Myelin Oligodendrocyte Glycoprotein Antibodies in Breast Milk in a Lactating Woman With Severe Enterovirus-Associated MOGAD

Abstract

Objectives

Myelin oligodendrocyte glycoprotein antibody (MOG-IgG)–associated disease (MOGAD) is a distinct neuroinflammatory disorder, which may be preceded by infections. Transmission of MOG-IgG from blood into breast milk has not been reported previously.

Methods

We present the case of a 38-year-old lactating woman with a severe manifestation of MOGAD after enterovirus gastroenteritis, who was still lactating at that time. Alongside with detailed clinical characteristics, we measured MOG-IgG in breast milk and serum from the breast-fed infant.

Results

Our patient experienced a severe progressive episode of CNS inflammation, affecting the optic nerve, brain, and spinal cord, before which she was partially breastfeeding. Her condition rapidly worsened to coma with status epilepticus, requiring intensive care admission. High serum MOG-IgG titers were detected and prompt attack therapy initiated with high-dose glucocorticosteroids followed by plasma exchange, which led to rapid clinical improvement. The patient's breast milk tested positive for MOG-IgG, whereas the serum of the breastfed infant was negative.

Discussion

Early diagnosis and therapeutic intervention at acute onset of MOGAD is important for attack remission. MOG-IgG may be detected in breast milk of lactating woman with MOGAD, but is probably not transmitted to the infant in detectable quantities.

Introduction

Myelin oligodendrocyte glycoprotein antibody (MOG-IgG)–associated disease (MOGAD) is a distinct demyelinating disorder of the CNS often presenting with optic neuritis and myelitis in adults and diagnosed according to recently proposed international consensus criteria.¹ Onset of MOGAD with preceding infections has been pointed out repeatedly.^2,3 Transmission of MOG-IgG from blood into breast milk has not been reported previously.

We present the case of a lactating and partially breastfeeding woman with life-threating manifestation of MOGAD after enterovirus infection. MOG-IgG tested positive in the patient's serum and in her breast milk, but was negative in the breastfed infant's serum.

Standard Protocol Approvals, Registrations, and Patient Consents

Written informed consent was obtained and ethical approval granted by the ethics committee of the medical faculty of LMU Munich.

Data Availability

Data not published in this article will be made available upon reasonable request.

Case Presentation

A 38-year-old woman presented in our emergency department with acute urinary retention (Figure 1); all dates are presented relative to day of admission. Patient history revealed preceding arthralgia, fatigue, and an episode of moderate gastrointestinal infection 14 days before symptom onset. Neurologic admission examination revealed thermal hypesthesia of the right leg and bladder dysfunction requiring catheterization. CSF analysis revealed lymphocytic pleocytosis (968 cells/μL, 8% neutrophiles, Table). Cranial CT displayed no abnormalities. Therapy with ceftriaxone, ampicilline, and aciclovir for possible infectious meningo-encephalo-myelitis was initiated; cranial and spinal MRI (c/sMRI) revealed multiple diffuse T2-hyperintense white matter lesions with emphasis around the thalamus and cerebellum, T2-hyperintense signals in the right optic nerve, long-extending transverse myelitis (LETM), predominantly in the thoracic gray matter including an H-sign, and lesions within the spinal conus (Figure 2, A and B). Because of the lesion distribution, MOGAD was included early as an important differential diagnosis and MOG-IgG testing requested besides broad infectious and autoimmune testing.

Figure 1. Summary of the Clinical Disease Course and Therapy.

Time course from initial symptom onset and admission to long-term outpatient follow-up 20 months later with relevant events including important sampling dates of MOG-IgG measurements. Days after admission are indicated. MOG-IgG = myelin oligodendrocyte glycoprotein-IgG.

Table.

CSF Parameters and MOG-IgG in Serum Over Time

Day postadmission	1	5	10	24	46
CSF parameters
CSF cells/μL	968	381	61	33	15
% Lymphocytes	64	59	85	90	94
% Neutrophiles	8	28	—	—	—
% Monocytes	23	13	15	10	6
CSF protein mg/dL	118	82	73	29	25
Oligoclonal bands	negative	positive	n.d	positive	negative
MOG-IgG in patient serum
Fixed cell-based assaya	1:320 (d 3)		1:1,000 (d 11)	1:100 (d 24)	1:320 (d 48)
Live cell-based assayb		1:81920 (serum) 1:2048 (CSF) d 13		1:1,280 (d 24)	1:1,280 (d 78)
MOG-IgG in breast milkc	positive (1:64) Sampling: 4 d after admission and before start of immunotherapy
MOG-IgG serum of infantc	Negative (0) Sampling: 24 d after admission

Abbreviation: MOGAD = myelin oligodendrocyte glycoprotein antibody-associated disease.

