Letter on “Real‐World Clinical Experience With Serum <scp>MOG</scp> and <scp>AQP4</scp> Antibody Testing by Live Versus Fixed Cell‐Based Assay”

We read with interest the important paper of Said et al. which compared the sensitivity and specificity of live cell (LCBA) and fixed cell-based assays (FCBA) for myelin oligodendrocyte antibody (MOG-IgG) in serum [1]. The authors used the Euroimmun FCBA available in their in-house laboratory. The commercially available fixed-cell assay widely used in the ‘real world’ setting (ARUP Laboratories, Salt Lake City, UT) utilizes the same Euroimmun platform; however, the test performance may differ from in-house laboratory based on the degree of experience and training of laboratory raters and investigators. We therefore undertook a comparison of FCBA and LCBA for MOG Antibody using exclusively clinically available results without recourse to research-based protocols. We identified all patients at NYU Multiple Sclerosis Care Center (New York City, New York) who were clinically tested for MOG-IgG by both Mayo (Mayo Laboratories, Rochester, MN) and ARUP between Jan 2023 and Dec 2024 (n = 200), but limited our analyses to patients with paired blood samples that were collected on the same day (n = 80). Of the same-day paired samples, 66 were negative by both assays (and therefore did not meet the current criteria for Myelin Oligodendrocyte Glycoprotein Antibody-Associated Disease [2]), while 14 were positive by Mayo only (n = 7), ARUP only (n = 2) or both (n = 5) assays. The charts of patients who were seropositive by at least one assay were reviewed by a neuroimmunologist who was blinded to antibody status to determine if they met criteria for one or more MOGAD-compatible clinical syndromes [2], and alternative diagnoses were appropriately excluded. Of the 14 seropositive patients, 9 met criteria for MOGAD [2], while the other 5 had alternative diagnoses (see Table 1 for clinical details on all seropositive patients). Mayo LCBA correctly identified all 9 MOGAD patients but also yielded 3 false positives, while ARUP FCBA correctly identified 4 out of 9 MOGAD patients and yielded 3 false positives. Thus, sensitivity and specificity of these two assays were: 40% sensitivity and 95.7% specificity for ARUP's FCBA, and 100% sensitivity and 95.7% specificity for Mayo's LCBA. Our results closely match the findings of Said et al., in which sensitivity of in-house FCBA was 45.7%, while specificity was close to 100% [1], as well as those of Tea et al., which yielded FCBA sensitivity of 50.7% in children and 52.3% in adults, with specificity at 100% [3]. In another study that directly compared commercial Euroimmun FCBA with LCBA, the sensitivity of FCBA was even lower −25% [4]. Taken together, these data indicate that FCBAs will miss half or more of MOGAD diagnoses. It should also be noted that the high level of agreement in a fixed vs. live cell-based assay for antibodies to myelin oligodendrocyte glycoprotein observed by Smith et al. [5] does not contradict the above results, as Smith et al. did not have access to clinical data and therefore were not able to calculate sensitivity and specificity for FCBA as we did. We conclude that all patients with clinical syndrome suggestive of MOGAD should be tested for MOG-IgG with LCBAs only; if FCBA is used and returns a negative result, re-testing with LCBA is appropriate. R.V., A.J.J. and I.K. collected and analyzed the data and contributed to the writing and editing of the manuscript. I.K. supervised the project. Raveena Vij and Dr. Alexander J. Jonokuchi have no conflicts of interest to disclose. Dr. Ilya Kister served on advisory boards for Biogen, Genentech, Horizon, and Alexion Pharmaceuticals, and received consulting fees from Roche and research support for investigator-initiated grants from Genentech and Novartis. The data that support the findings of this study are available from the corresponding author upon reasonable request.