Use of an islet cell antibody assay to identify type 1 diabetic patients with rapid decrease in C-peptide levels after clinical onset

OBJECTIVE - To investigate whether the presence of antibody markers at diag nosis could help predict the rapid decrease in residual beta-cell function noted in some, but not all, patients with recent-onset type 1 diabetes. RESEARCH DESIGN AND METHODS - We measured random C-peptide levels (radioimm unoassay); islet cell cytoplasmic antibodies (ICA) (indirect immunofluoresc ence); and antibodies against IA-2 protein, 65-kDa glutamate decarboxylase, and insulin (liquid-phase radiobinding assays) in 172 patients <40 years o f age with type 1 diabetes. The patients had been consecutively recruited a t diagnosis by the Belgian Diabetes Registry and were followed for 2 years. RESULTS - Two years after diagnosis, random C-peptide levels had decreased significantly (P < 0.001) in ICA(+) patients but not in ICA(-) patients. C- peptide values <50 pmol/l were noted in 88% of patients diagnosed before 7 years of age, in 45% of patients diagnosed between ages 7 and 15 years, and in 29% of patients diagnosed after 15 years of age (P < 0.001). In cases o f clinical onset before age 15 years, a rapid decline in random C-peptide v alues was observed almost exclusively in patients with high-titer ICA (grea ter than or equal to 50 Juvenile Diabetes Foundation [JDF] units) at diagno sis (69 vs. 17% in patients with lower ICA titers, P < 0.001). In patients diagnosed after 15 years of age, 36% of patients with ICA titers greater th an or equal to 12 JDF units developed low C-peptide levels compared with 14 % of patients with ICA titers <12 JDF units (P < 0.03), Multivariate analys is confirmed that C-peptide levels after 2 years were inversely correlated with ICA levels (P < 0.001) and to a lesser degree positively correlated wi th age at diagnosis (P < 0.02), regardless of the levels or number of molec ular autoantibodies. CONCLUSIONS - Young age at diagnosis and high-titer ICA identify a group of type 1 diabetic patients at high risk of rapidly losing residual beta-cell function. Using these selection criteria, it is possible to better target beta-cell-preserving interventions to patients with or without such rapid p rogression, depending on the nature of the tested substance. The ICA assay measures clinically relevant antibodies nor detected in antibody assays tha t use recombinant human autoantigens for substrate.