Letter to the editor re: “Pitfalls in the detection of N-methyl-daspartate-receptor (NMDA-R) antibodies” accurate evaluation of anti-NMDA-R cell-based assay avoids reporting of false positive results

Lars Komorowski, Winfried Stoecker, Johanna Fraune, Christian

Summary

Please cite this article as: Lars Komorowski, Winfried Stoecker, Johanna Fraune, Christian Probst , Letter to the editor re: “Pitfalls in the detection of N-methyl-d-aspartate-receptor (NMDA-R) antibodies” accurate evaluation of anti-NMDA-R cell-based assay avoids reporting of false positive results. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Clb(2016), doi: 10.1016/ j.clinbiochem.2016.12.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
With the Letter to the Editor „Pitfalls in the detection of N-methyl-D-aspartate-receptor (NMDA-R) antibodies” (2016) in Clinical Biochemistry, Gastaldi et al. state a common risk of reporting false positive anti-NMDAR antibody detection by using the commercially available fixed cell-based assay (CBA) manufactured by Euroimmun AG, Germany [1]. The conclusion was drawn from i) data obtained in an external quality assessment scheme (EQAS) for anti-NMDA-R conducted by the Italian Association for Neuroimmunology (AINI) and ii) data from two publications. We would like to respond to both aspects and briefly discuss this matter to avoid further misleading interpretation.
The aforementioned EQAS by AINI involved 5 Italian centers each using the Euroimmun CBA and encompassed a total of 3 samples (target values: a) negative, b) anti-LGI1 positive, c) negative). The authors refer to unpublished data from personal communication, suggesting the reporting of false positive findings among these analyses. To our knowledge, only a single center reported one of the negatively precharacterized samples to be positive for anti-NMDA-R IgG, whereas the remaining centers were in agreement with the target value. Hence, we would assume a mistake during evaluation and/or interpretation of the fluorescence signal in the respective case and not an error which could be blamed on the assay.
We are aware that interpretation of the anti-NMDA-R CBA requires a certain level of expertise. As an example of “dubious interpretation” of the anti-NMDA-R CBA, Gastaldi and colleagues mention a figure from de Witte et al. [2]. In that figure the depicted cells show an uncommon homogeneous fluorescence pattern interpreted as positive for anti-NMDA-R IgG. Retesting of the corresponding sample with different methods within the same study (including another batch of the Euroimmun CBA), however, did not confirm the presence of NMDA-R specific IgG. We agree with Gastaldi et al., that the initial interpretation of this particular fluorescence pattern is inaccurate. Even based on the published image, we would evaluate this staining pattern as anti-NMDA-R IgG negative. A comparison with the positive control, also given in de Witte et al. [2], reveals the differences between the fluorescence patterns. To our experience, a positive reaction with anti-NMDA-R IgG results in fine to coarse granular staining of the cytoplasm with characteristic cytoplasmatic processions and membranous extensions of the cells. This pattern mirrors the distribution of NMDAR within the cells which is constant between batches, and thus, mandatory. Other signals, e.g. homogenous or globular staining must not be reported as anti-NMDA-R IgG positive. This also applies to figure 1 published by Gastaldi et al. [1]. The origin of the respective sample remains unclear and the high microscopic magnification is likely impeding accurate interpretation. As far as we can assess from the figure, however, this result should not be interpreted as positive for anti-NMDA-R IgG.
At this point, we would like to emphasize that accurate reading of CBAs is not just a binary comparison of antigen expressing cells and control cells regarding the brightness of their staining.
Instead, the core requirement for correct classification of an anti-NMDA-R IgG positive sample is the presence of the typical staining pattern. Due to the worldwide availability of standardized slides and reference material every laboratory which is familiar with the use of indirect immunofluorescence assays has the ability to perform accurate evaluations of the CBA.
Moreover, Gastaldi and colleagues refer to a report of equally high frequencies of anti-NMDA-R antibodies in > 4,200 healthy individuals and neuropsychiatric ill patients by Dahm et al. [3] to imply that the anti-NMDA-R IgG assay yields false positive results. It is the largest assessment of antiNMDA-R in non-encephalitis cohorts up-to-date which has been possible through the availability of the validated commercial CBA [4]. Nevertheless, as already indicated by Lancaster et al. [5] in a responding letter, we as well would like to note that the majority of anti-NMDA-R antibodies reported in this scientific study were of class IgA and IgM. Opposed to the synopsis of the study, the presented data indicate that autoimmune encephalitis-associated IgG antibodies to NMAD-R are only observed rarely (≤ 1.3). The low frequency of the anti-NMDA-R IgG in these non-encephalitis cohorts is comparable to the prevalence of many well-defined IgG autoantibodies known to be associated with autoimmune diseases in control panels, and substantially better compared to the frequency reported with a cell-based assay using live cells in which 23% of patients with non-autoimmune or unlikely autoimmune disorders were considered to be anti-NMDA-R antibody positive [6], sometimes leading to unnecessary treatments [7]. Furthermore, the caveats of anti-NMDA-R antibody interpretation were recently described by multiple investigators, suggesting the use of confirmatory tests (tissue immunohistochemistry) particularly when serum testing is used alone [8]. Whether or not the detected anti-NMDA-R IgA or IgM are of diagnostic or even pathogenic importance for any disease remains to be elucidated. Definitely, one must critically distinguish between the three antibody classes, strictly delimiting the encephalitis-associated IgG from the IgA and IgM reactions in the CBA.

