User login
Urgency of treatment is something that many physicians may not fully appreciate when it comes to neuromyelitis optica (NMO), according to experts on this rare autoimmune demyelinating disorder. This may be partly due to its similar presentation to multiple sclerosis (MS), said Michael Levy, MD, PhD, associate professor, Harvard Medical School, research director, Division of Neuroimmunology & Neuroinfectious Disease, and director, Neuroimmunology Clinic and Research Laboratory, at Massachusetts General Hospital in Boston. But while the two conditions share many clinical characteristics, “immunologically, they are about as different as can be,” he warned.
The urgency of distinction is important because where MS is known to have a relatively gradual progression, NMO is now red-flagged to potentially cause rapid and irreversible damage. While the course of MS might be described as a slow burn, NMO should be treated like a wildfire.
“That message has gotten muddled, particularly because acute treatment in MS has never been shown to affect outcome,” said Jeffrey Bennett, MD, PhD, professor of neurology and ophthalmology at the University of Colorado School of Medicine, Aurora. In contrast, rapid diagnosis and treatment of NMO “means potentially preventing future devastating neurologic injury,” he said.
First described by Dr. Eugène Devic in 1894, and sometimes known as Devic’s disease, NMO is believed to have a prevalence that varies widely depending on ethnicity and gender. A recent report suggests a prevalence of approximately1/100,000 population among Whites with an annual incidence of less than 1/million in this population, while the prevalence is higher among East Asians (approximately 3.5/100,000), and may reach as high as 10/100,000 in Blacks.1 It has a high female-to-male ratio (up to 9:1) with a mean age of onset of about 40 years, although pediatric cases are described.
It has long been recognized that NMO lacks the “neurocerebritis” of MS, with inflammation predominant in the optic and spinal nerves, but it was not until 2004 that researchers at the Mayo Clinic identified serum aquaporin-4 immunoglobulin G antibodies (AQP4-IgG) that could reliably distinguish NMO from MS. In 2015, the international consensus diagnostic criteria for neuromyelitis optica2 cited core clinical characteristics required for patients with AQP4-IgG-positive NMO spectrum disorder (NMOSD) “including clinical syndromes or MRI findings related to optic nerve, spinal cord, area postrema, other brainstem, diencephalic, or cerebral presentations.” Rarely, NMO patients can be seronegative for AQP4-IgG, but are still considered to have NMOSD for which non-opticospinal clinical and MRI characteristics findings are described. MS patients testing negative for AQP4-IgG should also be tested for the related myelin oligodendrocyte glycoprotein antibody disease (MOGAD), which has a prevalence about four to five times greater than NMO, Dr. Bennett said.
Testing
Because both NMO and MOGAD can be identified by antibodies, they are less commonly misdiagnosed as MS compared to previously. But, prior to the identification of the AQP4-IgG antibody in 2004, the misdiagnosis rate of NMO was probably about 95% said Dr. Levy.
“Of course, before we had the antibody test or clinical criteria, we couldn’t confirm a diagnosis of NMO, so basically everyone had a diagnosis of MS, and after the antibody test became commercially available in 2005/2006, we could confirm the diagnosis, with our study in 2012 showing a much lower misdiagnosis rate of 30%.”3 More recently, the misdiagnosis rates are even lower, he added. A recent study out of Argentina found a rate of only 12%.4
The specificity and sensitivity of cell binding assay serum AQP4-IgG testing is roughly 99% and 90%, respectively, better than ELIZA testing (which has a sensitivity in the 60-65% range), said Dr. Bennett. “That’s why we highly emphasize to physicians, that if you have a suspicion for NMOSD you go to a cell binding assay, and make sure that where you’re sending the serum, the lab can do that procedure.” Still, because of the risk of false positives, he urges restraint in testing for the disorder in the absence of a high suspicion for it. “If you test a lot of people indiscriminately for a rare disorder, you get a lot of false positives because the actual true positives are a very small fraction of that group. So, even with a specificity of around 99% that means 1% of the people you test are falsely positive. And if you’re testing a group of people indiscriminately, then your true positives are less than 1% by far, so then most of the people that you pick up are not truly with disease.”
Acute Treatment
While misdiagnosis of NMO as MS is less common than previously, it is still a concern, not only because of the irreversible risks associated with delayed acute treatment, but also the risks of inappropriate preventive MS therapy, which could be harmful to patients with NMO.
Acute flare-ups of NMO and MOGAD are currently all treated with the same decades-old mainstays for acute MS — intravenous steroids and plasma exchange — but the approach is more aggressive. Retrospective studies have shown that, for NMO, plasma exchange has shown an increased likelihood of improvement versus steroids alone, said Dr. Bennett, but since time is of the essence, treatment should begin before a definitive diagnosis is confirmed.
