Autoantibody Discovery Sheds Light on NMO, MS Differentiation


Neurologists have long debated whether NMO is simply a variant of multiple sclerosis (MS) or a separate disease, and misdiagnosis is common.

Neuromyelitis optica (NMO) is a demyelinating disorder characterized by optic neuritis and longitudinal, transverse myelitis. It may follow a monophasic course with a single, acute attack, or it may be chronic and relapsing-remitting, with attacks separated by intervals ranging from a few weeks to as long as 3 decades.

Patients with NMO sometimes develop symptoms that cannot be attributed to either optic neuritis or transverse myelitis. Most patients have a normal brain MRI scan at disease onset, but about 10% show unusual lesions around the hypothalamus and in the areas immediately surrounding the third and fourth ventricles, and others have nonspecific brain lesions. Neurologists have long debated whether NMO is simply a variant of multiple sclerosis (MS) or a separate disease, and misdiagnosis is common.

Interest in the controversy was renewed when investigators at the Mayo Clinic announced the discovery of NMO-IgG, a unique serum autoantibody marker that supports the view of NMO as a condition distinct from MS. These authors have urged that the diagnostic criteria for NMO be revised, taking the patient's NMO-IgG status into account.1

The discovery of the NMO autoantibody may help shed light on the pathophysiology of NMO and how it differs from MS. "Keep in mind that the name 'MS' remains a 19th-century generic descriptor for what appears to be a mixed group of disorders for which there have been no specific diagnostic tests," says Vanda Lennon, MD, PhD, professor of neurology and immunology at the Mayo Clinic College of Medicine in Rochester, Minnesota, and leader of the team that identified the autoantibody. "The finding of NMO-IgG in a patient's blood allows delineation of a subgroup of this poorly understood group of disorders, better understanding of its cause, and the tailoring of more specific therapy."


NMO and MS are both demyelinating disorders. Optic neuritis and transverse myelitis can occur in patients with these or other demyelinating disorders. But in NMO, the pathophysiology usually is limited to the optic nerve and the spinal cord, whereas other demyelinating diseases may affect the brain as well. Still, misdiagnosis has been common because the symptoms of certain demyelinating diseases overlap considerably.

Differential diagnosis has important clinical implications; NMO and MS have different natural histories and respond to different drugs. In general, the symptoms of NMO are more severe than those of MS, and its prognosis is worse. A biomarker that allows physicians to distinguish one disease from the other would permit early, accurate diagnosis; help with treatment planning; and spare patients with MS inappropriate dosing with drugs best prescribed for patients with NMO.1

Two observations have led investigators to suspect that an antibody might be involved in the pathogenesis of NMO. First, unlike patients with MS, patients with NMO usually improve after plasmapheresis.2 Second, IgG, IgM, and the products of complement activation from patients with NMO are deposited in a vasculocentric pattern in tissue from the CNS, but not from the stomach, liver, or kidney.2

Furthermore, tied in with the challenge of distinguishing NMO from MS is uncovering the true identity of a condition known as Asian optic-spinal MS, which accounts for 15% to 40% of MS diagnoses in Japan.3 Its symptoms include frequent and severe vision loss as well as spinal cord and brain stem involvement. Many researchers have wondered whether Asian optic-spinal MS might not in fact be the same clinical entity as NMO. The discovery of an antibody common to both diseases would help unravel the mystery.3,4


In an effort to find a biomarker unique to NMO, the Mayo team tested masked serum samples from 102 North American patients who had either NMO or syndromes associated with a high risk for NMO development.3 The syndromes included (1) at least 1 attack of longitudinal transverse myelitis affecting at least 3 consecutive vertebral segments or (2) recurrent optic neuritis.

Of those 102 patients, 45 ultimately met the diagnostic criteria for NMO and 35 remained in the high-risk category. The remaining 22 never developed longitudinal myelitis and did not meet other diagnostic criteria for NMO, and so they received a final diagnosis of MS. Twelve Japanese patients with Asian optic-spinal MS also were tested.

