AQP4-IgG and MOG-IgG Related Optic Neuritis—Prevalence, Optical Coherence Tomography Findings, and Visual Outcomes: A Systematic Review and Meta-Analysis

Angeliki Filippatou, Loulwah Mukharesh, Shiv Saidha, Peter A. Calabresi, Elias S. SotirchosFrontiers in Neurology 2020 · DOI

TL;DR

This systematic review and meta-analysis clarifies the prevalence of AQP4-IgG and MOG-IgG in optic neuritis (ON), revealing higher rates in Asian populations and a significant presence in pediatric ON. Crucially, while both conditions cause similar retinal thinning, AQP4-IgG-related ON leads to significantly worse visual outcomes than MOG-IgG or MS-related ON.

What Problem Does This Paper Address?

Optic neuritis (ON) is a key symptom in several neurological conditions, including multiple sclerosis (MS), and diseases associated with antibodies to aquaporin-4 (AQP4-IgG) and myelin oligodendrocyte glycoprotein (MOG-IgG). However, for AQP4-IgG and MOG-IgG associated ON, particularly when it occurs in isolation (without other neurological signs), understanding its prevalence has been challenging due to inconsistent reporting across studies. Furthermore, direct comparisons of how these conditions affect vision and the physical structure of the retina (using optical coherence tomography, OCT) have been limited by small sample sizes, making it difficult to establish clear differences in disease severity and prognosis. This paper aims to consolidate existing research to provide a clearer picture of these prevalence rates and the structural and functional consequences of AQP4-IgG and MOG-IgG related ON compared to MS-related ON.

Methods

The authors conducted a systematic review and meta-analysis, a robust method for synthesizing evidence from multiple studies. They searched the PubMed database for relevant research using a comprehensive set of keywords related to optic neuritis, AQP4-IgG, MOG-IgG, and OCT. They identified 65 eligible studies for their analysis. Statistical analyses were performed using random effects models to account for potential variations across studies. The review focused on three main objectives:

  1. Prevalence: Determining the frequency of AQP4-IgG and MOG-IgG seropositivity in patients with isolated ON, differentiating between adult and pediatric populations, and by ethnicity (Asian vs. non-Asian).
  2. OCT Findings: Comparing structural retinal measures (specifically peri-papillary retinal nerve fiber layer (pRNFL) and macular ganglion cell + inner plexiform layer (GCIPL) thicknesses) between AQP4-ON, MOG-ON, and MS-ON eyes.
  3. Visual Outcomes: Comparing visual acuity (measured by logMAR) between AQP4-ON, MOG-ON, and MS-ON eyes.

Datasets

Not applicable. This paper is a systematic review and meta-analysis of existing published studies. No new datasets were collected or generated.

Key Insights and Ideas

  • Ethnicity and Age Matter: The prevalence of AQP4-IgG and MOG-IgG antibodies in ON is not uniform. Asian populations show significantly higher rates of AQP4-IgG positivity, while MOG-IgG is particularly prevalent in pediatric ON, even in non-Asian children. This highlights the importance of considering demographic factors in diagnosing ON.
  • Structural Damage is Similar, Functional Outcome is Not: A striking finding is that despite AQP4-ON and MOG-ON eyes showing comparable thinning of retinal layers (pRNFL and GCIPL) to each other and to MS-ON eyes, AQP4-ON eyes experience significantly worse visual outcomes. This suggests that the underlying pathology or the retina's response to injury differs between these conditions, leading to a disconnect between structural damage and functional vision.
  • MOG-ON: A Different Beast? The data suggests MOG-ON might have a different pathomechanism than AQP4-ON, potentially involving secondary damage rather than direct retinal pathology, given the preserved visual acuity despite structural thinning.

Key Results

  • Prevalence:
    • Adults with isolated ON: AQP4-IgG seroprevalence was 4% in non-Asian and 27% in Asian populations. MOG-IgG seroprevalence was 8% in non-Asian and 20% in Asian populations.
  • Children with isolated ON: AQP4-IgG seroprevalence was 0.4% in non-Asian and 15% in Asian populations. MOG-IgG seroprevalence was 47% in non-Asian and 31% in Asian populations.
  • OCT Findings (compared to MS-ON eyes):
    • AQP4-ON eyes showed significantly lower pRNFL (-11.7 µm) and GCIPL (-9.0 µm) thicknesses.
  • MOG-ON eyes also showed significantly lower pRNFL (-11.2 µm) and GCIPL (-6.1 µm) thicknesses.
  • However, AQP4-ON and MOG-ON eyes did not differ significantly in pRNFL or GCIPL thickness.
  • Visual Outcomes (logMAR difference):
    • AQP4-ON eyes were significantly worse than MOG-ON (0.60 difference) and MS-ON (0.68 difference).
  • MOG-ON and MS-ON eyes had similar visual outcomes (0.04 difference).
  • The risk of poor visual outcome (VA < 20/200) was higher in AQP4-ON compared to both MOG-ON and MS-ON.

