Coexistence of Myelin Oligodendrocyte Glycoprotein and Aquaporin-4 Antibodies in Adult and Pediatric Patients

Abstract

This study characterizes the frequency and sex ratio of positive test results for myelin oligodendrocyte glycoprotein–IgG and aquaporin-4–IgG in adult and pediatric patient populations in a single center.

Myelin oligodendrocyte glycoprotein (MOG)–IgG is a biomarker associated with central nervous system–demyelinating disorders termed MOG-IgG–associated disorders. These disorders have overlapping clinical features, including optic neuritis and myelitis, with aquaporin-4 (AQP4)–IgG–positive neuromyelitis optica spectrum disorders. These disorders are hypothesized to be biologically distinct; AQP4-IgG–positive neuromyelitis optica spectrum disorders are autoimmune astrocytopathies, whereas MOG-IgG–associated disorders are postulated to be autoimmune oligodendrocytopathies.¹ Despite their immunopathogenic differences, there are rare reports of patients with dual positivity of MOG-IgG and AQP4-IgG.^2,3 We aimed to determine the frequency, sex ratio, and coexistence of glial antibodies (AQP4-IgG and MOG-IgG) in adults and children undergoing evaluation for suspected central nervous system–demyelinating diseases.

Methods

This is a retrospective cohort study approved by the Mayo Clinic institutional review board and ethics committee with a waiver of informed consent because of the restricted, deidentified nature of the data used. All patients who had serum samples tested for AQP4-IgG and MOG-IgG from October 2017 to May 2019 were evaluated using a clinically validated flow cytometric assay (BD FACSCanto II [BD Biosciences]).⁴ All samples were repeated to confirm positivity. We have used basic demographic data (age and sex) provided through the clinical testing request forms. We used χ² tests to compare the proportions of female and male individuals within the antibody-positive groups. All tests were 2-sided, and P values less than .05 were considered significant. The statistical software R version 3.5.1 (R Foundation for Statistical Computing) was used.

Results

Over a 20-month period, 15 598 patients (1862 children [<18 years] and 13 736 adults) were tested for both AQP4-IgG and MOG-IgG (Figure 1). In 1291 patients (8.3%), MOG-IgG was detected; AQP4-IgG was detected in 387 patients (2.3%).

Figure 1. Flowchart of Cohort Selection and Identification of Glial Antibodies Within the Tested Population.

The frequency of detection of antibodies in 13 736 adults and 1862 children. AQP4-IgG indicates aquaporin-4 immunoglobulin G; MOG-IgG, myelin oligodendrocyte glycoprotein immunoglobulin G.

Of the adults, 899 (6.5%) were MOG-IgG positive, and 351 (2.6%) were AQP4-IgG positive (Figure 2A). Of the adults with MOG-IgG, 546 were female (60.7%); 303 of the adults with AQP4-IgG (86.3%) were female (P < .001) (Figure 2C). Of the children, 392 were MOG-IgG positive (21.1%), and 36 were AQP4-IgG positive (1.9%) (Figure 2B). Of all children with MOG-IgG, 214 were female (54.6%) vs 21 (58.3%) of all children with AQP4-IgG (P = .79) (Figure 2D).

Figure 2. Frequency of Myelin Oligodendrocyte Glycoprotein (MOG)–IgG and Aquaporin-4 (AQP4)–IgG Serum Titers on Flow Cytometry Assay.

A-D, sex ratio of glial antibodies in adults and children; E, AQP4-IgG–positive cohort; F, MOG-IgG–positive cohort; G, AQP4-IgG in individuals with dual positivity; and H, MOG-IgG in individuals with dual positivity.

In the subgroup of patients with serum testing completed on the same sample (n = 8276), similar proportions of antibody positivity were observed as in those with separate samples tested (Figure 1). The median serum titers for the cohort were 1:10 000 (range, 1:5 to 1:100 000) for AQP4-IgG and 1:100 (range, 1:20 to 1:100 000) for MOG-IgG (Figure 2E and F). The adult and children subgroups had the same median titers.

Of 15 598 patients, 10 patients (0.06%) had dual positivity. All 10 patients with dual positivity had high-titer AQP4-IgG (median, 1:10 000; range, 1:100 to 1:100 000) and low-titer MOG-IgG (median, 1:40; range, 1:20 to 1:100) (Figure 2G and H). All of the patients with dual positivity were adults (median age, 47 years [range, 30-61 years]), and 9 of the 10 patients were female (90%).

Conclusions

In this large cohort of patients undergoing evaluation for suspected central demyelinating diseases, MOG-IgG were detected almost 3 times as often as AQP4-IgG in adults; in children, MOG-IgG was detected more than 11 times as often as AQP4-IgG. In this study, MOG-IgG and AQP4-IgG rarely coexisted in a single patient (0.06%).

We have confirmed the previously described female predominance in adults who are positive for AQP4-IgG.⁵ Furthermore, we have demonstrated that there is a distinct difference in the female predominance in adults with AQP4-IgG compared with MOG-IgG. Together, these sex differences and the greater detection of MOG-IgG, especially in children, suggests that central demyelinating diseases associated with these biomarkers may have different drivers of autoimmunity.

The rare coexistence of these 2 antibody biomarkers also points to distinct immunopathogeneses of these diseases. It has been postulated that MOG-IgG is an epiphenomenon, occurring with exposure of antigen in the AQP4-IgG disease. However, the rarity of dual positivity is this large cohort is evidence against an epiphenomenon. All individuals with dual positivity had high titers of AQP4-IgG and low titers of MOG-IgG, suggesting the disease phenotype may be more compatible with AQP4-IgG–positive neuromyelitis optica spectrum disorders.

These findings suggest that MOG-IgG may occur more frequently than AQP4-IgG in patients tested for central demyelinating diseases, particularly in children. However, because the study population was derived from patients who had undergone a central demyelinating disease serological evaluation, these data should not be interpreted as representing seroprevalence in the general population. Further population-based studies are required to determine the prevalence and incidence of MOG-IgG–associated disorders. That MOG-IgG and AQP4-IgG rarely coexist highlights the immunopathogenic distinction of these biomarkers.