The sera from these 114 patients were compared with samples from a control group consisting of 19 patients with classic MS and 56 patients with other myelopathies, optic neuropathies, and miscellaneous disorders.

After exposing the sera to samples of mouse brain, midbrain, spinal cord, and liver tissue, the investigators identified an immunoglobulin that was specific to NMO and its high-risk syndromes; this antibody occurred with much less frequency in the control patients. "NMO-IgG yielded a characteristic immunohistochemical pattern of binding in mouse CNS tissues," the Mayo team wrote. "It was prominent in pia and subpia and outlined the Virchow-Robin space and microvessels in white and gray matter of the cerebellum, midbrain, and spinal cord. It also bound to subependymal white matter and the subpial layer of midbrain in a mesh pattern. No staining was associated with subarachnoid vessels or the choroid plexus." On further labeling studies, the autoantibody demonstrated an affinity for cerebral microvessels. It was not detected in gut mucosa, smooth muscle, kidney, or liver tissue.

Thirty-three (73%) of the 45 North American patients in whom NMO was diagnosed and 16 (46%) of the 35 high-risk patients tested positive for NMO-IgG. Two (9%) of the 22 patients given a final diagnosis of MS were seropositive (P < .0001, compared with the patients with NMO). All in all, a positive result had a sensitivity of 73% and a specificity of 91% in distinguishing between patients with NMO and those who had optic neuritis or myelitis but did not meet the other diagnostic criteria.

The findings were similar in the Japanese patients: of the 7 who were seropositive, NMO was diagnosed in 6, and a high-risk syndrome in 1, confirming that NMO and Asian optic-spinal MS are indeed the same disorder. None of the control patients were seropositive for NMO-IgG.

Also included in the study was a group of 14 patients with suspected paraneoplastic autoimmunity who had an initially unidentified serum antibody. After they identified NMO-IgG in the larger group, the investigators went back and analyzed the sera of these patients and determined that the unidentified antibody was NMO-IgG. Of those patients, 12 met the clinical criteria either for NMO or a high-risk syndrome.3

Neurologists have greeted the discovery with enthusiasm. "This is very exciting for us in the world of demyelinating diseases," says Victoria Pelak, MD, associate professor of neurology and ophthalmology at the University of Colorado Health Sciences Center in Denver. "It could shed light on a pathogenic mechanism and may help us find a similar antigen in MS."

"This is the first time anyone has identified a biomarker for an idiopathic demyelinating inflammatory disorder," adds Bianca

Weinstock-Guttman, MD, associate professor of neurology at the University at Buffalo, State University of New York, and director of the Baird Multiple Sclerosis Center of the Jacobs Neurological Institute. "It has important diagnostic implications, and it may help us monitor response to therapy and develop more targeted therapies for NMO."

"As the first recognized biomarker for any form of MS, detection of NMO-IgG enables more accurate subclassification of CNS demyelinating disorders-some previously thought to be classic MS-and others thought to be vascular complications of diseases like systemic lupus erythematosus," Lennon told Applied Neurology.

"Although this remains to be confirmed, it looks as if there are disorders that have the NMO autoantibody as a common denominator," says Aaron Miller, MD, medical director of the Corinne Goldsmith Dickinson Center for Multiple Sclerosis and professor of neurology at Mount Sinai School of Medicine in New York City. "That means they may respond to similar treatment modalities. Also, the autoantibody will let us study a more homogeneous group of patients."

"The finding of NMO-IgG adds to a growing body of clinical, laboratory, and immunopathologic evidence that NMO and MS are indeed separate disorders," commented Dean M. Wingerchuk, MD, professor of neurology at the Mayo Clinic Scottsdale and a member of the team that identi-fied the autoantibody.


