Yes | 1/2 (50%) |
| Ever new enhancing brain lesions at 24-month follow-up: Yes | 1/2 (50%) |
| Disease duration at last follow-up, years: Mean (SD) | 3.0 (2.2) |
| ARR at last follow-up: Estimate (CI) | 0.3 (0.1, 0.9) |
| EDSS at last follow-up: Median (min, max) | 1.5 (1.0, 3.0) |
Abbreviations: MOG, myelin oligodendrocyte glycoprotein; MOGAD, myelin oligodendrocyte glycoprotein antibody-associated disease; MS, multiple sclerosis; ADEM, acute demyelinating encephalomyelitis; CIS, clinically isolated syndrome; CIS-MS, clinically isolated syndrome with high risk for MS; DDNOS, demyelinating disease not otherwise specified; OCB, oligoclonal band; EBV-VCA, Epstein-Barr virus-viral capsid antigen; ARR, annualized relapse rate; EDSS, expanded disability status score; SD, standard deviation; CI, confidence interval.
Table 6:
Characteristics of MOG positive cases
| Patient number | First demyelinating disease diagnosis | Diagnosis at time of recruitment | Most recent demyelinating disease diagnosis | Disease duration at time of MOG titer, years | MOG titer value | CSF OCBs present | EDSS at last follow-up |
|---|---|---|---|---|---|---|---|
| 1 | CIS | CIS | MOGAD | 1.20 | 20 | No | 1.0 |
| 2 | CIS | CIS | MOGAD | 0.35 | 20 | No | 2.0 |
| 3 | MS | MS | MS | 1.38 | 20 | Yes | 3.0 |
| 4 | MS | MS | MS | 0.33 | 20 | Yes | 2.5 |
| 5 | CIS | MS | MS | 0.43 | 20 | Yes | 1.5 |
| 6 | CIS | MS | MOGAD | 0.39 | 20 | No | 1.5 |
| 7 | MS | MS | MS | 0.66 | 20 | Yes | 1.0 |
| 8 | CIS | MS | MOGAD | 1.34 | 20 | Yes | 1.5 |
| 9 | ADEM | MS | MOGAD | 2.59 | 40 | No | 2.0 |
| 10 | ADEM | MS | MOGAD | 3.84 | 40 | No | 3.5 |
| 11 | CIS | MS | MOGAD | 1.69 | 40 | Yes | 3.0 |
| 12 | ADEM | MS | MOGAD | 3.68 | 40 | No | 1.0 |
| 13 | CIS | MS | MOGAD | 1.38 | 40 | No | 2.0 |
| 14 | ADEM | MS | MOGAD | 2.82 | 40 | No | 3.5 |
| 15 | MS | MS | MOGAD | 0.03 | 80 | No | 1.0 |
| 16 | ADEM | MS | MOGAD | 1.82 | 80 | No | 1.0 |
| 17 | MS | MS | MOGAD | 0.00 | 80 | No | 2.5 |
| 18 | MS | MS | MOGAD | 0.23 | 80 | No | 1.5 |
| 19 | MS | MS | MOGAD | 3.46 | 160 | Yes | 2.0 |
| 20 | CIS | MS | MOGAD | 1.04 | 160 | No | 3.0 |
| 21 | MS | MS | MOGAD | 1.37 | 160 | No | 1.0 |
| 22 | DDNOS | MS | MOGAD | 2.29 | 160 | No | 1.0 |
| 23 | ADEM | MS | MOGAD | 1.88 | 160 | No | 0.0 |
| 24 | MS | MS | MS | 0.05 | 160 | Yes | 1.0 |
| 25 | CIS | CIS | MOGAD | 0.47 | 160 | Yes | 0.0 |
| 26 | MS | MS | MOGAD | 0.65 | 320 | No | 1.5 |
| 27 | MS | MS | MOGAD | 0.25 | 320 | Yes | 1.0 |
| 28 | CIS | CIS | MOGAD | 0.98 | 320 | No | 2.0 |
| 29 | ADEM | MS | MOGAD | 0.33 | 640 | No | ∙∙ |
| 30 | ADEM | MS | MOGAD | 4.16 | 1280 | No | 1.5 |
Abbreviations: MOG, myelin oligodendrocyte glycoprotein; MOGAD, myelin oligodendrocyte glycoprotein antibody-associated disease; CIS, clinically isolated syndrome; MS, multiple sclerosis; ADEM, acute disseminated encephalomyelitis; DDNOS, demyelinating disease not otherwise specified; EBV-VCA, Epstein-Barr virus-viral capsid antigen; OCB, oligoclonal band; EDSS, expanded disability status score.
