Search

ObjectiveTo evaluate neuropathy phenotypes and clinical outcomes associated with trisulfated heparin disaccharide (TS-HDS) autoantibodies.BackgroundTS-HDS autoantibody has been reported as a biomarker of immune-mediated neuropathy. However, studies evaluating the clinical associations of this autoantibody are limited.Design/MethodsElectronic medical records were reviewed to identify TS-HDS autoantibody seropositive patients and characterize their clinical and electrodiagnostic findings.ResultsSeventy-seven TS-HDS-IgM seropositive (titer range 9000-350,000) patients were identified (50 females; median age of onset was 48 years (range 9-83 years). Eleven patients were also positive for FGFR3-IgG (titer range 4000-19,000). 70% (54/77) had clinical/paraclinical evidence of neuropathy (54/77, 70% of TS-HDS-IgM alone; 10/11, 91% of TS-HDS-IgM with coexisting FGFR3-IgG). The managing physician characterized an immune-mediated neuropathy in 30% (23/77) and 54% (6/11) of the TS-HDS-IgM only and TS-HDS-IgM with coexisting FGFR3-IgG seropositive patients, respectively. Small fiber neuropathy presented in 58% (45/77) and 63% (7/11) of TS-HDS-IgM only, and both antibodies seropositive patients, respectively. Length-dependent neuropathy was the most common neuropathy phenotype amongst TS-HDS IgM (43/54, 80%) and dual seropositive cases (7/11, 63%). Forty-one (53%) patients received immunotherapy, predominantly: IVIG (n = 37), IV solumedrol (n = 7), oral prednisone (n = 14), and mycophenolate mofetil (n = 12). Among these, 43% (15/35) with TS-HDS-IgM seropositivity alone had improvement in inflammatory neuropathy cause and treatment (INCAT) disability score or modified Rankin Scale (mRS), while 33% (2/6) of patients with dual seropositivity had INCAT and mRS improvement. TS-HDS-IgM titers had low discriminative ability to identify immunotherapy response with an AUC of 0.621.ConclusionsNeuropathy associations and clinical phenotypes amongst TS-HDS-IgM seropositive cases are variable. Furthermore, only a minority of cases are immunotherapy responsive, limiting the value of this biomarker in identifying immune-mediated neuropathies.

ObjectiveTo define the clinicopathologic features of amphiphysin-immunoglobulin G (IgG)–mediated neuropathy.MethodsPatients examined at our institution from January 1, 1995, to September 30, 2018, with amphiphysin-IgG by indirect immunofluorescence and Western blot, were reviewed. Their phenotypes were compared to cases of coexisting collapsin response-mediator protein-5 (CRMP5)–IgG or anti-neuronal nuclear antibody type 1 (ANNA1-IgG) and CRMP5-IgG autoimmunity. Improvement in modified Rankin Scale (mRS) (≥1) on follow-up was considered a favorable outcome. Amphiphysin RNA expression was assessed in healthy nerves.ResultsFifty-three amphiphysin-IgG–positive cases were identified. Of 33 (60%) patients with neuropathy, 21 had amphiphysin-IgG alone, and 12 had coexisting autoantibodies (ANNA1-IgG, n = 8; CRMP5-IgG, n = 2; ANNA1-IgG and CRMP5-IgG, n = 2). The neuropathies in isolated amphiphysin-IgG autoimmunity included polyradiculoneuropathy (62%), diffuse sensory neuronopathy (35%), and facial neuropathy with gastroparesis (3%). Among these, pain (80%), breast cancer (63%), and CNS (57%) involvements commonly coexisted, and neuropathy frequently prompted breast cancer diagnosis (76%). Stiff-person spectrum disorder was the most common CNS accompaniment (45%). Nerve biopsies showed axonal loss (n = 6/6), subperineurial edema (n = 4/6), and CD4 predominant inflammation (n = 2/6). Median mRS score at last follow-up was 3.5; 58% of patients were immunotherapy-responsive. Patients with amphiphysin-IgG alone had more favorable immunotherapy response than patients with CRMP5-IgG polyneuropathy (n = 45) (44% vs 16%, p = 0.028, odds ratio 4.2, 95% confidence interval 1.1 to 15.5). Only 1/9 (11%) patients with amphiphysin-IgG with coexisting CRMP5-IgG or ANNA1-IgG had immunotherapy response. RNA amphiphysin expression occurred at low levels in nerve.ConclusionAmphiphysin-IgG autoimmune neuropathy has a recognizable phenotype, is frequently immune responsive, and can prompt early diagnosis of breast cancer.

The report of Ebright et al.1 unfortunately provided an outdated and inaccurate representation of the current state of neural autoantibody testing offered by the Mayo Clinic Neuroimmunology Laboratory.