“What’s limiting our patients is, number one, recognizing NMOSD when the attack is happening in your face. You’ve got to know, hey, this is NMOSD or I’m suspicious of NMOSD and hence, I need to treat it urgently because the outcome has a high probability of not being good. You’ve got to realize that this is NMOSD before the test comes back, because by the time it comes back positive in several days, you’re probably missing the optimal window to treat. The point is to know that the presentation in front of you, the MRI pictures in front of you, the laboratory tests that you might have done in terms of spinal fluid analysis, all highly suggest NMOSD. And so hence, I’m going to take the chance that I might be wrong, but I’m going to treat as if it is and wait for the test to come back.”
Realistically, the risks associated with this approach are minor compared with the potential benefit, Dr. Bennett said. “For plasma exchange, there’s the placement of the central line, and the complications that could happen from that. Plasma exchange can lead to metabolic ionic changes in the blood, fluid shifts that might have to be watched in the hospital setting.”
While waiting for diagnostic results, one clue that may emerge from acute treatment is recovery time. “The recovery from MOGAD attacks is really distinct,” said Dr. Levy. “They get better a lot faster. So, if they’re blind in the hospital, and 3 months later they can see again with treatment, that’s MOGAD.” On the other hand, comorbidities such as lupus strongly favor NMO, he added. And another underrecognized, unique symptom of NMO is that about 10% of people may present with protracted episodes of nausea, vomiting, or hiccups, added Dr. Bennett. “What’s important is not that the neurologist recognize this per se in the emergency department, because they’re not going to be called for that patient — the GI doctors will be called for that patient. But when you’re seeing a patient who may have another presentation: a spinal cord attack, a vision attack with optic neuritis, and you ask them simply ‘have you ever had an episode of protracted nausea, vomiting, or hiccups?’ — I can’t tell you how many times I can have someone say ‘that’s weird I was just in the ED 3 months ago for that.’ And then, I know exactly what’s going on.”
Prevention of Relapse
Treatment of NMO presents some particular challenges for clinicians because the old treatment, rituximab, an anti-CD20 monoclonal antibody which has been used since 2005, has been so affordable and successful. “It’s hard to get people off,” said Dr. Levy. “It’s still the most commonly used drug for NMO in the US, even though it’s not approved. It’s cheap enough, and so people get started on that as a treatment, and then they just continue it, even as an outpatient.”
But since 2019, four new FDA-approved therapies have entered the scene with even better efficacy: the anti-CD-19 targeted medication inebilizumab (Uplizna, Viela Bio, approved in 2020), which requires two 90-minute infusions per year; the interleukin-6 (IL-6) receptor inhibitor satralizumab (Enspryng, Roche, approved in 2020), which is administered subcutaneously once monthly; and the anti-complement C5 inhibitors eculizumab (Soliris, Alexion Pharmaceuticals, approved in 2019), and ravulizumab (Ultomiris, Alexion Pharmaceuticals, approved in 2024), which require infusions every 2 weeks or every 2 months, respectively.
Both experts point to compelling clinical evidence to prescribe the Food and Drug Administration–approved drugs for newly diagnosed NMO, and to switch existing patients from rituximab to the new drugs. “The data is pretty clear that there’s about a 35% failure rate with rituximab, as opposed to less than 5% with the new drugs,” explained Dr. Levy. But ironically, where insurance companies used to balk at covering rituximab because it was not FDA approved for NMO, they are now balking at the FDA-approved options because of the cost. “Even in an academic center, where we get a discount on the drugs, the biosimilar generic of rituximab costs about $890 per dose,” he said. “So overall, it’s less than $4,000 a year for rituximab. Compare that with the most expensive FDA-approved option, which is eculizumab. That’s $715,000 per year. And then the other three drugs are below that, but none are less than about $290,000 a year.”
Patients are also hesitant to switch from rituximab if they’ve been well-controlled on it. “There’s a process to it, and I always talk my patients through it, but I would say less than half make the switch,” said Dr. Levy. “Most people want to stay. It’s a whole different schedule, and mixing two drugs. Are you going to overlap and overly immune suppress? Is the insurance going to approve it? It becomes more complicated.”
“Insurance companies are sometimes inappropriately pushing physicians, asking for patients to fail rituximab before they’ll approve an FDA-approved drug, which is like playing doctor when they’re not a licensed physician,” added Dr. Bennett. “And I think that is absolutely inappropriate, especially in light of the fact that before there were approved drugs, insurance companies used to deny rituximab because it was ‘experimental’ and ‘too expensive’ — and now it’s a cheaper alternative.”
Requiring failure on rituximab is also unethical, given the potential for irreversible damage, Dr. Levy pointed out. “With NMO we don’t tolerate a failure. That’s also how the trials of the new drugs were done. It was considered unethical to have an outcome of annualized relapse rate, like we used to in MS, where we say, OK if you have two attacks a year, then the drug has failed. With NMO, one failure, one breakthrough, and that drug is worthless. We switch.”
