Low mortality rate in a large cohort of myelin oligodendrocyte glycoprotein antibody disease (MOGAD)

Abstract

The mortality rates of individuals with myelin oligodendrocyte glycoprotein antibody disease (MOGAD) are currently unknown. This study aimed to assess the mortality rate in a large cohort of patients with MOGAD. Since none of the patients in our cohort died, we estimated the upper limit of a 95% confidence interval of the crude mortality rate in the cohort to be 2.1%. These data suggest that mortality in MOGAD is lower than that reported in other neuroinflammatory diseases and comparable to the age‐adjusted mortality rates of the general population in the United States. Additional studies are warranted to confirm this observation.

Introduction

Myelin oligodendrocyte glycoprotein antibody disease (MOGAD) is an autoimmune disorder of the central nervous system (CNS) which can present with a variety of clinical manifestations including acute disseminated encephalomyelitis (ADEM), optic neuritis (ON), transverse myelitis (TM), and cortical and brainstem syndromes. ¹

While the long‐term outcomes in MOGAD are generally favorable, permanent disability may occur due to visual, sphincter, and/or erectile impairment. ² , ³ Mortality rates in MOGAD have not been previously reported. In this analysis, we calculated the mortality rate in a large MOGAD cohort and compared it to data reported in the literature for other demyelinating diseases.

Patients and Methods

This was a retrospective cohort study evaluating the living status of all MOGAD patients seen at the Neuroimmunology Clinic at the Massachusetts General Hospital (MGH). Patients were identified through the Mass General Brigham (MGB) electronic database on October 1, 2022. To be included in the study, patients had to have a positive MOG‐IgG antibody test verified by a cell‐based assay and a clinical syndrome compatible with MOGAD. ⁴ The clinical charts were evaluated for evidence of health care services (i.e., office visit, lab or imaging study, patient message, or documented phone call) provided in the 6 months preceding the index date. If such evidence was unavailable, the patients were contacted directly (by phone, email, or patient message) or through their primary care clinic to verify their living status. Additional data extracted from the patient's charts included age, sex, date of diagnosis, age at disease onset, clinical disease manifestations (i.e., optic neuritis, transverse myelitis, acute demyelinating encephalomyelitis (ADEM), brainstem syndrome, cerebral syndrome, and meningitis), duration of follow‐up, number of relapses, expanded disability status scale (EDSS) score at the last follow‐up, use of immunotherapy at the last follow‐up, and associated comorbidities (i.e., hypertension, hyperlipidemia, diabetes mellitus, ischemic heart disease, cardiac arrhythmias, lung disease, obesity, obstructive sleep apnea syndrome, and cancer). In cases whose living status was verified by direct contact, the duration of follow‐up was calculated as the time from disease onset to that contact, while the other variables were extracted from the last available in‐person office visit.

Total counts and proportions, medians, ranges, and interquartile ranges (IQR) are used to describe the main characteristics of the study cohort. Because none of the patients in our cohort died, we applied the “rule of three’ for the interpretation of data in which none of the events of concern has occurred with a 5% limit of credibility. We therefore applied the following formula to estimate the upper limit of a 95% confidence interval (CI) mortality rate in MOGAD: 3/n + 1, where n indicates the number of MOGAD patients in the cohort or person‐years of follow‐up. ⁵

The study was approved by the MGH institutional review board. Due to the retrospective study design and use of aggregate anonymous data, patients' informed consent was not obtained.

Results

One hundred fifty‐one patients with positive MOG‐IgG tests were identified. Of these, 9 were excluded due to a clinical presentation that was not consistent with the diagnosis of MOGAD. The remaining 142 patients were included in the study, of whom 117 had documentation of health care services provided in the past 6 months. In the other 25 patients, the living status was verified by direct contact with the patient, parent, or primary care physician. All the patients were verified to be alive. The patient selection and verification of the living status process of the study population are illustrated in Figure 1.

Figure 1.

Selection and verification process of the living status of the study cohort.

The median age in our cohort was 36.5 years (range 8–85, IQR 24–55), and the median disease duration was 4 years (range 1–41, IQR 3–10). Eighty‐seven (61.3%) were females. The median number of relapses was 2 (range 1–21, IQR 1–4), and the median EDSS was 1 (range 0–6, IQR 1–2). One hundred and four patients (73.3%) were treated with immunotherapies at the last follow‐up. Thirty patients (21.1%) had vascular comorbidities (i.e., hypertension, hyperlipidemia, diabetes mellitus, ischemic heart disease), and three (2.1%) had cancer. Table 1 summarizes the demographic and disease‐related clinical characteristics of the study cohort.

Table 1.

Demographic and disease‐related characteristics of the study cohort.