References

[1] Gastaldi M, Thouin A, Franciotta D, Vincent A. Pitfalls in the detection of N-methyl-d-aspartatereceptor (NMDA-R) antibodies. Clin Biochem (2016) doi: 10.1016/j.clinbiochem.2016.11.023.
[2] Witte, de LD, Hoffmann C, Mierlo, van HC, Titulaer MJ, Kahn RS, Martinez-Martinez P. Absence of N-methyl-d-aspartate receptor IgG autoantibodies in schizophrenia: the importance of crossvalidation studies. JAMA Psychiatry (2015) 72(7): 731-733.
[3] Dahm L, Ott C, Steiner J, Stepnial B, Teegen B, Saschenbrecker S, Hammer C, Borowski K, Begemann M, Lemke S, Rentzsch K, Probst C, Martens H, Wienands J, Spalletta G, Weissenborn K, Stöcker W, Ehrenreich H. Seroprevalence of autoantibodies against brain antigens in health and disease. Ann Neurol (2014) 76(1): 82-94.
[4] Suh-Lailam BB, Haven TR, Copple SS, Knapp D, Jaskowski TD, Tebo AE. Anti-NMDA-receptor N-Methyl-D-aspartic acid antibody encephalitis: Performance evaluation and laboratory experience with the anti-NMDAreceptor IgG assay. Clin Chim Acta (2013) 421: 1-6.
[5] Lancaster E, Leypoldt F, Titulaer MJ, Honnorat J, Waters PJ, Reindl M, Höftberger R. Immunoglobulin G antibodies to the N-methyl-d-aspartate receptor are distinct from immunoglobulin A and immunoglobulin M responses. Ann Neurol (2015) 77(1):183.
[6] Zandi MA, Paterson RW, Ellul MA, Jacobson L, Al-Diwani A, Jones JL, Cox AL, Lennox B, Stamelou M, Bhatia KP, Schott JM, Coles AJ, Kullmann DM, Vincent A. Clinical relevance of serum antibodies to extracellular N-methyl-D-aspartate receptor epitopes. J Neurol Neurosurg Psychiatry (2015) 86:708713.
[7] Armangue T, Santamaria J, Dalmau J. When a serum test overrides the clinical assessment. Neurology (2015) 84:1379-1381.
[8] Graus F, Titulaer MJ, Balu R, Benseler S, Bien CG, Cellucci T, Cortese I, Dale RC, Gelfand JM, Geschwind M, Glaser CA, Honnorat J, Höftberger R, Iizuka T, Irani SR, Lancaster E, Leypoldt F, Prüss H, Rae-Grant A, Reindl M, Rosenfeld MR, Rostásy K, Saiz A, Venkatesan A, Vincetn A, Wandinger KP, Waters P, Dalmau J. A clinical approach to diagnosis of autoimmune encephalitis. Lancet Neurol (2016) 15:391-404.