Depicts the time course of key CSF markers (upper part) including cell count with differentiation, protein, and oligoclonal bands. Transient presence of CSF-specific oligoclonal bands with delayed onset reflects central immunoglobin production and has been reported in MOGAD. MOG-IgG levels (lower part) in serum (patients and infant) and in breast milk are shown and results of fixed- vs live cell-based assay presented across the timeline (days after admission are indicated). First positive MOG-IgG result (day 3) turned back 10 days after admission and decreased after methylprednisolone and plasma exchange (day 24). Very high MOG IgG in serum was only detected by live cell-based assay (day 13).

^afixed cell-based assay, Euroimmun.

^bReindl lab, Medical University of Innsbruck, Austria.

^cBoth tested with live cell-based assay.

Figure 2. MOGAD-Associated Lesions on Cerebral and Spinal Cord MRI Over Time.

MRI of the brain (A) and spinal cord (B) at day 2, day 5 (only brain imaging), and on day 48. Unless mentioned otherwise, images are T2-weighted fluid attenuated inversion recovery coronal for cerebral and T2 turbo spin echo sagittal and axial for spinal sequences. Arrows indicate lesions. (A) MRI of the brain on day 2 after emergency department admission; (A) T2 hyperintense right thalamic lesion with diffuse margins and (B) T2 hyperintense right cerebellar peduncule lesion with diffuse margins; on day 5 with (C) progression of the thalamic lesion and (D) with several new white matter lesions in both hemispheres; on day 48 (E and F) regression of most lesions after therapy with steroids and plasma exchange (PE). (B) MRI of the spinal cord including the conus (right side) on day 2 with longitudinally extended transverse myelitis LETM (A) in the cervical spine and (B) with focus around thoracic levels (C) axial images depicts focus around central gray matter, forming the H-sign on cervical level 7 (D) longitudinal inflammatory lesions in the lower spinal cord and conus; on day 48 resolution/regression of (E) cervical, (F and G) thoracic level 9, including axial images and (G) lower thoracic/conus lesion. MOGAD = myelin oligodendrocyte glycoprotein antibody-associated disease.

Within 24 hours of admission, symptoms progressed rapidly, and the patient was transferred to the neurologic intensive care unit. Follow-up c/sMRI (after 5 days, Figure 2A) showed several new subcortical lesions and progressive thalamic signal alterations, but no new spinal cord lesions. Six days after hospital admission, she was comatose (EDSS: 9.5). EEG revealed nonconvulsive status epilepticus which was terminated after immediate treatment with levetiracetam.

Ten days after admission, serum MOG-IgG returned positive with a titer of 1:320 (1st sampling day 10 fixed cell-based assay [FCBA], Euroimmun), and MOGAD was diagnosed.¹ Subsequent live cell-based assay (LCBA) confirmed positive MOG-IgG at very high levels (2nd sampling day 13, serum: 1:81920, CSF: 1:2048; Reindl lab, Medical University of Innsbruck, Austria, Table). Aquaporin-4-IgG and other autoantibody panels were entirely negative. Enterovirus PCR tested positive in a stool sample (Coxsackie B5 variant), but not in serum or CSF. Serum enterovirus IgM and IgA were elevated with borderline positive IgG, congruent with recent gastroenteritis.

With confirmed diagnosis of MOGAD and negative neuroinfectious PCR panel diagnostic, calculated anti-infective therapy was terminated. Simultaneously, high-dose glucocorticoid (GC) therapy (1 g methylprednisolone per day for 5 days with oral taper) was initiated with subsequent 7 plasma exchange (PE) cycles because of persistent severe neurologic deficits. During PE, clinical symptoms improved significantly after the 3rd PE cycle.

However, bladder dysfunction persisted requiring continuous urinary catheterization. Owing to the severe onset and persistent high MOG-IgG serum titers, long-term immunosuppressive therapy with azathioprine (150 mg/d) was initiated and oral prednisolone further tapered until 8 months after disease onset. Repeated c/sMRI revealed resolution of most lesions without indication of new lesions (Figures 1 and 2, A and B), 20 months after onset the patient relied on intermittent bladder self-catheterization, but showed no further neurologic deficits (EDSS 3.0), whereas MOG-IgG remained positive (sample collected 20 months after onset, LCBA: 1: 1,280).

The patient was partially breastfeeding her healthy 16-months-old infant at admission. Lactation was terminated by bromocriptine on the intensive care unit by gynecologic recommendation. Breast milk was collected 11 days after admission (before start of immunotherapy) and tested positive (1:64) for MOG-IgG, whereas MOG-IgG was not detectable in the serum of the breastfed infant at the tested time point, 3 weeks after the breast milk was collected (day 24, LCBA, Reindl lab, Medical University of Innsbruck, Austria, Table).

