Article In Brief
Findings from a new study raise the question of whether MOG antibody testing for unexplained encephalitis should be added to the growing list of conditions in children with unexplained central nervous system conditions.
A team of Spanish investigators have identified antibodies against myelin oligodendrocyte glycoprotein (MOG) in children with unexplained encephalitis, expanding the number of syndromes associated with this antibody.
The study, published in March in The Lancet Neurology, raises the question of whether MOG antibody testing for these conditions should be added to the growing list of conditions in children with unexplained central nervous system conditions.
In the last five years, MOG antibodies have been linked to acute demyelinating syndromes that include optic neuritis, neuromyelitis optica spectrum disorder, acute disseminated encephalomyelitis (ADEM), and ADEM followed by optic neuritis. The course of treatment is immunotherapy and the majority of the children get better and have no residual effects. But some children go on to have relapses, and no one is sure why.
While most investigators have focused on demyelinating conditions, the Spanish Pediatric anti-MOG Study Group decided to also look for the antibody in pediatric cases of encephalitis without any changes in white matter.
“The spectrum of MOG-antibody specific diseases is wider than we thought,” said the senior study author Josep Dalmau, MD, PhD, professor of neurology at the Catalan Institute for Research and Advanced Studies in Barcelona, and professor of neurology at the Perelman School of Medicine at the University of Pennsylvania.
“At least in children, the spectrum should now include encephalitis, even if MRI doesn't show clear evidence of white matter involvement.”
Dr. Dalmau added that the current criteria and definition of MOG-antibody associated disease should be modified to include unexplained causes of encephalitis, some of the other findings about response to treatment, and the poor prognosis in some of the patients.
Study Design, Findings
The first author of the paper, Thaís Armangue, MD, of the pediatric neuroimmunology unit at the Sant Joan de Déu Children's Hospital at the University of Barcelona, led the collection of the data and the analysis. She reached out to doctors at 40 pediatric medical centers in Spain for this prospective, observational study.
She and her colleagues recruited 535 children from two different cohorts: 239 patients who had been diagnosed with demyelinating syndromes (cohort A) and 296 patients with encephalitis (cohort B). They conducted MOG-antibody testing and had access to imaging and clinical data. Clinical questionnaires on each child were filled out at the initial workup and every six months until the end of the study period—from June 2013 to December 2018. They identified syndromes in each patient, as well as their treatment and their response to therapy. They also looked at the number of relapses over the study period.
The research team found that 116 patients in the sample had MOG-antibodies: 94 patients (39 percent) in cohort A and 22 patients (7 percent) in cohort B. The median age of the children was 6.2 years old. The MOG-antibody positive children had a number of conditions, including 46 patients (68 percent) with ADEM; 22 patients (9 percent) with encephalitis other than ADEM; 20 (17 percent) with optic neuritis; 13 patients (11 percent) with myelitis; six (5 percent) with neuromyelitis optica spectrum disorders; and nine (8 percent) with other disorders.
Sixty-four patients in cohort B were ultimately diagnosed with autoimmune encephalitis and 22 patients (34 percent) had MOG-positive antibodies, which were more common than all of the neuronal antibodies combined.
In fact, of the 116 MOG-positive patients, more than half (59 percent) had clinical features of encephalitis. Forty-six (68 percent) met diagnostic criteria for ADEM, but 22 did not. These 22 patients had a range of symptoms, including decreased level of consciousness (100 percent), seizures (16 percent), fever (59 percent), motor deficits (40 percent), abnormal behavior (50 percent), and brainstem cerebellar dysfunction (23 percent).
The brain MRIs were also different from those diagnosed with ADEM. Twelve of the 22 patients had extensive bilateral cortical involvement and five had basal ganglia or thalamic changes; four others had isolated thalamic or basal ganglia involvement; and four had a single cortical or cortical/subcortical lesion. Only two of the children had normal brain MRIs.
Following the children over time proved very important, said Dr. Dalmau. By 42 months, 33 patients (28 percent) of the 116 MOG-antibody positive patients had relapses, including 17 out of 100 diagnosed at their first episode.
At the first episode 16 were not treated and 100 received first-line immunotherapy; 86 of these 100 patients received treatment with steroids, intravenous immunoglobulin, or plasma exchange. Additional treatments included rituximab (in three patients, one with anakinra). At relapse, all but one of them received another round of treatments, with additional treatment with rituximab in a third of the patients who relapsed. The majority, 85 percent, recovered completely but the remaining patients had moderate to severe problems. One of the children had severe cranial hypertension and died.
Those with the worst prognosis included individuals with ADEM-like relapses that turned into a leukodystrophy-like pattern on MRI, cortical encephalitis, and brain atrophy.