A Wealth of Treatment Choices
Patients opting for an FDA-approved treatment now have a “wildly effective” array of new drugs, said Dr. Levy, but choosing can be difficult when each has its own set of advantages and disadvantages. “I have equal numbers of patients on all the drugs, and I show all the data to my patients: efficacy, safety, logistics, cost, and then I ask ‘What are your priorities? Which of these things that I say really rings with you? Is it the infusion schedule? Is it the efficacy? Is it the safety concern? Is it the cost? What are you most concerned about?’ And then we start to have the conversation that way. It’s a shared decision-making process.”
There is definitely an art to finding the best fit for each patient, agreed Dr. Bennett, “both with the urgency of controlling the disease, the particular patient in front of you, their ability to adhere to certain therapies, their ability to have access to infusions, or to self-inject, or to get transported to an infusion center or have access to home infusion.”
Patient empowerment in the decision is very important, added Dr. Levy. “When people make the decision on their own, they’re much more likely to be compliant, rather than me telling them they have to do this. And that’s why I think we haven’t had a single relapse on the new drugs. There have been switches because of intolerance, and cost, and all those issues, but not because of a breakthrough attack.”
Future
Looking ahead in the field, Dr. Bennett sees the biggest potential for improvement is in the management of acute attacks, which he describes as “a major treatment gap.” Although plasma exchange is immediately effective in limiting the amount of circulating pathogenic AQP4-IgG “there are other approaches that could be even more beneficial,” he said. “A promising strategy is to use drugs that act immediately on arms of the immune response that are directly injuring brain tissue. These include serum complement and cells such as neutrophils and natural killer cells that release destructive enzymes and inflammatory mediators,” he explained. “Complement inhibitors, such as the C5 inhibitors eculizumab and ravulizumab, currently approved for NMOSD relapse prevention, act immediately to inhibit complement-mediated tissue injury. Similarly, high doses of antihistamines could be used to rapidly stop the release of the destructive enzyme elastase from neutrophils and natural killer cells, while elastase inhibitors could be given to minimize cell injury. Direct clinical studies are needed to find both the optimal treatment window and regimen.”
References
1. Hor JY et al. Epidemiology of Neuromyelitis Optica Spectrum Disorder and Its Prevalence and Incidence Worldwide. Front Neurol. 2020 Jun 26:11:501. doi: 10.3389/fneur.2020.00501.
2. Wingerchuk DM et al. International Consensus Diagnostic Criteria for Neuromyelitis Optica Spectrum Disorders. Neurology. 2015 Jul 14;85(2):177-89. doi: 10.1212/WNL.0000000000001729.
3. Mealy MA et al. Epidemiology of Neuromyelitis Optica in the United States: A Multicenter Analysis. Arch Neurol. 2012 Sep;69(9):1176-80. doi: 10.1001/archneurol.2012.314.
4. Contentti EC et al. Frequency of NMOSD Misdiagnosis in a Cohort From Latin America: Impact and Evaluation of Different Contributors. Mult Scler. 2023 Feb;29(2):277-286. doi: 10.1177/13524585221136259.
Urgency of treatment is something that many physicians may not fully appreciate when it comes to neuromyelitis optica (NMO), according to experts on this rare autoimmune demyelinating disorder. This may be partly due to its similar presentation to multiple sclerosis (MS), said Michael Levy, MD, PhD, associate professor, Harvard Medical School, research director, Division of Neuroimmunology & Neuroinfectious Disease, and director, Neuroimmunology Clinic and Research Laboratory, at Massachusetts General Hospital in Boston. But while the two conditions share many clinical characteristics, “immunologically, they are about as different as can be,” he warned.
The urgency of distinction is important because where MS is known to have a relatively gradual progression, NMO is now red-flagged to potentially cause rapid and irreversible damage. While the course of MS might be described as a slow burn, NMO should be treated like a wildfire.
“That message has gotten muddled, particularly because acute treatment in MS has never been shown to affect outcome,” said Jeffrey Bennett, MD, PhD, professor of neurology and ophthalmology at the University of Colorado School of Medicine, Aurora. In contrast, rapid diagnosis and treatment of NMO “means potentially preventing future devastating neurologic injury,” he said.
First described by Dr. Eugène Devic in 1894, and sometimes known as Devic’s disease, NMO is believed to have a prevalence that varies widely depending on ethnicity and gender. A recent report suggests a prevalence of approximately1/100,000 population among Whites with an annual incidence of less than 1/million in this population, while the prevalence is higher among East Asians (approximately 3.5/100,000), and may reach as high as 10/100,000 in Blacks.1 It has a high female-to-male ratio (up to 9:1) with a mean age of onset of about 40 years, although pediatric cases are described.