Discussion
In this cross-sectional seroprevalence study with prospective case follow-up we demonstrate the absence of AQP4-IgG and MOG-IgG positivity among a large number of healthy pediatric controls. A small proportion of study participants with a prior working diagnosis of POMS after case ascertainment had a final diagnosis of MOGAD (5%), but none had a final diagnosis of AQP4-NMOSD. Clinical features associated with a final MOGAD diagnosis include younger age at disease onset, ADEM phenotype at first demyelination, and lack of EBV exposure. False-positives among cases with a final diagnosis of MS were rare: 1% for MOG-IgG and 0% for AQP4-IgG, confirming a high specificity for MOG-IgG and even higher specificity of AQP4-IgG.
The absence of MOG-IgG and AQP4-IgG among a large group of pediatric healthy controls is novel and has not, to our knowledge, been previously demonstrated. This highlights the utility of these antibody biomarkers in children with CNS demyelinating disease. The absence of AQP4-IgG detection among our controls and cases is consistent with exceedingly rare false-positives using AQP4-IgG cell-based assays.4 In contrast to AQP4-IgG, MOG-IgG false-positives with cell-based assays account for up to a quarter of positives identified in clinical practice; these false-positives often occur at low end-titer and may reflect indiscriminate ordering of this highly specific test in very low probability situations.11 This is further emphasized by a study showing that approximately 1% of neurologic disease controls without MOGAD admitted to hospital were MOG-IgG positive at low titer, similar to our results.35 While false-positives occur at low titer, raising the screening cutoff would likely exclude many true MOGAD cases.11 The risk of false-positives in pediatric patients appears to be lower than in adults possibly reflecting a higher pre-test probability in children in which MOGAD accounts for a greater proportion of CNS demyelinating disease.11 Our data suggests that false-positive MOG-IgG results, though very rare, are more likely to occur in POMS than pediatric healthy controls and further studies in large numbers of adults are needed to determine if that is true into adulthood.
Among our cohort of children with a working diagnosis of POMS, none were positive for AQP4-IgG using the latest generation live cell-based assay. This suggests that AQP4-NMOSD may be easier to discriminate from MS. Conversely, this finding may relate to the lower proportion of AQP4-NMOSD among pediatric CNS demyelinating diseases or reflect the availability of this biomarker during the timeline of the study (in contrast to MOG-IgG which was not generally available in clinical practice), which may have led to cases testing positive not being enrolled. While 45% of all cases tested negative prior to enrollment, it is likely they were tested with older generation, less sensitive assays.
