Serum inflammatory proteomic signatures define chronic inflammatory demyelinating polyneuropathy and inform on disease activity

BACKGROUND: Chronic Inflammatory Demyelinating Polyneuropathy (CIDP) is an immune-mediated neuropathy with heterogeneous presentation and variable treatment response. Blood-based biomarkers remain lacking, and interpretation is confounded by widespread immunoglobulin (IG) therapy. The primary aim was to identify serum inflammatory proteomic signatures associated with CIDP; exploratory aims included associations with disease activity, phenotype, and treatment response. METHODS: In this case-control study, 92 proteins were quantified using multiplex proximity extension assay and ELISA in 358 sera samples from patients with CIDP, multiple sclerosis (MS), and IG-treated myasthenia gravis (MG) as well as age- and sex- matched healthy controls (HC). Logistic regression with Bonferroni correction, principal component analysis, and correlation analyses were applied to identify differential expression across diagnoses and CIDP subgroups. FINDINGS: Seven proteins, TNFSF14, CD40, SIRT2, CCL3, IL8, TGFα, and uPA, consistently distinguished CIDP from HC, MS, and IG-treated MG, representing disease-specific signatures independent of IG exposure. Six additional proteins (STAMBP, CASP-8, EN-RAGE, OSM, HGF, and CXCL6) were IG-responsive. CXCL9, CDCP1, and CCL20 correlated with muscle weakness, while calprotectin and neurofilament light chain reflected broader inflammatory and axonal injury signatures. Patients with unstable active disease exhibited higher CCL4, IL-8, CCL3, FGF-21, and IL-17A. IL-8, EN-RAGE, and CASP-8 predicted clinical improvement, and IL-5 tracked longitudinal disability changes. INTERPRETATION: CIDP is characterised by distinct serum inflammatory proteomic signatures comprising disease-specific and IG-responsive signatures. These candidate biomarkers may support improved patient stratification, disease monitoring, and prediction of treatment response, advancing individualised therapy in CIDP. FUNDING: Erling-Persson Foundation, Sweden (#2022_0030 to ARP) and the German Research Foundation, Germany (KFO5023 TP8 to AM).