It has long been recognized that NMO lacks the “neurocerebritis” of MS, with inflammation predominant in the optic and spinal nerves, but it was not until 2004 that researchers at the Mayo Clinic identified serum aquaporin-4 immunoglobulin G antibodies (AQP4-IgG) that could reliably distinguish NMO from MS. In 2015, the international consensus diagnostic criteria for neuromyelitis optica2 cited core clinical characteristics required for patients with AQP4-IgG-positive NMO spectrum disorder (NMOSD) “including clinical syndromes or MRI findings related to optic nerve, spinal cord, area postrema, other brainstem, diencephalic, or cerebral presentations.” Rarely, NMO patients can be seronegative for AQP4-IgG, but are still considered to have NMOSD for which non-opticospinal clinical and MRI characteristics findings are described. MS patients testing negative for AQP4-IgG should also be tested for the related myelin oligodendrocyte glycoprotein antibody disease (MOGAD), which has a prevalence about four to five times greater than NMO, Dr. Bennett said.
Testing
Because both NMO and MOGAD can be identified by antibodies, they are less commonly misdiagnosed as MS compared to previously. But, prior to the identification of the AQP4-IgG antibody in 2004, the misdiagnosis rate of NMO was probably about 95% said Dr. Levy.
“Of course, before we had the antibody test or clinical criteria, we couldn’t confirm a diagnosis of NMO, so basically everyone had a diagnosis of MS, and after the antibody test became commercially available in 2005/2006, we could confirm the diagnosis, with our study in 2012 showing a much lower misdiagnosis rate of 30%.”3 More recently, the misdiagnosis rates are even lower, he added. A recent study out of Argentina found a rate of only 12%.4
The specificity and sensitivity of cell binding assay serum AQP4-IgG testing is roughly 99% and 90%, respectively, better than ELIZA testing (which has a sensitivity in the 60-65% range), said Dr. Bennett. “That’s why we highly emphasize to physicians, that if you have a suspicion for NMOSD you go to a cell binding assay, and make sure that where you’re sending the serum, the lab can do that procedure.” Still, because of the risk of false positives, he urges restraint in testing for the disorder in the absence of a high suspicion for it. “If you test a lot of people indiscriminately for a rare disorder, you get a lot of false positives because the actual true positives are a very small fraction of that group. So, even with a specificity of around 99% that means 1% of the people you test are falsely positive. And if you’re testing a group of people indiscriminately, then your true positives are less than 1% by far, so then most of the people that you pick up are not truly with disease.”
Acute Treatment
While misdiagnosis of NMO as MS is less common than previously, it is still a concern, not only because of the irreversible risks associated with delayed acute treatment, but also the risks of inappropriate preventive MS therapy, which could be harmful to patients with NMO.
Acute flare-ups of NMO and MOGAD are currently all treated with the same decades-old mainstays for acute MS — intravenous steroids and plasma exchange — but the approach is more aggressive. Retrospective studies have shown that, for NMO, plasma exchange has shown an increased likelihood of improvement versus steroids alone, said Dr. Bennett, but since time is of the essence, treatment should begin before a definitive diagnosis is confirmed.
“What’s limiting our patients is, number one, recognizing NMOSD when the attack is happening in your face. You’ve got to know, hey, this is NMOSD or I’m suspicious of NMOSD and hence, I need to treat it urgently because the outcome has a high probability of not being good. You’ve got to realize that this is NMOSD before the test comes back, because by the time it comes back positive in several days, you’re probably missing the optimal window to treat. The point is to know that the presentation in front of you, the MRI pictures in front of you, the laboratory tests that you might have done in terms of spinal fluid analysis, all highly suggest NMOSD. And so hence, I’m going to take the chance that I might be wrong, but I’m going to treat as if it is and wait for the test to come back.”
Realistically, the risks associated with this approach are minor compared with the potential benefit, Dr. Bennett said. “For plasma exchange, there’s the placement of the central line, and the complications that could happen from that. Plasma exchange can lead to metabolic ionic changes in the blood, fluid shifts that might have to be watched in the hospital setting.”
While waiting for diagnostic results, one clue that may emerge from acute treatment is recovery time. “The recovery from MOGAD attacks is really distinct,” said Dr. Levy. “They get better a lot faster. So, if they’re blind in the hospital, and 3 months later they can see again with treatment, that’s MOGAD.” On the other hand, comorbidities such as lupus strongly favor NMO, he added. And another underrecognized, unique symptom of NMO is that about 10% of people may present with protracted episodes of nausea, vomiting, or hiccups, added Dr. Bennett. “What’s important is not that the neurologist recognize this per se in the emergency department, because they’re not going to be called for that patient — the GI doctors will be called for that patient. But when you’re seeing a patient who may have another presentation: a spinal cord attack, a vision attack with optic neuritis, and you ask them simply ‘have you ever had an episode of protracted nausea, vomiting, or hiccups?’ — I can’t tell you how many times I can have someone say ‘that’s weird I was just in the ED 3 months ago for that.’ And then, I know exactly what’s going on.”