MOG-IgG was positive among 6% of our cohort of children with an initial working diagnosis of POMS and short disease duration at enrollment (mean 1.3 years, SD 1.3). Among these MOG-IgG positive children, 17% (or 1% of all cases) were determined to have false-positive MOG-IgG, based on clinical, MRI, and/or CSF features more consistent with MS after review by PIs and a panel of POMS experts. These findings reinforce prior reports, which have identified MOG antibodies in 7–17% of children with POMS.14,15 Data from a Canadian cohort of children similarly showed that while MOG antibodies were detected in the serum of 17% of those fulfilling criteria for POMS, their clinical and MRI courses were atypical for MS and at most recent evaluation none had a disease course consistent with MS.14
Our study findings indicate that the following features should alert clinicians of a possible diagnosis of MOGAD in patients with a working diagnosis of POMS: first demyelinating disease diagnosis of ADEM, younger age (≤12 years) at disease onset, and lack of EBV exposure. These features were significant in univariate analysis and remained significant in multivariate analysis. These results are similar to those of prior studies,36 including a recent study that found that EBV seronegativity in children with an MS-like presentation may suggest an alternate diagnosis, particularly MOGAD.37
Additional features that may suggest a MOGAD diagnosis over MS include: lack of new MRI T2 brain lesions around 12-month timepoint and resolution of most or all T2-lesions with follow-up (Figure 1C), as shown in recent studies,38,39 as well as female sex (significant in multivariable model only), initial optic nerve localization (significant in univariable model only), and lack of CSF OCBs and initial spinal cord localization (both significant in univariable model only). While MOGAD cases were less likely to have CSF OCBs, OCBs were present in one-fifth of these patients; therefore, the presence of CSF OCBs should not exclude a MOGAD diagnosis.40
The limitations of this study include the time between first demyelinating event and MOG-IgG testing. MOG-IgG antibody clinical testing was not available when many of the subjects initially presented. The disease duration at time of recruitment was up to four years in some subjects, though mean disease duration at time of recruitment was 1.0 year for MS and 1.5 years for MOGAD. In addition, we did not have serial MOG antibody testing. Our study therefore may not account for some transient (< 6 months) MOG-IgG seropositive cases, previously demonstrated in up to 57% of children with acquired demyelination.14 Also, end-titers may have been higher if tested at the time of their initial attack, as end-titer tend to reduce over time or when tested outside of attacks.14 End-titers may also, in theory, have been impacted by steroid use in the month before sampling (31% of MS or MS-CIS cases and 44% of MOGAD cases). Additionally, other important phenotypes of MOGAD such as unilateral cortical encephalitis were likely not accounted for in this study, which focuses on patients that had manifestations that overlapped with MS.20 Finally, it should be noted that the features associated with MOGAD identified in our cohort may not be generalizable to all pediatric CNS demyelinating diseases as our population was pre-selected for POMS. This may, for example, explain the female predominance seen in our MOGAD group, a finding not previously reported.
Strengths of this study include the large cohort of children with MS with prospective long-term follow-up from across the United States and controls who were enrolled at the same institutions as the cases. All cases were reviewed by POMS specialists at recruitment and at serial follow-ups. Cases with MOG-IgG positivity were re-reviewed by the PIs and those determined to have MS or unclear final diagnosis were again reviewed by a panel of POMS experts, using proposed consensus criteria/typical clinical and imaging findings for MOGAD16,17,28,33 to reach consensus. A live cell-based MOG-IgG assay was utilized which has been reported to be superior to other types of cell-based assay,8 and the reproducibility of this assay has previously been demonstrated when compared to assays from other international centers.7,8 Screening of both AQP4-IgG and MOG-IgG was performed blinded to knowledge of clinical data in a single neuroimmunology laboratory (Mayo Clinic). Additionally, EBV testing was completed for all cases. Finally, our study represents the largest POMS and control population published to date who had AQP4-IgG and MOG-IgG screening and longitudinal clinical follow-up.
In summary, our study demonstrates that MOG-IgG and AQP4-IgG are not found among healthy pediatric controls. MOG-IgG false-positivity at low-end titer (≤1:160) may, however, rarely occur in POMS at a frequency of approximately 1%. Additionally, our study shows that MOGAD accounts for a small proportion (5%) of patients initially assigned a diagnosis of POMS. Conversely, AQP4-NMOSD is very unlikely to mimic POMS. Clinical features associated with a final MOGAD diagnosis in suspected POMS cases include initial ADEM phenotype, younger age (≤12 years) at disease onset, and lack of EBV exposure.