NMO also is called Devic syndrome or Devic disease after the French physician Eugene Devic, who described a series of cases in 1894 and 1895.4 In Western nations, NMO is much less common than MS, accounting for fewer than 1% of all demyelinating disorders.4 Recent evidence, however, suggests that its presumed lesser prevalence compared with MS may be the result of misdiagnosis because of a lack of clear diagnostic criteria that distinguishes it from MS.5 It is less common in white persons than in those of other races and is more common in non-Western countries, such as India and Japan.5 Like MS, NMO typically occurs in young adults, although it has been described in patients whose ages range from infancy to the ninth decade. Also like MS, the disease is more common in women.5

NMO presents a diagnostic challenge for 2 reasons. First, its hallmarks-optic neuritis and transverse myelitis-also occur in persons with MS. The tip-off is that symptoms are more severe in NMO, and the disease may take a fulminant course.5 As many as 50% of patients lose vision in 1 eye or the ability to walk independently within 5 years of the first episode.3 Second, the visual and spinal symptoms may strike days, months, or even years apart, yet both must occur for a definite diagnosis of NMO to be made.

The proximity of these attacks offers a clue to prognosis, says Wingerchuk. "Patients who present with optic neuritis and myelitis nearly simultaneously-within a few days-have a greater chance of having a monophasic course, the caveat being that we have noted interattack intervals varying from weeks to 3 decades, so a 'monophasic' diagnosis must be viewed with caution," he explained.

"If the interattack interval is a few weeks or months, the patient will almost certainly develop a relapsing course," said Wingerchuk. The relapsing form is more common, affecting 70% of patients by some estimates.5 It is associated with a 5-year survival rate of 68%, compared with a rate of 90% among patients who experience monophasic illness.4


The basic clinical features of NMO include vision loss in 1 or both eyes, paralysis, and possible loss of bowel or bladder function, depending on the vertebrae affected.4,5 Acute transverse myelitis, defined as a severe, bilateral inflammation of the spinal cord that traverses at least 3 contiguous vertebrae, is a typical presenting symptom.5 Some patients also experience painful spasms lasting from 15 seconds to 2 minutes that may occur multiple times a day.5

Symptoms may worsen over the next few hours or days, although 78% to 88% of patients regain at least some function in the affected areas. Some patients develop only minor visual loss, but as many as 40% may experience complete blindness in the affected eye. They may recover some vision, but the losses can accumulate with subsequent attacks.5 Prodromal symptoms, including fever, headache, malaise, myalgia, sore throat, and respiratory or GI symptoms, occur in 30% to 50% of patients.4,5

Cognitive impairment, a prominent feature of MS, is seldom reported in patients with NMO. However, "when patients are tested with formal, neuropsychiatric tests, you do find it," Pelak said. She predicts that highly sensitive tests will be able to ferret out evidence of cognitive dysfunction, even in patients with no apparent brain lesions. "There's something going on below the level of the MRI scan," she said.

Table - Differentiating NMO from MS
Affected regions
Optic nerve/spinal cord
Any white matter
Attack severity
Mild to moderate
Cranial MRI findings
Normal or nonspecific*
Spinal MRI lesions
≥ 3 central contiguous segments
< 1 segment
CSF cells
≥ 50, polymorphonucleocytes
< 50, lymphocytes
CSF bands
Usually absent
Usually present
Concurrent autoimmune disease
Present in 50% to 70% of cases; mostly SLE or Sjögren syndrome
NMO-IgG seropositivity
Present in >70% of cases
Rare, even if patient has myelitis or optic neuritis
Cognitive impairment
Thought to be rare to nonexistent; now under study
Affects 45% to 65% of patients
Prednisone; immunomodulators; plasmapheresis
Prednisone; interferon-beta 1a or 1b; glatiramer acetate;immunomodulators
Life expectancy
5-year mortality ≈ 10% with monophasic form; ≈ 30% with multiphasic form; death usually results from cardiac arrest or respiratory failure
Usually normal, although marked by morbidity
NMO, neuromyelitis opitica; MS, multiple sclerosis; CSF, cerebrospinal fluid; SLE, systemic lupus erythematosus. *Lesions resembling those of MS are seen in ≤ 10% of NMO patients.