Prevention of Relapse
Treatment of NMO presents some particular challenges for clinicians because the old treatment, rituximab, an anti-CD20 monoclonal antibody which has been used since 2005, has been so affordable and successful. “It’s hard to get people off,” said Dr. Levy. “It’s still the most commonly used drug for NMO in the US, even though it’s not approved. It’s cheap enough, and so people get started on that as a treatment, and then they just continue it, even as an outpatient.”
But since 2019, four new FDA-approved therapies have entered the scene with even better efficacy: the anti-CD-19 targeted medication inebilizumab (Uplizna, Viela Bio, approved in 2020), which requires two 90-minute infusions per year; the interleukin-6 (IL-6) receptor inhibitor satralizumab (Enspryng, Roche, approved in 2020), which is administered subcutaneously once monthly; and the anti-complement C5 inhibitors eculizumab (Soliris, Alexion Pharmaceuticals, approved in 2019), and ravulizumab (Ultomiris, Alexion Pharmaceuticals, approved in 2024), which require infusions every 2 weeks or every 2 months, respectively.
Both experts point to compelling clinical evidence to prescribe the Food and Drug Administration–approved drugs for newly diagnosed NMO, and to switch existing patients from rituximab to the new drugs. “The data is pretty clear that there’s about a 35% failure rate with rituximab, as opposed to less than 5% with the new drugs,” explained Dr. Levy. But ironically, where insurance companies used to balk at covering rituximab because it was not FDA approved for NMO, they are now balking at the FDA-approved options because of the cost. “Even in an academic center, where we get a discount on the drugs, the biosimilar generic of rituximab costs about $890 per dose,” he said. “So overall, it’s less than $4,000 a year for rituximab. Compare that with the most expensive FDA-approved option, which is eculizumab. That’s $715,000 per year. And then the other three drugs are below that, but none are less than about $290,000 a year.”
Patients are also hesitant to switch from rituximab if they’ve been well-controlled on it. “There’s a process to it, and I always talk my patients through it, but I would say less than half make the switch,” said Dr. Levy. “Most people want to stay. It’s a whole different schedule, and mixing two drugs. Are you going to overlap and overly immune suppress? Is the insurance going to approve it? It becomes more complicated.”
“Insurance companies are sometimes inappropriately pushing physicians, asking for patients to fail rituximab before they’ll approve an FDA-approved drug, which is like playing doctor when they’re not a licensed physician,” added Dr. Bennett. “And I think that is absolutely inappropriate, especially in light of the fact that before there were approved drugs, insurance companies used to deny rituximab because it was ‘experimental’ and ‘too expensive’ — and now it’s a cheaper alternative.”
Requiring failure on rituximab is also unethical, given the potential for irreversible damage, Dr. Levy pointed out. “With NMO we don’t tolerate a failure. That’s also how the trials of the new drugs were done. It was considered unethical to have an outcome of annualized relapse rate, like we used to in MS, where we say, OK if you have two attacks a year, then the drug has failed. With NMO, one failure, one breakthrough, and that drug is worthless. We switch.”
A Wealth of Treatment Choices
Patients opting for an FDA-approved treatment now have a “wildly effective” array of new drugs, said Dr. Levy, but choosing can be difficult when each has its own set of advantages and disadvantages. “I have equal numbers of patients on all the drugs, and I show all the data to my patients: efficacy, safety, logistics, cost, and then I ask ‘What are your priorities? Which of these things that I say really rings with you? Is it the infusion schedule? Is it the efficacy? Is it the safety concern? Is it the cost? What are you most concerned about?’ And then we start to have the conversation that way. It’s a shared decision-making process.”
There is definitely an art to finding the best fit for each patient, agreed Dr. Bennett, “both with the urgency of controlling the disease, the particular patient in front of you, their ability to adhere to certain therapies, their ability to have access to infusions, or to self-inject, or to get transported to an infusion center or have access to home infusion.”
Patient empowerment in the decision is very important, added Dr. Levy. “When people make the decision on their own, they’re much more likely to be compliant, rather than me telling them they have to do this. And that’s why I think we haven’t had a single relapse on the new drugs. There have been switches because of intolerance, and cost, and all those issues, but not because of a breakthrough attack.”