Summary for Social Media if Published:
Author Twitter handles: @EoinFlanagan14
What is the current knowledge on this topic: Approximately one third of pediatric patients with acquired CNS demyelinating syndromes have been shown to be MOG antibody positive. However, information of MOG-IgG positivity in large cohorts with a diagnosis of pediatric-onset MS is limited.
What question does this study address: This study investigated the frequency of MOG and aquaporin-4 antibodies among healthy pediatric controls and pediatric onset multiple sclerosis patients, the clinical characteristics and outcomes in seronegative vs seropositive children, and clinical features associated with final disease diagnosis.
What does this study add to our knowledge: Our study showed that (1) MOG and aquaporin-4 antibodies were not found among healthy pediatric controls, (2) 5% of patients initially labelled as pediatric MS had MOG antibody associated disease (MOGAD), (3) 1% had a positive MOG antibody but their final diagnosis by consensus remained MS and thus they represented false positive results.
How might this potentially impact the practice of neurology: Our findings, taken together with prior studies, suggest that both MOG and aquaporin-4 antibodies are generally not identified in pediatric healthy individuals, but occasionally false-positive MOG antibody results occur in children with MS at a frequency of approximately 1%.
Acknowledgements
We thank Jessica Sagen for her assistance with study coordination. We thank the patients and families who participated in the study. This study was funded by the National Multiple Sclerosis Society and National Institute of Neurological Disorders and Stroke (R01NS113828, 1R01NS071463, S1-1808-32326).
Footnotes
Potential Conflicts of Interest
E.P.F. and S.J.P. are working in the Mayo Clinic Neuroimmunology laboratory clinical service that is commercially offering AQP4-IgG and MOG-IgG but receive no personal income from this testing. S.J.P. is a named inventor on patents (#12/678,350 filed 2010 and #12/573,942 filed 2008), which relate to functional AQP4/NMO-IgG assays and NMO-IgG as a cancer marker.
References
- 1.Fadda G, Armangue T, Hacohen Y, Chitnis T, Banwell B. Paediatric multiple sclerosis and antibody-associated demyelination: clinical, imaging, and biological considerations for diagnosis and care. Lancet Neurol 2021; 20(2): 136–49. [DOI] [PubMed] [Google Scholar]
- 2.Waters PJ, McKeon A, Leite MI, et al. Serologic diagnosis of NMO: a multicenter comparison of aquaporin-4-IgG assays. Neurology 2012; 78(9): 665–71; discussion 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.McKeon A, Lennon VA, Lotze T, et al. CNS aquaporin-4 autoimmunity in children. Neurology 2008; 71(2): 93–100. [DOI] [PubMed] [Google Scholar]
- 4.Fryer JP, Lennon VA, Pittock SJ, et al. AQP4 autoantibody assay performance in clinical laboratory service. Neurol Neuroimmunol Neuroinflamm 2014; 1(1): e11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jurynczyk M, Geraldes R, Probert F, et al. Distinct brain imaging characteristics of autoantibody-mediated CNS conditions and multiple sclerosis. Brain 2017; 140(3): 617–27. [DOI] [PubMed] [Google Scholar]
- 6.Hacohen Y, Wong YY, Lechner C, et al. Disease Course and Treatment Responses in Children With Relapsing Myelin Oligodendrocyte Glycoprotein Antibody-Associated Disease. JAMA Neurol 2018; 75(4): 478–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Waters PJ, Komorowski L, Woodhall M, et al. A multicenter comparison of MOG-IgG cell-based assays. Neurology 2019; 92(11): e1250–e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Reindl M, Schanda K, Woodhall M, et al. International multicenter examination of MOG antibody assays. Neurol Neuroimmunol Neuroinflamm 2020; 7(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hennes EM, Baumann M, Schanda K, et al. Prognostic relevance of MOG antibodies in children with an acquired demyelinating syndrome. Neurology 2017; 89(9): 900–8. [DOI] [PubMed] [Google Scholar]