Discussion

Here, we demonstrate that MOG-IgG is detectable in breast milk of a breastfeeding woman with a severe onset of MOGAD and very high MOG-IgG titer in serum. Passive transfer of (auto)-antibodies to breast milk is a facultative process underlying a temporal dynamic with multiple contributors including molecule size, antibody type, and titers. Neonatal Fc receptor (FcRn) facilitates IgG transfer; just recently, the FcγRIIB was identified as a key transporter of maternal IgG transfer across the mammary gland.⁴ Breast milk IgG concentration is usually the highest in the early colostrum and decreases along the first year of lactation, before increasing again during prolonged lactation.^5,6 The very high amount of MOG-IgG in the patient's serum and the very late stage of lactation for the 18-month-old infant probably contributed to the detectability of MOG-IgG in breast milk. The MOG-IgG ratio breast milk to serum was 1:1,280, in line with the ratio of total breast milk IgG (approximately 15 mg/L) and serum IgG (approximately 15,000 mg/L).⁶ The numerical value of the end point titers must also be interpreted within the immunoglobulin milieu of the tested compartment. Given the approximately 1:1,000 lower total IgG in breast milk, a titer of 1:64 represents a robust, specific response. MOG-IgG was not detected in the serum of her infant at the time point of sampling. Usually only a small amount, if any, of the milk antibodies is transferred into the infantile circulation by mode of lactation.⁷

In addition, this case illustrates key features and caveats of MOGAD. The management of severe manifestations of MOGAD requires rapid diagnosis and early initiation of intensive attack therapy.^8,9 Acute symptomatic seizures are observed in up 10% of patients with MOGAD, particularly nonconvulsive epileptic status—as seen in this case—is a rare MOGAD presentation for which EEG and seizure-suppressing medication are vital in parallel to immune therapy.¹ Here, a gastrointestinal infection preceded neurologic symptoms and an infectious etiology was initially suspected because of the high CSF cell count. Infectious diseases of the CNS including enterovirus-associated encephalomyelitis may present with similar symptoms as MOGAD, which further challenges diagnosis.¹⁰ However, the onset of MOGAD is frequently preceded by infections; elevated cell counts including CSF granulocytes had been reported.¹¹ Key MRI features with LETM and multiple “cloudy” T2-hyperintense lesions within deep white matter and cerebellar peduncles further supported the diagnosis of MOGAD (Figure 2A). MRI lesions progressed within 5 days, in accordance with recently described intra-attack lesion dynamics as a typical feature of MOGAD.¹² Positive serum MOG-IgG led to immediate initiation of GC plus PE therapy with improvement of the patient condition. Nevertheless, severe bladder dysfunction persisted, which can be observed frequently in patients with MOGAD with LETM and conus involvement.¹³

MOG-IgG titers were extremely high initially, as detected exclusively by the LCBA unlike lower titers in the FCBA. This is in line with results from a previous study, which showed a lower correlation of quantitative MOG-IgG results between the FCBA and the LCBA but excellent correlation between different LCBAs pointing out the importance of LCBA for MOG-IgG testing.^14,15

Limitations are the single samples of breast milk and of serum from the breastfed infant at different time points and the missing measurement of quantitative total IgG concentration in the breast milk.

Our case underlines MOGAD as an essential differential diagnosis in patients with suggested infectious CNS disease and highlights typical MRI features. MOG-IgG may be transferred into breast milk.

Author Contributions

S. Kammermeier: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data. S. Heck: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data. H. Zimmermann: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. D. Engels: drafting/revision of the manuscript for content, including medical writing for content. M. Reindl: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. T. Kümpfel: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data.

Study Funding

The authors report no targeted funding.

Disclosure

D. Engels received speaker honoraria from Alexion, Amgen, Merck, and Roche, all not related to this work. T. Kümpfel has received speaker honoraria and/or personal fees for advisory boards from Alexion/AstraZeneca, UCB, Merck and Biogen, and for lectures/education from Alexion/AstraZeneca, Novartis Pharma, Roche Pharma, Horizon Therapeutics/Amgen, and Chugai Pharma. The institution she works for has received compensation for her service as a member of a steering committee from Roche. T. Kümpfel is a site principal investigator in several randomized clinical trials (Novartis Pharma, Roche Pharma, BMS and Sanofi Genzyme) and in a randomized clinical trials supported by the BMBf (funding code: 01 GM1908E), and her institution has received compensation for clinical trials all outside the present work. The other authors report no relevant disclosures. Go to Neurology.org/NN for full disclosures.