Future
Looking ahead in the field, Dr. Bennett sees the biggest potential for improvement is in the management of acute attacks, which he describes as “a major treatment gap.” Although plasma exchange is immediately effective in limiting the amount of circulating pathogenic AQP4-IgG “there are other approaches that could be even more beneficial,” he said. “A promising strategy is to use drugs that act immediately on arms of the immune response that are directly injuring brain tissue. These include serum complement and cells such as neutrophils and natural killer cells that release destructive enzymes and inflammatory mediators,” he explained. “Complement inhibitors, such as the C5 inhibitors eculizumab and ravulizumab, currently approved for NMOSD relapse prevention, act immediately to inhibit complement-mediated tissue injury. Similarly, high doses of antihistamines could be used to rapidly stop the release of the destructive enzyme elastase from neutrophils and natural killer cells, while elastase inhibitors could be given to minimize cell injury. Direct clinical studies are needed to find both the optimal treatment window and regimen.”
References
1. Hor JY et al. Epidemiology of Neuromyelitis Optica Spectrum Disorder and Its Prevalence and Incidence Worldwide. Front Neurol. 2020 Jun 26:11:501. doi: 10.3389/fneur.2020.00501.
2. Wingerchuk DM et al. International Consensus Diagnostic Criteria for Neuromyelitis Optica Spectrum Disorders. Neurology. 2015 Jul 14;85(2):177-89. doi: 10.1212/WNL.0000000000001729.
3. Mealy MA et al. Epidemiology of Neuromyelitis Optica in the United States: A Multicenter Analysis. Arch Neurol. 2012 Sep;69(9):1176-80. doi: 10.1001/archneurol.2012.314.
4. Contentti EC et al. Frequency of NMOSD Misdiagnosis in a Cohort From Latin America: Impact and Evaluation of Different Contributors. Mult Scler. 2023 Feb;29(2):277-286. doi: 10.1177/13524585221136259.
Urgency of treatment is something that many physicians may not fully appreciate when it comes to neuromyelitis optica (NMO), according to experts on this rare autoimmune demyelinating disorder. This may be partly due to its similar presentation to multiple sclerosis (MS), said Michael Levy, MD, PhD, associate professor, Harvard Medical School, research director, Division of Neuroimmunology & Neuroinfectious Disease, and director, Neuroimmunology Clinic and Research Laboratory, at Massachusetts General Hospital in Boston. But while the two conditions share many clinical characteristics, “immunologically, they are about as different as can be,” he warned.
The urgency of distinction is important because where MS is known to have a relatively gradual progression, NMO is now red-flagged to potentially cause rapid and irreversible damage. While the course of MS might be described as a slow burn, NMO should be treated like a wildfire.
“That message has gotten muddled, particularly because acute treatment in MS has never been shown to affect outcome,” said Jeffrey Bennett, MD, PhD, professor of neurology and ophthalmology at the University of Colorado School of Medicine, Aurora. In contrast, rapid diagnosis and treatment of NMO “means potentially preventing future devastating neurologic injury,” he said.
First described by Dr. Eugène Devic in 1894, and sometimes known as Devic’s disease, NMO is believed to have a prevalence that varies widely depending on ethnicity and gender. A recent report suggests a prevalence of approximately1/100,000 population among Whites with an annual incidence of less than 1/million in this population, while the prevalence is higher among East Asians (approximately 3.5/100,000), and may reach as high as 10/100,000 in Blacks.1 It has a high female-to-male ratio (up to 9:1) with a mean age of onset of about 40 years, although pediatric cases are described.
It has long been recognized that NMO lacks the “neurocerebritis” of MS, with inflammation predominant in the optic and spinal nerves, but it was not until 2004 that researchers at the Mayo Clinic identified serum aquaporin-4 immunoglobulin G antibodies (AQP4-IgG) that could reliably distinguish NMO from MS. In 2015, the international consensus diagnostic criteria for neuromyelitis optica2 cited core clinical characteristics required for patients with AQP4-IgG-positive NMO spectrum disorder (NMOSD) “including clinical syndromes or MRI findings related to optic nerve, spinal cord, area postrema, other brainstem, diencephalic, or cerebral presentations.” Rarely, NMO patients can be seronegative for AQP4-IgG, but are still considered to have NMOSD for which non-opticospinal clinical and MRI characteristics findings are described. MS patients testing negative for AQP4-IgG should also be tested for the related myelin oligodendrocyte glycoprotein antibody disease (MOGAD), which has a prevalence about four to five times greater than NMO, Dr. Bennett said.
Testing
Because both NMO and MOGAD can be identified by antibodies, they are less commonly misdiagnosed as MS compared to previously. But, prior to the identification of the AQP4-IgG antibody in 2004, the misdiagnosis rate of NMO was probably about 95% said Dr. Levy.