- 10.Duignan S, Wright S, Rossor T, et al. Myelin oligodendrocyte glycoprotein and aquaporin-4 antibodies are highly specific in children with acquired demyelinating syndromes. Dev Med Child Neurol 2018; 60(9): 958–62. [DOI] [PubMed] [Google Scholar]
- 11.Sechi E, Buciuc M, Pittock SJ, et al. Positive Predictive Value of Myelin Oligodendrocyte Glycoprotein Autoantibody Testing. JAMA Neurol 2021; 78(6): 741–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cobo-Calvo A, d’Indy H, Ruiz A, et al. Frequency of myelin oligodendrocyte glycoprotein antibody in multiple sclerosis: A multicenter cross-sectional study. Neurol Neuroimmunol Neuroinflamm 2020; 7(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Reindl M, Waters P. Myelin oligodendrocyte glycoprotein antibodies in neurological disease. Nat Rev Neurol 2019; 15(2): 89–102. [DOI] [PubMed] [Google Scholar]
- 14.Waters P, Fadda G, Woodhall M, et al. Serial Anti-Myelin Oligodendrocyte Glycoprotein Antibody Analyses and Outcomes in Children With Demyelinating Syndromes. JAMA Neurol 2020; 77(1): 82–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Armangue T, Olivé-Cirera G, Martínez-Hernandez E, et al. Associations of paediatric demyelinating and encephalitic syndromes with myelin oligodendrocyte glycoprotein antibodies: a multicentre observational study. Lancet Neurol 2020; 19(3): 234–46. [DOI] [PubMed] [Google Scholar]
- 16.Bruijstens AL, Lechner C, Flet-Berliac L, et al. E.U. paediatric MOG consortium consensus: Part 1 - Classification of clinical phenotypes of paediatric myelin oligodendrocyte glycoprotein antibody-associated disorders. Eur J Paediatr Neurol 2020; 29: 2–13. [DOI] [PubMed] [Google Scholar]
- 17.Baumann M, Bartels F, Finke C, et al. E.U. paediatric MOG consortium consensus: Part 2 - Neuroimaging features of paediatric myelin oligodendrocyte glycoprotein antibody-associated disorders. Eur J Paediatr Neurol 2020; 29: 14–21. [DOI] [PubMed] [Google Scholar]
- 18.Yazbeck E, Maurey H, Leroy C, et al. Progressive Leukodystrophy-Like Demyelinating Syndromes with MOG-Antibodies in Children: A Rare Under-Recognized Phenotype. Neuropediatrics 2021. [DOI] [PubMed] [Google Scholar]
- 19.Banks SA, Morris PP, Chen JJ, et al. Brainstem and cerebellar involvement in MOG-IgG-associated disorder versus aquaporin-4-IgG and MS. J Neurol Neurosurg Psychiatry 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ogawa R, Nakashima I, Takahashi T, et al. MOG antibody-positive, benign, unilateral, cerebral cortical encephalitis with epilepsy. Neurol Neuroimmunol Neuroinflamm 2017; 4(2): e322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kim SH, Kim W, Li XF, Jung IJ, Kim HJ. Does interferon beta treatment exacerbate neuromyelitis optica spectrum disorder? Mult Scler 2012; 18(10): 1480–3. [DOI] [PubMed] [Google Scholar]
- 22.Jacob A, Hutchinson M, Elsone L, et al. Does natalizumab therapy worsen neuromyelitis optica? Neurology 2012; 79(10): 1065–6. [DOI] [PubMed] [Google Scholar]
- 23.Min JH, Kim BJ, Lee KH. Development of extensive brain lesions following fingolimod (FTY720) treatment in a patient with neuromyelitis optica spectrum disorder. Mult Scler 2012; 18(1): 113–5. [DOI] [PubMed] [Google Scholar]
