Background: We reviewed our autoimmune
encephalitis neural antibody testing using brain tissue indirect
immunofluorescence (TIIF) and cell-based assays (CBAs) after one year.
Methods: Samples were tested from March
2019–March 2020 by TIIF and CBA for anti-NMDAR, LGI1, CASPR2, AMPAR,
GABA(B)R, DPPX, IgLON5 and GAD65. Weakly positive or positive CBA, with or
without corresponding TIIF positivity, was reported positive. Clinical
questionnaires were submitted for clinical-serological correlation. Patients
with a compatible clinical phenotype and no more likely alternative
diagnosis were classified as true-positives, while all others were flagged
as possible false-positives. Results: Twenty of 373
patients (5.4%) had a positive neural antibody. All anti-LGI1 (N=4), GAD65
(N=4), and GABA(B)R (N=1) were classified as true-positives. In contrast,
only 3/6 anti-CASPR2 and 3/5 anti-NMDAR were classified as true-positives.
Among true-positives, 2/4 anti-LGI and 3/3 anti-CASPR2 were positive by CBA
only. All possible false-positive results exhibited only weak serum staining
by CBA, with negative serum TIIF and negative CSF CBA/TIIF (if available).
Conclusions: Clinical sensitivity of CBA seems
higher than TIIF for neural antibodies studied herein, but may come at some
expense to clinical specificity. Among patients with weak serum staining by
CBA, correlation with serum TIIF, CSF CBA/TIIF, and clinical presentation is
recommended.