ARTICLE IN BRIEF
4 Investigators have developed a test to quantify antibodies that define anti-N-methyl-D-aspartic acid receptor encephalitis.
CHICAGO —Anti-N-methyl-D-aspartic acid receptor (NMDAR) encephalitis is not rare and is probably under-diagnosed, according to Josep Dalmau, MD, PhD, who identified and coined the name of the disorder.
In a platform session here at the AAN annual meeting in April, Dr. Dalmau, professor of neurology at the University of Pennsylvania in Philadelphia, described the symptoms of 90 patients with anti-NMDAR encephalitis and the fruits of his NIH-funded research in identifying the epitope-region of the NMDA receptor where the antibodies bind and trigger the immune disorder.
Anti-NMDAR encephalitis is of interest to several specialties, Dr. Dalmau explained, including psychiatrists, neurologists, and oncologists. Symptoms typically include headache and fever, usually in a young person; and within a few days, bizarre personality changes emerge. The condition can be misdiagnosed as acute psychosis or drug abuse, he noted.
The clinical symptoms worsen, if untreated, and patients experience reduced levels of consciousness and catatonia-like symptoms. About 68 percent require ventilation, about 90 percent have dyskinesia, and autonomic instability occurs in 70 percent. Oncologists are consulted if a tumor is found.
Tumors, usually ovarian teratoma, are present in about 60 percent of cases, he said. People with anti-NMDAR encephalitis (with or without a tumor) tend to be young. In the series of 90 patients he recruited for his latest study, the youngest was age five and the oldest 76; the median age was 26.
BIRTH OF A DIAGNOSIS
In an interview, Dr. Dalmau explained how he came to identify this disorder. About five years ago, he treated a young woman with encephalitis in the ICU. CSF analysis suggested an inflammatory or autoimmune process. After months of treatment and with supportive care, she recovered fully. Dr. Dalmau kept samples of her CSF and serum.
A few months later, he encountered another patient with similar clinical features, and then a few years later he found another four cases. In 2007, he published a study in the Annals of Neurology on 12 cases of anti-NMDAR encephalitis. He and his colleagues found the anti-NMDAR antibodies in all 12 cases.
“Finding the specific subunit and epitope-region was important,” he explained. “Several conditions had been related to NMDAR antibodies, including lupus, stroke, viral encephalitis, and chronic epilepsy. Our goal was to identify the specific region [of the receptor] where antibodies responsible for anti-NMDAR encephalitis bind,” he explained.
The test, he said, is highly sensitive and specific. There have been no false positives. It includes three different techniques that exclude the presence of other antibodies, and one of them is specific for antibodies to NMDAR. It involves using human embryonic kidney (HEK) cells that have been transfected with NR1 and NR2 subunits of the NMDA receptor. When these cells are not transfected, they do not react with the NMDAR antibodies; once transfected, the HEK cells react to the antibodies and the diagnosis is established.
Routine neurological testing for these symptoms includes MRI and CSF analysis, but MRI is not diagnostic, Dr. Dalmau noted. CSF is abnormal in 95 percent of cases, with lymphocytic pleiocystosis in 91 percent. Dr. Dalmau has now identified 103 patients with anti-NMDAR encephalitis.
“As a result of our study, we know that anti-NMDAR encephalitis manifests with certain characteristics and a predictable clinical picture. This has contributed to recognition of this disorder,” Dr. Dalmau explained. “Every time I give a talk at a university or an institution, the number of patients identified increases. This leads me to believe it is under-diagnosed and under-recognized.”
Figure. DR. JOSEP DA...Image Tools
RESULTS OF SERIES
Dr. Dalmau described the series of 90 patients: 81 were women; 58 percent had a tumor; 62 percent of the women (n=47) had a teratoma, almost always in the ovary; and 22 percent of the men (n=2) had tumors — one testicular tumor and one small cell lung cancer.
Patients with tumors are easier to treat, Dr. Dalmau said. After surgical removal of the tumor, immunotherapy is instituted, most often with corticosteroids, IVIg, and plasma exchange. If patients do not respond well to these measures, rituximab and cyclophosphamide can be used, he said. If no tumor is present, then the same immunotherapy is used.
Dr. Dalmau explained that those with no tumor generally have a longer and more up-and-down recovery. Early identification of tumor and early initiation of treatment are associated with improved outcomes.
Of 87 patients whose outcome was assessed, 54 (62 percent) had a full recovery or only mild residual deficits, 17 (19.5 percent) are recovering, 10 (11.5 percent) had severe deficits, and six (7 percent) died. These six patients were diagnosed retrospectively using archived serum or CSF after the disorder was discovered. Thirteen patients had a relapse (six with tumor and seven without tumor). Early diagnosis and intervention were associated with fewer relapses. Dr. Dalmau declined to give other numbers and percentages because the study is being submitted for publication.
“We are working on commercializing the diagnostic test,” he said. “This usually requires a test that allows antibody quantitation, such as ELISA or immunoprecipitation, but as occurs with other antibodies, the tests can be less sensitive or specific. We have developed an ELISA to quantify the NMDAR antibodies, and the technique has been optimized to the point that the specificity is close to what would be required for clinical care,” he explained.
MORE ANTIBODY DISCOVERIES TO COME
Dr. Dalmau's discoveries, and other antibodies that have been identified in patients with neurologic disorders, are important, “because if you detect antibodies in a patient's serum or CSF that target cell surface proteins like receptors or ion channels, it suggests that there may be a tumor and that the patient may respond to immunosuppressive therapy,” explained Angela Vincent, MD, professor of neuroimmunology and head of the department of neurology at Oxford University in the UK. Immunosuppression is not a cure, she explained, but it will keep these patients in remission, and the level of antibodies may begin to fall if a tumor is removed.
“We found antibodies to the potassium channels in a treatment-responsive form of limbic encephalitis, usually without a tumor, and antibodies to aquaporin-4 have been identified in neuromyelitis optica by Dr. Vanna Lennon and the Mayo group,” Dr. Vincent said. “The NMDA antibody is also proving to be important clinically. It seems likely that more antibodies will be found.”
Testing for the NMDA antibody is available at a few specialized academic centers, including the University of Pennsylvania and Oxford University. Dr. Dalmau is currently in discussions about commercializing the test for NMDA antibodies. Dr. Vincent said the process of getting a test to market is cumbersome and it will probably take a long time.
ANTIBODIES TO NR1/NR2 HEROMERS OF NMDA RECEPTOR
In an earlier study in the Feb. 12 Neurology, Dr. Josep Dalmau reported anti-NMDA receptor encephalitis in four Japanese women.
Panels A and C show nonpermeabilized cultures of hippocampal neurons incubated with sera of two patients at the onset of symptoms. Panels B and D are similar cultures incubated with sera from the same patients obtained at follow-up seven years, three months, and six years later.
Sera obtained at symptom presentation show intense immunolabeling of the cell surface of neurons and neuronal processes, whereas sera obtained at follow-up are not reactive. Insets for Panels A and C correspond to sera from the same patients incubated with HEK293 cells expressing NR1/NR2B heromers of the NMDA receptor. Sera from the time of symptom presentation (insets: A and C), but not from follow-up (insets: B and D) react with cells expressing NR1/NR2B heromers (green cell membrane and cytoplasmic staining). Sera from controls did not show any reactivity with neurons of NR1/NR2B expression cells (not shown).
Source: Neurology 2008;70-504-511.