The detection of aquaporin-4 (AQP4) antibodies in neuromyelitis optica (NMO) led to a breakthrough in diagnosing NMO. To date, different assays to detect these antibodies are available. Sensitivities of these assays differ between 58-80%, resulting in a substantial amount of false-negative results. Multiple factors influence the test results including the assay technique, the moment the serum sample was obtained (e.g. at clinical activity or under treatment) and the type of disease (recurrent or monophasic). We recently developed two anti-AQP4 assays: a fluorescence immunoprecipitation assay (FIPA) and a cellbased flowcytometric assay (CBA). Both assays showed high specificity (> 99%) and sensitivity between 56% (to detect all patients with NMO) and 74% (to detect only the patients with recurrent NMO). Corresponding results were found in both assays, except in one patient: This woman presented at the age of 31 years with clinical symptoms corresponding to an elongated spinal cord lesion, followed by blindness of one eye, fulfilling the diagnostic criteria for NMO. She experienced multiple relapses with spinal cord and optic nerve symptoms. Her illness was severe with serious decline, failing different treatments including prednisone, azathioprine, plasma exchange, mitoxantrone, rituximab and immunoglobulin intravenously. She is now clinically stable using mycophenolate mofetil. Eight serum samples were obtained during her disease course at different time points and tested on AQP4 antibodies. Irrespective of disease activity or treatment, we consistently found that all of these samples were evidently positive in the CBA (with high serum titres) and negative in the FIPA. No correlation between anti-AQP4 titers and disease activity was observed. Additional testing on glutamate transporter EAAT2 was negative in all samples, ruling out that in this patient antibodies bind to the co-expressed EAAT2 rather than AQP4. A negative assay does not necessarily exclude a diagnosis of NMO. Although different reasons for falsenegative results exist, our observation indicates that in this particular NMO patient an epitope is present that is only recognised on the intact membrane-expressed molecule and that is not present in the immunoprecipitation assay. This can be a variant of the disease previously described in AQP4 positive NMO patients. An alternative possibility is that there are still other NMO associated auto-antibodies to be identified besides anti-AQP4.
Read More: Antibodies against aquaporin-4 in a patient with neuromyelitis optica: positive or not?