Testing for antibodies to human aquaporin-4 by ELISA: Sensitivity, specificity, and direct comparison with immunohistochemistry

Abstract Background Several assays have been developed to detect antibodies to aquaporin-4 (NMO-IgG/AQP4-Ab). However, many of these assays require sophisticated techniques and are thus only available at specialized laboratories. This is problematic since NMO-IgG/AQP4-Ab testing has important prognostic and therapeutic implications. Objective To evaluate a newly developed, commercial, enzyme-linked immunosorbent assay (ELISA) for detecting NMO-IgG/AQP4-Ab. Methods Serum samples from 261 patients with NMO spectrum disorders (NMOSD; n = 108) and controls (n = 153) were tested for AQP4-Ab by using ELISA. Of these patients, 207 were tested in parallel using a standard immunohistochemical (IHC) assay. Results Fifty of 66 (75.8%) patients with NMO, 17/25 (68%) with LETM, 3/14 (21.4%) with ON, 2/3 (66.7%) with ON and non-extensive transverse myelitis, and 2/153 (1.3%) controls tested positive in the ELISA. Of those NMOSD patients tested by both ELISA and IHC, 10 were positive only in the ELISA and 3 exclusively in the IHC assay, suggesting that the overall sensitivity of the ELISA was higher than that of the standard IHC assay. The ELISA yielded very good intra- and inter-run reproducibility with regard to AQP4-Ab detection and good intrarun, but only moderate inter-run reproducibility with regard to AQP4-Ab quantification. Anti-AQP4 serum concentrations correlated with disease activity (p < 0.00001), but did not differ between patients with NMO and patients with isolated LETM or ON. Conclusion The ELISA evaluated here provides a relatively sensitive and easy-to-use diagnostic tool for detecting antibodies to AQP4 and could make AQP4-Ab testing, which is of high clinical relevance, more widely available.