Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a heterogeneous disease in which diverse autoantibodies have been described but systematic screening has never been performed. Detection of CIDP-specific antibodies may be clinically useful. We developed a screening protocol to uncover novel reactivities in CIDP. Sixty-five CIDP patients and 28 controls were included in our study. Three patients (4.6%) had antibodies against neurofascin 155, four (6.2%) against contactin-1 and one (1.5%) against the contactin-1/contactin-associated protein-1 complex. Eleven (18.6%) patients showed anti-ganglioside antibodies, and one (1.6%) antibodies against peripheral myelin protein 2. No antibodies against myelin protein zero, contactin-2/contactin-associated protein-2 complex, neuronal cell adhesion molecule, gliomedin or the voltage-gated sodium channel were detected. In IgG experiments, three patients (5.3%) showed a weak reactivity against motor neurons; 14 (24.6%) reacted against DRG neurons, four of them strongly (7.0%), and seven (12.3%) reacted against Schwann cells, three of them strongly (5.3%). In IgM experiments, six patients (10.7%) reacted against DRG neurons, while three (5.4%) reacted against Schwann cells. However, results were not statistically significant when compared to controls. Immunoprecipitation experiments identified CD9 and L1CAM as potential antigens, but reactivity could not be confirmed with cell-based assays. In summary, we describe a diverse autoantibody repertoire in CIDP patients, reinforcing the hypothesis of CIDP's pathophysiological heterogeneity.Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a disabling disease with a pathogenesis that remains largely unknown 1 . CIDP response to immune therapies and scarce experimental evidence on passive transfer animal models suggest that humoral factors play a role in its pathogenesis 2 . CIDP diagnosis is based on clinical and electrophysiological criteria 3 that allow the inclusion of a broad spectrum of patients within CIDP, including typical and atypical variants. This heterogeneity has hindered the description of disease-specific biomarkers, despite intensive research efforts 4 .The response of CIDP patients to intravenous immunoglobulin (IVIg) and plasma exchange (PlEx) suggests that humoral factors are involved in its pathogenesis. The search of autoantibodies has been the most important laboratory research topic in CIDP. Initial focus was placed on myelin antigens. Classical studies, using diverse techniques, detected higher frequencies of antibodies against myelin protein zero (MPZ), peripheral myelin protein 2 (PMP2) or peripheral myelin protein 22 (PMP22) [5][6][7][8][9][10] . However, meaningful clinical-immunological correlations with those antigens were not established. CIDP patients harboring antibodies against LM1-containing

BackgroundThe aim of the research is to study the human leukocyte antigen (HLA) class II allele frequencies in chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) associated with anti-neurofascin 155 (NF155) antibodies.MethodsThirteen anti-NF155+ and 35 anti-NF155 negative (anti-NF155neg) CIDP patients were included in a case-control study. The frequencies of the DRB1 HLA allele were analyzed in all patients while DQ frequencies were only studied in patients sharing the DRB1*15 allele. In silico HLA-peptide binding and NF155 antigenicity, predictions were performed to analyze overlap between presented peptides and antigenic regions.ResultsDRB1*15 alleles (DRB1*15:01 and DRB1*15:02) were present in 10 out of 13 anti-NF155+ CIDP patients and in only 5 out of 35 anti-NF155neg CIDP patients (77 vs 14%; OR = 20, CI = 4.035 to 99.13). DRB1*15 alleles appeared also in significantly higher proportions in anti-NF155+ CIDP than in normal population (77 vs 17%; OR = 16.9, CI = 4.434 to 57.30). Seven anti-NF155+ CIDP patients (53%) and 5 anti-NF155neg CIDP patients had the DRB1*15:01 allele (OR = 7, p = 0.009), while 3 anti-NF155+ CIDP patients and none of the anti-NF155neg CIDP patients had the DRB1*15:02 allele (OR = 23.6, p = 0.016). In silico analysis of the NF155 peptides binding to DRB1*15 alleles showed significant overlap in the peptides presented by the 15:01 and 15:02 alleles, suggesting functional homology.ConclusionsDRB1*15 alleles are the first strong risk factor associated to a CIDP subset, providing additional evidence that anti-NF155+ CIDP patients constitute a differentiated disease within the CIDP syndrome.Electronic supplementary materialThe online version of this article (10.1186/s12974-017-0996-1) contains supplementary material, which is available to authorized users.