OBJECTIVE: We tested the hypothesis that insulin therapy rather than sulfonylurea (SU) treatment is preferable to reverse or preserve beta-cell function among patients with slowly progressive insulin-dependent (type 1) diabetes (SPIDDM) or latent autoimmune diabetes in adults. METHODS: This multicenter, randomized, nonblinded clinical study screened 4089 non-insulin-dependent diabetic patients for glutamic acid decarboxylase autoantibodies (GADAb). Sixty GADAb-positive non-insulin-requiring diabetic patients with a 5-yr duration or shorter of diabetes were assigned to either the SU group (n = 30) or the insulin group (n = 30). Serum C-peptide responses to annual oral glucose tolerance tests were followed up for a mean of 57 months. The primary endpoint was an insulin-dependent state defined by the sum of serum C-peptide values during the oral glucose tolerance test (SigmaC-peptide) less than 4 ng/ml (1.32 nmol/liter). RESULTS: The progression rate to an insulin-dependent state in the insulin group (three of 30, 10%) was lower than that in the SU group (13 of 30, 43%; P = 0.003, log-rank). Longitudinal analysis demonstrated that SigmaC-peptide values were better preserved in the insulin group than in the SU group. Multiple regression analysis demonstrated that insulin treatment, a preserved C-peptide response, and a low GADAb titer at entry were independent factors in preventing progression to an insulin-dependent state. Subgroup analysis suggested that insulin intervention was highly effective for SPIDDM patients with high GADAb titers [> or =10 U/ml (180 World Health Organization U/ml)] and preserved beta-cell function [SigmaC-peptide > or = 10 ng/ml (3.31 nmol/liter)] at entry. No severe hypoglycemic episodes occurred during the study. CONCLUSIONS: Insulin intervention to preserve beta-cell function is effective and safe for patients with SPIDDM or latent autoimmune diabetes in adults.