Dean M. Wingerchuk, MD, MSc, FRCP(C)

A Historical Perspective on NMO
Neuromyelitis optica (NMO), also known as Devic’s disease, was identified more than one hundred years ago as a severe disorder affecting the spinal cord (causing paralysis) and both optic nerves (causing blindness). Until recently, it was considered a rare and unusual form of multiple sclerosis (MS) because the affected spinal cord and optic nerve tissues showed lesions containing inflammation and demyelination similar to that seen in MS. Modern scientific and medical technology, including development of magnetic resonance imaging (MRI) and techniques to study the pathological changes in the nervous and immune systems, have led to the discovery that NMO is not only more severe than MS, but is probably a distinct disease. The most convincing evidence to support this argument is the recent discovery of a blood antibody called NMO-IgG, which is detected in NMO but not MS, and is now used to assist in diagnosis. This article will summarize the characteristics, diagnosis, and treatment of NMO.

Progress in understanding NMO began with careful study of the symptoms and signs experienced by patients with the disease and correlating those with the spinal cord and optic nerve abnormalities seen at autopsy. Affected individuals typically experienced an attack of optic neuritis which caused rapid visual loss (over hours to days) in one or both eyes, often with poor recovery of vision. At approximately the same time, they also had a myelitis attack, resulting in paraplegia, loss of sensation of the trunk and legs, and loss of bowel and bladder function. It was originally thought that the disease struck only once, leaving the affected person with severe disability, but reports accumulated throughout the 20th century showing that most patients experienced relapses of the disease. These recurrent attacks of optic neuritis and myelitis were not predictable, occurred months or years apart, and varied in severity and the amount of recovery of function after the attack ended. Examination of the optic nerves and spinal cord after death revealed multiple lesions containing inflammation and demyelination similar that seen in MS, though usually more severe and, importantly, the lesions did not extend to the brain itself. The pathological findings, however, were otherwise similar to MS.

The widespread availability of MRI studies in the last two decades helped foster re-evaluation of NMO. Neurologists recognized patients who seemed to have severe MS with repeated relapses of optic neuritis and myelitis but whose brain MRI scans lacked the typical white matter lesions found in nearly all other MS cases. In contrast, these patients had unusually extensive lesions on their spinal MRI scans taken during acute attacks of myelitis. It was recognized these patients fit the profile of relapsing NMO. Other characteristics, such as the presence of unusual white blood cells (neutrophils) and lack of oligoclonal bands in the cerebrospinal fluid (which 90% of patients with MS have), allowed further differentiation of these patients from typical MS. It became clear that a high proportion of affected people were non-Caucasians, such as African-Americans, Asians, or Hispanics, groups in which typical MS is uncommon, and that approximately 90% are female (compared with about two-thirds female for MS). New diagnostic criteria that allowed neurologists to differentiate NMO from MS were proposed by several groups in the 1990s and there was an increasing awareness of NMO, including the fact that the attacks tend to be very severe and can result in permanent blindness or lack of ability to ambulate within a few years of disease onset. The features of NMO and MS, however, still overlapped enough that most people diagnosed with NMO received the same therapies used to treat MS (Wingerchuk, Lennon, Pittock, Lucchinetti, & Weinshenker, 2006).

Recent Progress in NMO Research
Recent progress in two major research areas has led to a tremendous increased interest in NMO. The first area is pathology. New techniques to study how the immune system can cause nervous system injury showed that NMO lesions are distinct from those of MS. Although both diseases cause focal areas of nervous system inflammation and demyelination, NMO is associated with deposits of both antibodies and complement, a particularly powerful trigger of immune reactions, around blood vessels. Antibodies and complement are thought to also play a role in MS, but they appear to be primarily responsible for NMO lesions (Lucchinetti et al., 2002).

