A Systematic Review and Meta-Analysis of Autoantibodies for Diagnosis and Prognosis in Patients With Chronic Inflammatory Demyelinating Polyradiculoneuropathy.

To review the available evidence on sensitivity and specificity of anti-NF155 antibody detection in diagnosing a specific subset of patients with chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) and to calculate the frequencies of different autoantibodies to paranodal proteins. Diagnosis of CIDP relies on clinical and neurophysiologic criteria and lacks useful diagnostic biomarkers. A subset of CIDP patients exhibit atypical clinical phenotypes and impaired response to conventional treatments. These patients were reported as having autoantibodies targeting paranodal protein neurofascin isoform 155 (NF155), contactin-1 (CNTN1), and contactin-associated protein-1 (CASPR1). Here, we conducted a meta-analysis to summarize evidence on the diagnostic and prognostic value of these autoantibodies, especially for anti-NF155 antibody. We searched the following electronic bibliographic databases: PubMed, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), and Web of Science. Eligible studies provided information to calculate the frequencies of anti-NF155 antibody and anti-CNTN1 antibody, the sensitivity and specificity of anti-NF155 antibody, and the incidence of improvement and deterioration among anti-NF155 antibody seropositive CIDP patients. Heterogeneity was assessed using Q and statistics. The pooled frequency of anti-NF155 autoantibody across 14 studies was 7% [95% confidence interval (CI): 0.05-0.10] with high heterogeneity; the overall pooled sensitivity and specificity of anti-NF155 antibody for the diagnosis of a specific subgroup of CIDP patients were 0.45 (95% CI: 0.29-0.63) and 0.93 (95% CI: 0.86-0.97), respectively. For diagnosing of a specific subset of CIDP characterized by poor response to intravenous immunoglobulin (IVIg), we found a moderate sensitivity and a high specificity. The anti-NF155 antibody test should be used as a confirmatory test rather than a screening test. PROSPERO, identifier: CRD42020203385 and CRD42020190789.