Limitations and Future Work

The authors acknowledge several limitations:

  • Referral Bias: Many studies were conducted at tertiary referral centers, potentially overrepresenting more severe or atypical cases and thus overestimating prevalence.
  • Heterogeneity: Significant heterogeneity was observed across studies in OCT devices, scanning protocols, and patient demographics, which the meta-analysis attempted to mitigate but could still influence results.
  • Assay Variability: While cell-based assays (CBAs) were preferred, some studies used ELISA for AQP4-IgG, which has lower specificity.
  • Limited Pediatric Data: The analysis for pediatric ON was underpowered due to a small number of studies.
  • Recurrence Data: The impact of recurrent ON attacks on OCT and visual outcomes could not be fully accounted for due to inconsistent reporting.

Future work should focus on:

  • Investigating the pathoetiology behind the structural-functional disconnect in MOG-ON.
  • Clarifying the role of microcystoid macular pathology (MMP) in ON outcomes.
  • Conducting studies with larger, more diverse cohorts, particularly including individuals of African ancestry, given the known higher prevalence of NMOSD in these groups.
  • Developing standardized protocols for OCT and visual outcome assessment to reduce heterogeneity.
  • Exploring optimal acute and chronic treatment strategies for these distinct ON entities.

Summary updated

Abstract

<b>Background:</b> Optic neuritis (ON) is a cardinal manifestation of multiple sclerosis (MS), aquaporin-4 (AQP4)-IgG-, and myelin oligodendrocyte glycoprotein (MOG)-IgG-associated disease. However, the prevalence of AQP4-IgG seropositivity and MOG-IgG seropositivity in isolated ON is unclear, and studies comparing visual outcomes and optical coherence tomography (OCT)-derived structural retinal measures between MS-ON, AQP4-ON, and MOG-ON eyes are limited by small sample sizes. <b>Objectives:</b> (1) To assess the prevalence of AQP4-IgG and MOG-IgG seropositivity among patients presenting with isolated ON; (2) to compare visual outcomes and OCT measures between AQP4-ON, MOG-ON, and MS-ON eyes. <b>Methods:</b> In this systematic review and meta-analysis, a total of 65 eligible studies were identified by PubMed search. Statistical analyses were performed with random effects models. <b>Results:</b> In adults with isolated ON, AQP4-IgG seroprevalence was 4% in non-Asian and 27% in Asian populations, whereas MOG-IgG seroprevalence was 8 and 20%, respectively. In children, AQP4-IgG seroprevalence was 0.4% in non-Asian and 15% in Asian populations, whereas MOG-IgG seroprevalence was 47 and 31%, respectively. AQP4-ON eyes had lower peri-papillary retinal nerve fiber layer (pRNFL; -11.7 μm, 95% CI: -15.2 to -8.3 μm) and macular ganglion cell + inner plexiform layer (GCIPL; -9.0 μm, 95% CI: -12.5 to -5.4 μm) thicknesses compared with MS-ON eyes. Similarly, pRNFL (-11.2 μm, 95% CI: -21.5 to -0.9 μm) and GCIPL (-6.1 μm, 95% CI: -10.8 to -1.3 μm) thicknesses were lower in MOG-ON compared to MS-ON eyes, but did not differ between AQP4-ON and MOG-ON eyes (pRNFL: -1.9 μm, 95% CI: -9.1 to 5.4 μm; GCIPL: -2.6 μm, 95% CI: -8.9 to 3.8 μm). Visual outcomes were worse in AQP4-ON compared to both MOG-ON (mean logMAR difference: 0.60, 95% CI: 0.39 to 0.81) and MS-ON eyes (mean logMAR difference: 0.68, 95% CI: 0.40 to 0.96) but were similar in MOG-ON and MS-ON eyes (mean logMAR difference: 0.04, 95% CI: -0.05 to 0.14). <b>Conclusions:</b> AQP4-IgG- and MOG-IgG-associated disease are important diagnostic considerations in adults presenting with isolated ON, especially in Asian populations. Furthermore, MOG-IgG seroprevalence is especially high in pediatric isolated ON, in both non-Asian and Asian populations. Despite a similar severity of GCIPL and pRNFL thinning in AQP4-ON and MOG-ON, AQP4-ON is associated with markedly worse visual outcomes.
BibTeX
@inproceedings{filippatou2020aqp,
  author = {Angeliki Filippatou and Loulwah Mukharesh and Shiv Saidha and Peter A. Calabresi and Elias S. Sotirchos},
  title = {{AQP4-IgG and MOG-IgG Related Optic Neuritis—Prevalence, Optical Coherence Tomography Findings, and Visual Outcomes: A Systematic Review and Meta-Analysis}},
  booktitle = {Frontiers in Neurology},
  year = {2020},
  doi = {10.3389/fneur.2020.540156},
}
Figure