The second advance was the discovery of a blood marker for NMO. This autoantibody, which is called NMO-IgG, is found in about 75% of people with NMO and is not detected in people with typical MS. The antibody is so strongly associated with NMO that it is now part of revised criteria to diagnose the disease (Table 1). It has also allowed a better understanding of the spectrum of the disease. For example, we now know that some patients with NMO have symptoms or brain MRI lesions that reflect damage occurring outside of the optic nerves and spinal cord, something that would have previously led to discarding NMO as a possible diagnosis. It has reinforced the belief that detecting a longitudinally extensive spinal MRI lesion, one that extends over three or more segments of the cord, can help distinguish NMO from typical MS, in which the lesions are much shorter (Figure 1). Furthermore, NMO-IgG has helped us appreciate a broader spectrum of NMO. Some people with recurrent optic neuritis and recurrent transverse myelitis (especially with the “longitudinally extensive” cord lesion) also have detectable NMO- IgG in their blood and belong in the “NMO spectrum” (Wingerchuk, Lennon, Lucchinetti, Pittock, & Weinshenker, 2007). Finally, Japanese investigators have recognized two seemingly distinct types of MS in Japan: Western MS and opticospinal MS, a form that is largely restricted to the optic nerves and spinal cord. More than half of opticospinal patients with MS are positive for NMO-IgG and it is likely the same disease as NMO. We also know that NMO-IgG can predict relapse. If NMOIgG is positive in a patient who experiences their first-ever attack of transverse myelitis associated with a longitudinally extensive MRI lesion, his or her risk of another attack in the next year is greater than 50%.

The target of the NMO-IgG has already been identified: it is a protein called aquaporin-4. There is now great interest in determining whether NMO-IgG actually causes NMO. The discovery of NMO-IgG has allowed for more advances in understanding the immunology and pathology of both NMO and MS. Two very important recent findings include: 1) aquaporin-4 is lost in NMO lesions but not MS lesions, and 2) NMO-IgG can trigger activation of complement, suggesting that it has the potential to cause inflammation and demyelination (Lennon, Kryzer, Pittock, Verkman, & Hinson, 2005; Lennon et al., 2004).

The discovery that aquaporin-4 was the target of NMO-IgG was surprising because aquaporin-4 is located on support cells called astrocytes, rather than the nerve cells themselves, which are covered by the myelin damaged by the immune attack, or the oligodendrocytes, which make myelin. Aquaporins are water channels located on the membranes of cells and are important for maintaining water balance between the inside of a cell and its external environment. It was recently reported that high densities of aquaporin-4 are found on the ends of astrocytes, right next to nodes of Ranvier, which are areas between segments of myelin on nerves (Hinson 2007). The nodes are very important for initiating and maintaining conduction of electrical impulses along nerves. Therefore, immune damage around these sites might cause demyelination even though the primary target is on the astrocyte.

Understanding the exact mechanism of the immune damage and whether NMO-IgG is that cause is an important next step to understanding the disease and finding more effective therapies.

Treatment of NMO

It is important to recognize that treatment recommendations for NMO stem from observations made by patients and their treating neurologists, not from high-quality scientific studies such as randomized controlled trials. Such trials are difficult to organize and support because NMO is not a common disease and therefore there are no Food and Drug Administration (FDA)-approved treatments for NMO. Nevertheless, the rapid increase in understanding of the immunopathology of NMO, together with clinical experience with NMO from several large centers around the world, has resulted in more sophisticated and targeted approaches to NMO treatment (Wingerchuk & Weinshenker, 2005).

Treatment of Attacks

Intravenous corticosteroids, typically methylprednisolone (Solu-Medrol®), are usually used to treat new attacks of optic neuritis or myelitis. Corticosteroids are anti-inflammatory drugs that allow for more rapid improvement of symptoms. Steroids are discussed in detail in MSQR, Volume 26, Number 4, Winter 2007 issue. A commonly used regimen is 1,000 mg daily for five consecutive days. If a severe attack does not improve, or even worsens, despite corticosteroid therapy, it is reasonable to consider a trial of plasmapheresis. This procedure entails removal of blood through an intravenous line, separation of blood cells from the liquid component (serum), discarding the serum and replacing it with albumin from blood donors, combining the albumin with the retained blood cells, and returning the mixture to the patient. This procedure was shown to improve severe attacks that did not respond to corticosteroids and may be particularly effective in NMO. The hypothesis is that plasmapheresis removes substances in the blood that are causing immune injury; it may be that removal of NMO-IgG or related blood antibodies that helps treat NMO attacks. The standard regimen is seven exchanges performed over a period of approximately 2 weeks.

Therapy for Attack Prevention

Because patients with NMO develop virtually all of their permanent neurological disability from ON and myelitis attacks, preventive therapy is necessary once the diagnosis of NMO is secure. In addition, treatment is recommended for patients with one episode of “longitudinally extensive” myelitis who are NMO-IgG positive because they are at very high risk of relapse.