“Of course, before we had the antibody test or clinical criteria, we couldn’t confirm a diagnosis of NMO, so basically everyone had a diagnosis of MS, and after the antibody test became commercially available in 2005/2006, we could confirm the diagnosis, with our study in 2012 showing a much lower misdiagnosis rate of 30%.”3 More recently, the misdiagnosis rates are even lower, he added. A recent study out of Argentina found a rate of only 12%.4
The specificity and sensitivity of cell binding assay serum AQP4-IgG testing is roughly 99% and 90%, respectively, better than ELIZA testing (which has a sensitivity in the 60-65% range), said Dr. Bennett. “That’s why we highly emphasize to physicians, that if you have a suspicion for NMOSD you go to a cell binding assay, and make sure that where you’re sending the serum, the lab can do that procedure.” Still, because of the risk of false positives, he urges restraint in testing for the disorder in the absence of a high suspicion for it. “If you test a lot of people indiscriminately for a rare disorder, you get a lot of false positives because the actual true positives are a very small fraction of that group. So, even with a specificity of around 99% that means 1% of the people you test are falsely positive. And if you’re testing a group of people indiscriminately, then your true positives are less than 1% by far, so then most of the people that you pick up are not truly with disease.”
Acute Treatment
While misdiagnosis of NMO as MS is less common than previously, it is still a concern, not only because of the irreversible risks associated with delayed acute treatment, but also the risks of inappropriate preventive MS therapy, which could be harmful to patients with NMO.
Acute flare-ups of NMO and MOGAD are currently all treated with the same decades-old mainstays for acute MS — intravenous steroids and plasma exchange — but the approach is more aggressive. Retrospective studies have shown that, for NMO, plasma exchange has shown an increased likelihood of improvement versus steroids alone, said Dr. Bennett, but since time is of the essence, treatment should begin before a definitive diagnosis is confirmed.
“What’s limiting our patients is, number one, recognizing NMOSD when the attack is happening in your face. You’ve got to know, hey, this is NMOSD or I’m suspicious of NMOSD and hence, I need to treat it urgently because the outcome has a high probability of not being good. You’ve got to realize that this is NMOSD before the test comes back, because by the time it comes back positive in several days, you’re probably missing the optimal window to treat. The point is to know that the presentation in front of you, the MRI pictures in front of you, the laboratory tests that you might have done in terms of spinal fluid analysis, all highly suggest NMOSD. And so hence, I’m going to take the chance that I might be wrong, but I’m going to treat as if it is and wait for the test to come back.”
Realistically, the risks associated with this approach are minor compared with the potential benefit, Dr. Bennett said. “For plasma exchange, there’s the placement of the central line, and the complications that could happen from that. Plasma exchange can lead to metabolic ionic changes in the blood, fluid shifts that might have to be watched in the hospital setting.”
While waiting for diagnostic results, one clue that may emerge from acute treatment is recovery time. “The recovery from MOGAD attacks is really distinct,” said Dr. Levy. “They get better a lot faster. So, if they’re blind in the hospital, and 3 months later they can see again with treatment, that’s MOGAD.” On the other hand, comorbidities such as lupus strongly favor NMO, he added. And another underrecognized, unique symptom of NMO is that about 10% of people may present with protracted episodes of nausea, vomiting, or hiccups, added Dr. Bennett. “What’s important is not that the neurologist recognize this per se in the emergency department, because they’re not going to be called for that patient — the GI doctors will be called for that patient. But when you’re seeing a patient who may have another presentation: a spinal cord attack, a vision attack with optic neuritis, and you ask them simply ‘have you ever had an episode of protracted nausea, vomiting, or hiccups?’ — I can’t tell you how many times I can have someone say ‘that’s weird I was just in the ED 3 months ago for that.’ And then, I know exactly what’s going on.”
Prevention of Relapse
Treatment of NMO presents some particular challenges for clinicians because the old treatment, rituximab, an anti-CD20 monoclonal antibody which has been used since 2005, has been so affordable and successful. “It’s hard to get people off,” said Dr. Levy. “It’s still the most commonly used drug for NMO in the US, even though it’s not approved. It’s cheap enough, and so people get started on that as a treatment, and then they just continue it, even as an outpatient.”
But since 2019, four new FDA-approved therapies have entered the scene with even better efficacy: the anti-CD-19 targeted medication inebilizumab (Uplizna, Viela Bio, approved in 2020), which requires two 90-minute infusions per year; the interleukin-6 (IL-6) receptor inhibitor satralizumab (Enspryng, Roche, approved in 2020), which is administered subcutaneously once monthly; and the anti-complement C5 inhibitors eculizumab (Soliris, Alexion Pharmaceuticals, approved in 2019), and ravulizumab (Ultomiris, Alexion Pharmaceuticals, approved in 2024), which require infusions every 2 weeks or every 2 months, respectively.