- 24.Jitprapaikulsan J, Chen JJ, Flanagan EP, et al. Aquaporin-4 and Myelin Oligodendrocyte Glycoprotein Autoantibody Status Predict Outcome of Recurrent Optic Neuritis. Ophthalmology 2018; 125(10): 1628–37. [DOI] [PubMed] [Google Scholar]
- 25.Lopez-Chiriboga AS, Sechi E, Buciuc M, et al. Long-term Outcomes in Patients With Myelin Oligodendrocyte Glycoprotein Immunoglobulin G-Associated Disorder. JAMA Neurol 2020; 77(12): 1575–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Deschamps R, Pique J, Ayrignac X, et al. The long-term outcome of MOGAD: An observational national cohort study of 61 patients. Eur J Neurol 2021; 28(5): 1659–64. [DOI] [PubMed] [Google Scholar]
- 27.Wingerchuk DM, Banwell B, Bennett JL, et al. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology 2015; 85(2): 177–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Jarius S, Paul F, Aktas O, et al. MOG encephalomyelitis: international recommendations on diagnosis and antibody testing. J Neuroinflammation 2018; 15(1): 134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mar S, Liang S, Waltz M, et al. Several household chemical exposures are associated with pediatric-onset multiple sclerosis. Ann Clin Transl Neurol 2018; 5(12): 1513–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011; 69(2): 292–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Krupp LB, Tardieu M, Amato MP, et al. International Pediatric Multiple Sclerosis Study Group criteria for pediatric multiple sclerosis and immune-mediated central nervous system demyelinating disorders: revisions to the 2007 definitions. Mult Scler 2013; 19(10): 1261–7. [DOI] [PubMed] [Google Scholar]
- 32.Nourbakhsh B, Rutatangwa A, Waltz M, et al. Heterogeneity in association of remote herpesvirus infections and pediatric MS. Ann Clin Transl Neurol 2018; 5(10): 1222–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Marignier R, Hacohen Y, Cobo-Calvo A, et al. Myelin-oligodendrocyte glycoprotein antibody-associated disease. Lancet Neurol 2021; 20(9): 762–72. [DOI] [PubMed] [Google Scholar]
- 34.Lui A, Chong J, Flanagan E, et al. High titers of myelin oligodendrocyte glycoprotein antibody are only observed close to clinical events in pediatrics. Mult Scler Relat Disord 2021; 56: 103253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Held F, Kalluri SR, Berthele A, Klein A-K, Reindl M, Hemmer B. Frequency of myelin oligodendrocyte glycoprotein antibodies in a large cohort of neurological patients. Multiple Sclerosis Journal - Experimental, Translational and Clinical 2021; 7(2): 20552173211022767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Hacohen Y, Mankad K, Chong WK, et al. Diagnostic algorithm for relapsing acquired demyelinating syndromes in children. Neurology 2017; 89(3): 269–78. [DOI] [PubMed] [Google Scholar]
- 37.Nourbakhsh B, Cordano C, Asteggiano C, et al. Multiple Sclerosis Is Rare in Epstein-Barr Virus-Seronegative Children with Central Nervous System Inflammatory Demyelination. Ann Neurol 2021; 89(6): 1234–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Fadda G, Banwell B, Waters P, et al. Silent New Brain MRI Lesions in Children with MOG-Antibody Associated Disease. Ann Neurol 2021; 89(2): 408–13. [DOI] [PubMed] [Google Scholar]
- 39.Sechi E, Krecke KN, Messina SA, et al. Comparison of MRI Lesion Evolution in Different Central Nervous System Demyelinating Disorders. Neurology 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Armangue T, Capobianco M, de Chalus A, Laetitia G, Deiva K, consortium EUpM. E.U. paediatric MOG consortium consensus: Part 3 - Biomarkers of paediatric myelin oligodendrocyte glycoprotein antibody-associated disorders. Eur J Paediatr Neurol 2020; 29: 22–31. [DOI] [PubMed] [Google Scholar]