The optimal preventative therapy regimen for NMO is not known. Experience has suggested that current MS therapies such as beta-interferons and glatiramer acetate, are probably ineffective. This makes sense because those treatments do not have their main effects on the “humoral” arm of the immune system, which is the one that manufactures antibodies, the main driver of NMO. For these reasons, treatment regimens that suppress the immune system are recommended.

Chronic immunosuppression may be achieved with general therapies, which have broad immune system effects, or more targeted ones that impact a specific immune system function. The goal is to impact the “humoral” arm of the immune system and prevent new areas of inflammation that cause NMO attacks. Until recently, available therapies were limited to the general immunosuppression category. These include azathioprine (Imuran®; oral therapy, target dose: 2-3 mg/kg daily), mycophenolate mofetil (Cellcept®; oral therapy, target dose 1,000 mg twice daily), and mitoxantrone (Novantrone®; intravenous therapy administered using varying regimens). The oral therapies take some time, usually 4 to 6 months, to exert their full effect on the immune system, so they are often combined with oral prednisone during that time. Once the azathioprine or mycophenolate is fully effective, the prednisone may be tapered away. Mitoxantrone can only be administered temporarily because the risk of heart muscle injury. Other chemotherapeutic drugs have immunosuppressive effects and could theoretically benefit NMO. Another general approach utilizes intravenous infusions of immune globulin (IVIG), which are antibodies pooled from blood donors. It is thought that the infused antibodies may interact with those causing the disease, effectively neutralizing their effects.

The growing evidence that suggests that NMO is driven mainly by antibodies suggests an opportunity for more selective therapy. One candidate is rituximab, an intravenous drug that specifically depletes peripheral B cells, which are the ones involved in production of antibodies. In one series of eight patients with very active NMO, rituximab was able to reduce or eliminate attacks for up to 18 months. When B cell counts recover, typically in 6 to 12 months after treatment, re-treatment may occur (Cree et al., 2005).

It is not known which of the chronic treatments listed above represents the best NMO therapy. Each treatment has advantages and risks. It is recommended that patients with NMO obtain treatment from a clinician with experience in prescribing and monitoring the effects of immunological therapies.

Conclusion

Major advances toward understanding the cause of NMO have occurred within the past five years. The blood test for NMO-IgG is available to assist with diagnosis and to distinguish NMO from MS, and laboratory research focused on NMO-IgG and its target, aquaporin-4, promise to yield further rapid progress. Currently, treatments aimed at suppressing the humoral arm of the immune system appear effective for attack prevention, but more specific, safe, and well-tolerated therapies are needed for NMO.

References

Cree, B. A. C., Lamb, S., Morgan, K., Chen, A., Waubant, E., & Genain, C. (2005). An open label study of the effects of rituximab in neuromyelitis optica. Neurology, 64, 1270-1272.

Hinson, S. R., Pittock, S. J., Lucchinetti, S. F., Roemer, S. F., Fryer, J. P., Kryzer, T. J. et al. (2007). Pathogenic potential of IgG binding to water channel extracellular domain in neuromyelitis optica. Neurology, 69, 2221-2231.

Lennon, V. A., Kryzer, T. J., Pittock, S. J., Verkman, A. S., & Hinson, S. R. (2005). IgG marker of optic-spinal multiple sclerosis binds to the aquaporin-4 water channel. Journal of Experimental Medicine, 202, 473-477.

Lennon, V. A., Wingerchuk, D. M., Kryzer, T. J., Pittock, S. J., Lucchinetti, C. F., Fujihara, K., et al. (2004). A serum autoantibody marker of neuromyelitis optica: Distinction from multiple sclerosis. Lancet Neurology, 364, 2106-2112.

Lucchinetti, C. F., Mandler, R.N., McGavern, D., Bruck, W., Gleich, G., Ransohoff, R. M., et al. (2002). A role for humoral mechanisms in the pathogenesis of Devic’s neuromyelitis optica. Brain, 125, 1450-1461.

Wingerchuk, D. M., Lennon, V. A., Lucchinetti, C. F., Pittock, S. J., & Weinshenker, B. G. (2007). The spectrum of neuromyelitis optica. Lancet Neurology, 6, 805-815.

Wingerchuk, D. M. & Weinshenker, B. G. (2005). Neuromyelitis Optica. Current Treatment Options in Neurology, 7(3), 173-182.

This entry was posted on Monday, February 4th, 2008 at 8:59 am and is filed under All Articles, Featured. You can follow any responses to this entry through the RSS 2.0 feed.