Clinicians have proposed several sets of diagnostic criteria since NMO was first described. They all include optic neuritis and transverse myelitis, with later criteria that emerged with progress in laboratory and imaging technology.

In 1999, Wingerchuk and colleagues suggested that 3 absolute criteria must be present for a diagnosis of NMO: optic neuritis, acute myelitis, and no symptoms implicating other CNS regions. Also, the patient had to have at least 1 of these 3 supportive criteria1:

  • A brain MRI scan at disease onset that is normal or does not meet the imaging criteria for MS.
  • Spinal cord lesions extending over at least 3 vertebral segments.
  • A finding of at least 50 white blood cells/microL or 5 neutrophils/µL in the cerebrospinal fluid.

However, these criteria did not account for patients with neurologic symptoms originating outside the optic nerve or spinal cord or for the very few patients with brain MRI lesions resembling those of MS. Furthermore, some patients who met the criteria actually did turn out to have MS.

After they identified the NMO-IgG autoantibody, the Mayo Clinic investigators determined whether they could enhance its diagnostic precision. They calculated the sensitivity and specificity of each of the 1999 supportive criteria, as well as the autoantibody, in 96 patients with a final clinical diagnosis of NMO and 33 patients with MS. They then performed similar calculations using those criteria in every possible combination. The best combination was a 99% sensitivity and 90% specificity for NMO and consisted of at least 2 of these criteria:

  • NMO-IgG seropositivity.
  • A longitudinally extensive spinal cord lesion (greater than or equal to 3 contiguous vertebral segments).
  • Brain MRI scan at onset that was nondiagnostic for MS (ie, was normal or showed nonspecific lesions).

With a specificity of 93%, NMO-IgG seropositivity was the best single rule-in criterion. The absence of a long spinal cord lesion was the best rule-out criterion, with a sensitivity of 97%.

Based on these findings, the authors proposed that the diagnostic criteria be revised to include optic neuritis, myelitis, and at least 2 of the 3 supportive criteria. CNS involvement beyond the optic nerves and spinal cord would not be incompatible with the diagnosis.1

By lifting the requirement that neurologic manifestations be limited to the optic nerve and spinal cord, the revised criteria reflect a view among a growing number of neurologists that NMO may actually encompass a spectrum of illnesses. "Our data demonstrate that a wide variety of neurologic symptoms may precede or accompany NMO and may or may not be associated with an identifiable CNS lesion," the authors wrote.1

The discovery of the autoantibody supports this position. "Not everyone who is NMO-IgG-positive has classic Devic disease, but they most likely have a related disorder," said Miller. The Mayo team has urged that the concept of "pure" NMO be abandoned (Table).1,6


NMO has several important pathologic characteristics7:

  • Severe inflammatory or necrotic lesions.
  • Infiltration by eosinophils and neutrophils.
  • Vascular hyalinization.
  • Perivascular immunoglobulin reactivity, especially of IgM.
  • Perivascular complement activation.

Several lines of evidence point to a humoral, B-cell-mediated pathogenesis.8-10

Among the studies is an analysis by Lucchinetti and colleagues10 of 82 lesions from 9 autopsy-confirmed cases of NMO. An identical pathology that included active, extensive demyelination in the central portion of the spinal cord was identified in each case. Lesions also were seen in the optic nerves but were inactive and in some cases showed evidence of remyelination. Inflammatory infiltrates from the active spinal lesions contained large numbers of macrophages associated with perivascular granulocytes, eosinophils, and rare CD3+ T cells. Also detected were pronounced perivascular deposition of complement C9 neoantigen (a marker of complement activation) and immunoglobulins, with IgM outweighing IgG.

The lesions also showed evidence of vascular fibrosis and vessel proliferation. The study authors concluded that the extent of complement activation, eosinophil infiltration, and vascular fibrosis suggested that humoral immunity plays a role in the pathogenesis of NMO.

