Introduction: Anti-GAD65 is an established diagnostic biomarker for autoimmune diabetes and neurological syndromes associated with GAD antibody-spectrum disorders. Automated chemiluminescence-based immunoassays such as Maglumi are increasingly used, but their behavior at high antibody concentrations remains poorly characterized. We report the analytical concordance between Maglumi and radioimmunoassay (RIA), the cut-off titer that differentiates neurological from diabetic patients, and the hook effect of the immunoassay.
Materials and methods: Anti-GAD65 levels were analyzed in samples from 90 patients with suspected autoimmune diabetes or neurological syndromes, using Maglumi 2000® and RIA (Medipan®). Analytical concordance was assessed by Passing-Bablok regression, Bland-Altman plots, and Cohen's kappa. ROC curve analysis identified diagnostic thresholds, and dilution studies were performed in samples with unexpectedly low values to assess a potential hook effect.
Results: Maglumi results were approximately five times higher than RIA results (slope = 5.364; R2 = 0.908); agreement was higher in endocrine (R2 = 0.998) than in neurological patients (R2 = 0.891). A cutoff of 6778 IU/mL discriminated neurological patients from diabetic patients (AUC = 0.96). Sixteen samples showed a hook effect, with direct (undiluted) values ≤273 IU/mL, while the corrected titers after dilution exceeded 20,000 IU/mL. A linear regression confirmed an inverse relationship between direct and true titers (r = -0.79, p < 0.05).
Discussion: Maglumi anti-GAD65 immunoassay provides reliable discrimination between endocrine and neurological patients but requires method-specific cutoffs to avoid misclassification. Hook effect was frequent in high-titer neurological samples, potentially leading to misdiagnosis, underscoring the need for routine dilution protocols to improve diagnostic reliability.
Conclusions: Standardized dilution protocols and appropriate cutoffs are essential to ensure reliable anti-GAD65 testing with Maglumi chemiluminescent immunoassay, particularly in neurological indications.
Keywords: Anti-GAD; Chemiluminescence immunoassay; Hook effect; Maglumi 2000; Radioimmunoassay; Stiff Person syndrome.
Copyright © 2025 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.