Limitations of a Commercial Assay as Diagnostic Test of Autoimmune Encephalitis

Abstract

Detection of neuronal surface antibodies (NSAb) is important for the diagnosis of autoimmune encephalitis (AE). Although most clinical laboratories use a commercial diagnostic kit (Euroimmun, Lübeck, Germany) based on indirect immunofluorescence on transfected cells (IIFA), clinical experience suggests diagnostic limitations. Here, we assessed the performance of the commercial IIFA in serum and CSF samples of patients with suspected AE previously examined by rat brain immunohistochemistry (Cohort A). Of 6213 samples, 404 (6.5%) showed brain immunostaining suggestive of NSAb: 163 (40%) were positive by commercial IIFA and 241 (60%) were negative. When these 241 samples were re-assessed with in-house IIFA, 42 (18%) were positive: 21 (9%) had NSAb against antigens not included in the commercial IIFA and the other 21 (9%) had NSAb against antigens included in the commercial kit (false negative results). False negative results occurred more frequently with CSF (29% vs 10% in serum) and predominantly affected GABA_BR (39%), LGI1 (17%) and AMPAR (11%) antibodies. Results were reproduced in a separate cohort (B) of 54 AE patients with LGI1, GABA_BR or AMPAR antibodies in CSF which were missed in 30% by commercial IIFA. Patients with discordant GABA_BR antibody results (positive in-house but negative commercial IIFA) were less likely to develop full-blown clinical syndrome; no significant clinical differences were noted for the other antibodies. Overall, NSAb testing by commercial IIFA led to false negative results in a substantial number of patients, mainly those affected by anti-LG1, GABA_BR or AMPAR encephalitis. If these disorders are suspected and commercial IIFA is negative, more comprehensive antibody studies are recommended.

Keywords: autoimmune encephalitis (AE); brain immunohistochemistry; diagnostic test; immunofluorescent assay; neuronal antibodies.

Figure 1

Antibody detection in patients from Cohort A. Workflow used to identify IgG neuronal surface antibodies in a cohort of 6231 samples. IIFA, Indirect immunofluorescent assay; NSAb, Neuronal surface antibodies.

Figure 2

Discrepancies identifying GABA_BR antibodies by IIFA. (A) One of the two patients’ CSF (P2) demonstrating GABA_BR immunoreactivity on rat brain immunohistochemistry. Hippocampus (left panel) and cerebellum (right panel) staining patterns. (B) Upper panels show patient 1’s CSF (P1) reactivity on commercial IIFA and in-house IIFA GABA_BR transfected cells; lower panels show patient 2’s CSF (P2) reactivity on commercial IIFA (negative staining) and in-house IIFA (positive staining) GABAbR transfected cells.

Figure 3

Discrepancies identifying LGI1 antibodies by IIFA. (A) A patient’s CSF demonstrating LGI1 immunoreactivity on rat brain immunohistochemistry. Hippocampus (left panel) and cerebellum (right panel) staining patterns. (B) Left panel shows patient’s CSF reactivity on commercial IIFA LGI1 transfected cells (negative staining); Right panels show CSF reactivity on in-house IIFAs using LGI1 plus ADAM23 transfected cells (upper panels; positive staining) and GPI-LGI1 transfected cells (lower panels, negative staining). GPI: glycosyl-phosphatidylinositol anchored LGI1 (to display the protein on the cell surface).