	MOGAD Study Cohort (n = 142)
Age (in years; median, (range, IQR))	36.5 (8–85, 24–55)
Female sex (n, %)	87 (61.3)
Disease duration (in years; median, (range, IQR))	4 (1–41, 3–10)
Clinical manifestations (n, %) a
Optic neuritis	108 (76.1)
Transverse myelitis	58 (40.1)
ADEM	20 (14.1)
Brainstem syndrome	17 (12)
Cerebral syndrome	15 (10.6)
Meningitis	5 (3.5)
EDSS (median, (range, IQR))	1 (0–6, 1–2)
Number of relapses (median, (range, IQR))	2 (1–21, 1–4)
Treated with immunotherapy at last follow up (n, %)	104 (73.3)
Associated comorbidities (n, %) b
Hypertension	19 (13.4)
Hyperlipidemia	11 (7.8)
Diabetes mellitus	10 (7)
Ischemic heart disease	3 (2.1)
Cardiac arrhythmias c	3 (2.1)
Lung disease d	6 (4.2)
Obesity	2 (1.4)
OSAS	3 (2.1)
Cancer e	3 (2.1)

ADEM, acute demyelinating encephalomyelitis; OSAS, obstructive sleep apnea syndrome.

^a Some patients had more than one clinical manifestation.

^b Some patients had more than one associated comorbidity.

^c Cardiac arrhythmias include atrial fibrillation (n = 2) and supraventricular tachycardia (n = 1).

^d Lung diseases include asthma (n = 5) and chronic obstructive pulmonary disease (COPD, n = 1).

^e One‐ prostatic carcinoma, one‐ ovarian carcinoma, one‐ papillary thyroid carcinoma.

Based on the estimated mortality equation, the upper limit of a 95% CI for the mortality rate in MOGAD is not greater than 2.1%. When person‐years of follow‐up are used as the denominator instead of the number of patients in the cohort, the estimated upper limit of our cohort's 95% CI mortality rate is much lower at 3.1 per 1000 person‐years.

Discussion

This study reports low estimated mortality rates in MOGAD, with an upper limit of a 95% CI for the crude mortality rate of 2.1%. This rate is lower compared to aquaporin 4 (AQP4)‐positive neuromyelitis optica spectrum disorder (NMOSD) ⁶ , ⁷ and multiple sclerosis (MS), ⁸ , ⁹ and is largely comparable to the age‐adjusted mortality rates reported for the general population in the United States. ¹⁰ Of note, our estimation is based on a conservative method where the total number of people in the cohort instead of person‐years of exposure is used as the denominator. If person‐years of exposure were used, the estimated upper limit of our cohort's 95% CI mortality rate would be as low as 3.1 per 1000 person‐years.

Several considerations may explain the low mortality rate in MOGAD. First, disease‐related mortality in neuroinflammatory diseases is usually ascribed to severe brainstem or upper cervical cord involvement leading to respiratory failure. ⁶ , ⁷ The fact that MOGAD mainly involves the optic nerves and spinal cord involvement in MOGAD often affects the thoracic region and the conus medullaris ² , ¹¹ may contribute to its lower mortality rate. However, even MOGAD‐related CNS attacks requiring ventilatory support were shown to have a better outcome compared to AQP4‐positive NMOSD. ¹² Therefore, the better outcome cannot be attributed solely to disease localization. Infections are another major cause of death in neuroinflammatory diseases. ⁶ The risk of infections in these patient populations is particularly high in patients with severe motor disabilities, especially if treated with immunotherapy. ¹³ MOGAD patients are not always treated with immunotherapies and are less likely to develop severe disabilities ² so they may therefore be less prone to severe and life‐threatening infections. Lastly, mortality rates increase with age and the presence of other comorbidities, like cardiovascular disease and cancer. ¹⁰ The relatively young age and low rate of comorbidities in our cohort may have also contributed to the low mortality rate.

While none of the patients in our cohort died, we are aware of several other reports of MOGAD patients who have passed away. However, these reports are very limited; when reviewing other historical MOGAD cohorts, ² , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ we were able to identify only two MOGAD cases with a fatal outcome – Jarius et al. reported one patient (out of 50) who died due to severe brainstem encephalitis, ¹⁵ and Cobo‐Calvo et al. reported another single patient (out of 197) who died, although the cause of death in the latter case was not reported. ¹⁴ Overall, the low mortality rate observed in our cohort seems in line with the low number of deaths reported in the literature.

Several limitations of this study should be taken into account. First, because none of the patients in our cohort died, we could only report the estimated upper limit of a 95% CI rather than a true mortality rate. Second, the median observational period in our cohort was relatively short, and longer periods of follow‐up are needed in future studies. Nevertheless, we believe the information reported in this study has valuable clinical relevance, as it is the first study to report an estimated low mortality rate in MOGAD.

In conclusion, based on the current data, the mortality rate in MOGAD is lower than in AQP4‐positive NMOSD and in MS and is comparable to the general population. Larger studies are warranted to confirm these observations.

Author Contributions

IL‐ Conception and design of the study, acquisition and analysis of data, drafting the manuscript. GR‐ Acquisition of data, drafting the manuscript. RS‐ Acquisition of data, drafting the manuscript. NM‐ Acquisition of data, drafting the manuscript. AV‐ Acquisition of data, drafting the manuscript. MA‐ Acquisition of data, drafting the manuscript. PAB‐ Acquisition of data, drafting the manuscript. GC‐‐ Conception and design of the study, analysis of data, drafting the manuscript. ML‐ Conception and design of the study, acquisition and analysis of data, drafting the manuscript.

Conflict of Interest

IL, GR, RS, NM, AV, MA, PAB, GC and ML‐ Nothing to report.