The Positive Predictive Value of Aquaporin-4 Antibody Live Cell-based Assay in a Tertiary Referral Center
Mayra Montalvo1, Sean Pittock1, Elia Sechi1, James Fryer1, Andrew McKeon1, John Mills1, John Chen1, Eoin P. Flanagan1
1Mayo Clinic

We sought to assess the false positivity rate and positive predictive value of aquaporin-4 antibody (AQP4-IgG) using live cell-based flow cytometry assay for a diagnosis of neuromyelitis optica spectrum disorder (NMOSD) at a tertiary referral center.


AQP4-IgG is a serum diagnostic biomarker of neuromyelitis optica spectrum disorder (NMOSD). Antibody detection methods have improved but false positive rates of up to 5% are reported with some assays (e.g., enzyme-linked immunosorbent assay).


We searched our neuroimmunology laboratory database (1/1/2018-12/31/2019) and identified consecutive Mayo Clinic patients that tested positive for AQP4 IgG over the two year study period. We used a live cell-based fluorescent activated cell sorting (FACS) using the M1 isoform of aquaporin-4.  IgG-binding-index ≥2 was the cut-off during screening and an end titration ≥5 confirmed a positive result. Clinical records were reviewed independently by two neurologists to determine if positives met diagnostic criteria for NMOSD with AQP4-IgG.


Of 1704 Mayo Clinic patients consecutively tested for AQP4-IgG, 37 (2%) were positive. All 37 fulfilled 2015 NMOSD diagnostic criteria for NMOSD with AQP4-IgG with 100% agreement between two neurologists evaluating the medical records. Thus, no false positives occurred. The positive predictive value and specificity of AQP4-IgG for NMOSD were 100%. Among the AQP4-IgG positive samples, the median IgG-binding-index was 11.7 (range, 2.1-71.7) and median end titer was 1000 (range, 5-10000). The median age of AQP4-IgG seropositive patients (81% female) was 56 years (range, 4-90). AQP4-IgG NMOSD included one or more attacks of: transverse myelitis, 28; optic neuritis, 14; or area postrema syndrome, 8.


AQP4-IgG, assessed by live cell-based FACS assay, is a highly specific diagnostic biomarker of NMOSD and did not yield false positive results in this study despite a large testing volume.