Another group of investigators examined 4 patients who developed NMO after undergoing thymectomy for myasthenia gravis (MG).11 They noted that despite the rarity of both disorders, MG is 150 times more prevalent among patients with NMO than in the general population. "Dysregulation of B-cell autoimmunity in MG, possibly exacerbated by loss of control over autoreactive cells as a result of thymectomy, may predispose patients to the development of NMO," they wrote.

As mentioned, NMO-IgG has attracted attention not only for its diagnostic potential but for the insights it may provide into the pathogenesis of NMO and related disorders. After their initial identification of the autoantibody, the Mayo investigators went on to show that the autoantibody binds selectively to the aquaporin-4 (AQP-4) water channel, a component of the dystroglycan protein complex located in the astrocytic foot processes at the blood-brain barrier.2 AQP-4 is the predominant water channel protein in the CNS. These findings also make it the first autoantigen to be implicated in an inflammatory demyelinating disorder of the human CNS. Two possible mechanisms may explain the pathogenic process: an inflammatory response elicited by the binding of NMO-IgG to the AQP-4 channel or dysregulation of tissue water homeostasis, leading to inflammation and demyelination without complement activation.

In short, the researchers maintained, "NMO may represent the first example of a novel class of autoimmune channelopathies."2


The goals of treatment are to manage the acute attack, prevent relapse, treat any residual symptoms, and rehabilitate patients who have persistent disabilities. Early, appropriate treatment is essential to limit damage, lower the risk of future attacks, and minimize their severity when they do occur. Most approaches are based on clinical experience rather than formal evidence. Immunosuppression, corticosteroids to control inflammation, and plasmapheresis are the current treatment mainstays.

At most centers, the first-line treatment for acute flare-ups is intravenous methylprednisone in a dosage of 1000 mg/d for 3 to 7 days. This may be followed by a course of oral corticosteroids, with doses tapering over 8 weeks.

Several controlled double-blind trials have shown that patients with NMO as well as those with MS and other inflammatory demyelinating disorders who do not respond to corticosteroids may experience significant functional improvement following plasma exchange.12-14 In at least 1 of the studies, patients with NMO derived the greatest benefit.12 A typical course of treatment consists of 55 mL/kg/exchange performed every second day for a total of 7 exchanges.

Azathioprine (Azasan) is often administered during the acute phase along with the corticosteroid, with an eye toward preventing future episodes. The usual target dosage is 2 to 3 mg/kg/d, administered over several months, tapering to maintenance doses of 100 to 200 mg/d. The corticosteroid also may be tapered and finally discontinued once the patient is clinically stable. Should a flare-up occur while the patient is taking azathioprine, he or she may have to continue taking the corticosteroid at the lowest dose possible to maintain remission.

The patient who relapses despite these efforts presents a challenge. Mitoxantrone (Novantrone), cyclophosphamide, chlorambucil (Leukeran), and intravenous immunoglobulin all have shown some effectiveness, but only in small studies of limited duration.

Rituximab (Rituxan), originally developed to treat non-Hodgkin lymphoma, showed promise in an open-label study of 8 patients with NMO who received the drug to achieve B-cell depletion.15 Relapses stopped in 6 of the 8 patients, and the median attack rate dropped from 2.6 to 0 attacks/patient/year. Seven of the patients experienced significant recovery of neurologic function over 1 year of follow-up. A larger double-blind study is scheduled to begin in 2007.

Symptomatic treatment consists of the medications and therapies necessary to treat any pain, stiffness, spasms, or bowel or bladder problems that result from the attacks. Patients with major, persistent disabilities may require help from physical and occupational therapists and social workers, as well as doctors and nurses. Weinstock-Guttman warns that immobilized patients should be advised to take an aspirin daily to lower the risk of pulmonary embolism.


Discovery of the NMO-IgG has furthered understanding of the pathogenesis of NMO and, by inference, that of other inflammatory demyelinating disorders. It now seems clear that NMO is a humorally mediated illness that responds to treatment with B-cell suppression. It is also apparent that NMO and Asian optic-spinal MS are the same condition.