Dr. Dalmau said that there were several surprises in the study. The first was that they identified MOG-antibodies in non-ADEM encephalitis cases.
“Some of these patients would not have been properly diagnosed,” he said. This study also highlighted the importance of continued antibody testing. If MOG-antibody is still detected several months after the treatment ended, there is evidence that these patients have a higher risk for relapse.
Dr. Dalmau's team found several patients with multiphasic ADEM who had more than two relapses—some up to five. This means that the criteria for this disorder as it stands, which allows for only two relapses, should be modified, said Dr. Dalmau. It could alter the way these children are treated, he added.
Finally, he said, antibody testing for MOG should be included in the initial workup of pediatric encephalitis once viral infections and other causes of encephalopathy are ruled out. He said that no one knows what leads the body's B cells to make antibody against the MOG protein.
“I think our study will be helpful to many physicians and their patients,” he said. “Despite the number of patients we followed, this is still a relatively small study. This has to be verified by other investigators.”
In an accompanying editorial in The Lancet Neurology, Romain Marignier, MD, PhD, a neurologist at the Lyon Neurological Hospital in France, wrote: “By contrast with previous works in the field, the authors have not limited their analysis to children with demyelinating disorders, but also included a large cohort of children presenting with encephalitis other than acute disseminated encephalomyelitis. The first interesting aspect of this study was the confirmation of important characteristics of MOG antibodies in pediatric demyelinating disorders at a large scale. Second, the study found that the clinical presentation of MOG antibody-associated syndromes is age dependent, with acute disseminated encephalomyelitis predominant in young children and neuromyelitis optica including optic neuritis or myelitis predominant in older children. Third, Armangue and colleagues observed that MOG antibodies are not associated with multiple sclerosis, as already reported in a previous cohort of children and more recently in adults.”
“The work by Armangue and colleagues is a milestone in the understanding of MOG antibody-associated syndromes, as it showed that MOG antibodies are not restricted to demyelinating syndrome but should also be considered in children presenting with encephalitis,” wrote Dr. Marignier, who specializes in neuro-inflammatory disorders with a special focus on neuromyelitis optica.
“In view of the very broad clinical spectrum now associated with MOG autoimmunity, the next challenge will be to identify the optimal therapeutic strategy for each clinical presentation. This objective is closely connected to a better understanding of the pathogenic role of MOG antibodies, and the need for early, robust, and specific prognostic factors of relapse and disability.”
Brenda Banwell, MD, FAAN, chief of child neurology and professor of neurology and pediatrics at The Children's Hospital of Philadelphia and the Perelman School of Medicine at the University of Pennsylvania, has been the principal investigator of a longitudinal study that has been following children presenting to Canadian pediatric neurology clinics with their first demyelinating attack. The study is now in its 17th year.
She and her colleagues recently went back to the archive samples and ran MOG antibody tests on 1400 samples: 31 percent were positive for MOG-antibody. The Canadian study did not enroll children with non-demyelinating illness or infectious encephalitis. Her study was published last September in JAMA Neurology.
“The Spanish cohort study is so valuable because it highlights the need for MOG-antibody testing in children with unexplained seizures with encephalitis, and in presumed autoimmune encephalitis not meeting criteria for ADEM, as well as in children with demyelinating attacks of the central nervous system. MOG-antibody is emerging as a biomarker in children with a lot of different presentations. We should think about testing for MOG in patients with a more diverse set of clinical symptoms.”
Dr. Banwell said that some children who present with seizures and encephalitis could be missed and not get the proper immunomodulating treatments. She cautions that it is important to get proper MOG testing in a reliable laboratory. The cell-based assay is much more accurate and reliable, she said.
“The jury is still out on the pathogenic role of the antibody,” said Michael Sweeney, MD, assistant professor of neuroimmunology at the University of Louisville in Kentucky. “The story is still being written how MOG antibody leads to disease. Until now, we have been missing the identification of this antibody in our patients with encephalitis. We still need to dissect what these patients look like and how they respond to treatment and understand their prognosis.”
He added: “I think it will be important to better define which patients have MOG antibodies and what it means.”
“This is a major finding because it is not commonly part of the typical evaluation for autoimmune encephalitis,” added Kyle M. Blackburn, MD, an instructor in neuroimmunology at UT Southwestern Medical Center. “We typically associate MOG antibodies with demyelinating presentations, like ADEM and optic neuritis. Given the high rate of MOG-positive patients with encephalitis in this study, these findings suggest MOG antibodies should become part of the evaluation in children with encephalitis.”
Drs. Dalmau, Banwell, Sweeney, and Blackburn had no disclosures.