Both experts point to compelling clinical evidence to prescribe the Food and Drug Administration–approved drugs for newly diagnosed NMO, and to switch existing patients from rituximab to the new drugs. “The data is pretty clear that there’s about a 35% failure rate with rituximab, as opposed to less than 5% with the new drugs,” explained Dr. Levy. But ironically, where insurance companies used to balk at covering rituximab because it was not FDA approved for NMO, they are now balking at the FDA-approved options because of the cost. “Even in an academic center, where we get a discount on the drugs, the biosimilar generic of rituximab costs about $890 per dose,” he said. “So overall, it’s less than $4,000 a year for rituximab. Compare that with the most expensive FDA-approved option, which is eculizumab. That’s $715,000 per year. And then the other three drugs are below that, but none are less than about $290,000 a year.”
Patients are also hesitant to switch from rituximab if they’ve been well-controlled on it. “There’s a process to it, and I always talk my patients through it, but I would say less than half make the switch,” said Dr. Levy. “Most people want to stay. It’s a whole different schedule, and mixing two drugs. Are you going to overlap and overly immune suppress? Is the insurance going to approve it? It becomes more complicated.”
“Insurance companies are sometimes inappropriately pushing physicians, asking for patients to fail rituximab before they’ll approve an FDA-approved drug, which is like playing doctor when they’re not a licensed physician,” added Dr. Bennett. “And I think that is absolutely inappropriate, especially in light of the fact that before there were approved drugs, insurance companies used to deny rituximab because it was ‘experimental’ and ‘too expensive’ — and now it’s a cheaper alternative.”
Requiring failure on rituximab is also unethical, given the potential for irreversible damage, Dr. Levy pointed out. “With NMO we don’t tolerate a failure. That’s also how the trials of the new drugs were done. It was considered unethical to have an outcome of annualized relapse rate, like we used to in MS, where we say, OK if you have two attacks a year, then the drug has failed. With NMO, one failure, one breakthrough, and that drug is worthless. We switch.”
A Wealth of Treatment Choices
Patients opting for an FDA-approved treatment now have a “wildly effective” array of new drugs, said Dr. Levy, but choosing can be difficult when each has its own set of advantages and disadvantages. “I have equal numbers of patients on all the drugs, and I show all the data to my patients: efficacy, safety, logistics, cost, and then I ask ‘What are your priorities? Which of these things that I say really rings with you? Is it the infusion schedule? Is it the efficacy? Is it the safety concern? Is it the cost? What are you most concerned about?’ And then we start to have the conversation that way. It’s a shared decision-making process.”
There is definitely an art to finding the best fit for each patient, agreed Dr. Bennett, “both with the urgency of controlling the disease, the particular patient in front of you, their ability to adhere to certain therapies, their ability to have access to infusions, or to self-inject, or to get transported to an infusion center or have access to home infusion.”
Patient empowerment in the decision is very important, added Dr. Levy. “When people make the decision on their own, they’re much more likely to be compliant, rather than me telling them they have to do this. And that’s why I think we haven’t had a single relapse on the new drugs. There have been switches because of intolerance, and cost, and all those issues, but not because of a breakthrough attack.”
Future
Looking ahead in the field, Dr. Bennett sees the biggest potential for improvement is in the management of acute attacks, which he describes as “a major treatment gap.” Although plasma exchange is immediately effective in limiting the amount of circulating pathogenic AQP4-IgG “there are other approaches that could be even more beneficial,” he said. “A promising strategy is to use drugs that act immediately on arms of the immune response that are directly injuring brain tissue. These include serum complement and cells such as neutrophils and natural killer cells that release destructive enzymes and inflammatory mediators,” he explained. “Complement inhibitors, such as the C5 inhibitors eculizumab and ravulizumab, currently approved for NMOSD relapse prevention, act immediately to inhibit complement-mediated tissue injury. Similarly, high doses of antihistamines could be used to rapidly stop the release of the destructive enzyme elastase from neutrophils and natural killer cells, while elastase inhibitors could be given to minimize cell injury. Direct clinical studies are needed to find both the optimal treatment window and regimen.”
References
1. Hor JY et al. Epidemiology of Neuromyelitis Optica Spectrum Disorder and Its Prevalence and Incidence Worldwide. Front Neurol. 2020 Jun 26:11:501. doi: 10.3389/fneur.2020.00501.
2. Wingerchuk DM et al. International Consensus Diagnostic Criteria for Neuromyelitis Optica Spectrum Disorders. Neurology. 2015 Jul 14;85(2):177-89. doi: 10.1212/WNL.0000000000001729.
3. Mealy MA et al. Epidemiology of Neuromyelitis Optica in the United States: A Multicenter Analysis. Arch Neurol. 2012 Sep;69(9):1176-80. doi: 10.1001/archneurol.2012.314.
4. Contentti EC et al. Frequency of NMOSD Misdiagnosis in a Cohort From Latin America: Impact and Evaluation of Different Contributors. Mult Scler. 2023 Feb;29(2):277-286. doi: 10.1177/13524585221136259.