Many neurologists no longer consider NMO to be a single, "pure" disease. Instead, they see it as a spectrum of disorders that have a common underlying pathology that may respond to similar forms of treatment.


1. Wingerchuk DM, Lennon VA, Pittock SJ, et al. Revised diagnostic criteria for neuromyelitis optica. Neurology. 2006;66:1485-1489.

2. Lennon VA, Kryzer TJ, Pittock SJ, et al. IgG marker of optic-spinal multiple sclerosis binds to the aquaporin-4 water channel. J Exp Med. 2005;202:473-477.

3. Lennon VA, Wingerchuk DM, Kryzer TJ, et al. A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet. 2004;364:2106-2112.

4. Cree BA, Goodin DS, Hauser SL. Neuromyelitis optica. Semin Neurol. 2002;22:105-122.

5. Wingerchuk DM. Devic's disease. Syllabus presentation.

6. Kelly J. New biomarker differentiates neuromyelitis optica from MS. Neurol Rev. 2005:13. Available at: Accessed August 7, 2006.

7.Wingerchuk DM. Neuromyelitis optica: progress in diagnosis and management. Presented at: Rare Neuroimmunologic Disorders Symposium; August 20, 2004; Baltimore.

8. Dalakas MC. B-cells in the pathophysiology of autoimmune neurological disorders: a credible therapeutic target. Pharmacol Ther. 2006 April 24; [Epub ahead of print].

9. Wingerchuk DM. Evidence for humoral autoimmunity in neuromyelitis optica. Neurol Res. 2006;28:348-353.

10. Lucchinetti CF, Mandler RN, McGavern D, et al. A role for humoral mechanisms in the pathogenesis of Devic's neuromyelitis optica. Brain. 2002;125:1450-1461.

11. Kister I, Gulati S, Boz C, et al. Neuromyelitis optica in patients with myasthenia gravis who underwent thymectomy. Arch Neurol. 2006;63:851-856.

12. Weinshenker BG, O'Brien PC, Petterson TM, et al. A randomized trial of plasma exchange in acute central nervous system inflammatory demyelinating disease. Ann Neurol. 1999;46:878-886.

13. Keegan M, Pineda AA, McClelland RL, et al. Plasma exchange for severe attacks of CNS demyelination: predictors of response. Neurology. 2002;58:143-146.

14. Bennetto L, Totham A, Healy P, et al. Plasma exchange in episodes of severe inflammatory demyelination of the central nervous system. A report of six cases. J Neurol. 2004;251:1515-1521.

15. Cree BAC, Lamb S, Morgan K, et al. An open label study of the effects of rituximab in neuromyelitis optica. Neurology. 2005;64:1270-1272.

NORRA MacREADY is a freelance medical news writer in Sherman Oaks, California.

Recent Videos
Chelsie Monroe, MSN, APN, PMHNP-BC, and Karl Doghramji, MD, FAASM, DFAPA
Chelsie Monroe, MSN, APN, PMHNP-BC, and Karl Doghramji, MD, FAASM, DFAPA
Chelsie Monroe, MSN, APN, PMHNP-BC, and Karl Doghramji, MD, FAASM, DFAPA
Video 8 - "Treatment Augmentation in a Patient with Narcolepsy and ADHD"
Video 7 - "Complex Case of a 23-Year-Old Male College Student Suffering From Narcolepsy Symptoms"
Video 6 - "Patient-Centered Approach: Adapting Narcolepsy Treatments to Address Adverse Events and Mitigate Misuse Risks"
Video 5 - "Clinical Treatment Strategies for a Patient Suffering from EDS and Hypnagogic Hallucinations"
Video 2 - "Narcolepsy Evaluation, Management, and Treatment Considerations"
Video 2 - "Diagnostic Practices for Narcolepsy"
© 2024 MJH Life Sciences

All rights reserved.