OBJECTIVE: In type I diabetes mellitus, early markers of beta cell damage are needed in order to detect the infraclinical development of the disease. The reg protein may be a good candidate, as the reg gene has been proposed to play a role in the pancreatic beta cell destruction/regeneration process during diabetogenesis in animal models of autoimmune diabetes. The aim of this study was to test the hypothesis whether serum reg protein level could be representative of either the destructive or regenerative process at the beta cell level during the early phases of type I diabetes in humans. DESIGN AND METHODS: We used a highly specific immunoassay to measure serum reg protein level in controls and in three groups of either diabetes prone or diabetic subjects: recently diagnosed diabetic patients, long-standing diabetic patients and islet cell antibody-positive non-diabetic subjects. RESULTS: We found no significant difference between the values observed in these three groups in comparison with control group (90.7+/-18.1ng/ml, 83.1+/-5.6ng/ml, 98.7+/-24.5ng/ml vs 85.5+/- 5.6ng/ml respectively). Moreover, when the insulin reserve was evaluated at 6 months in the recently diagnosed group, serum reg protein levels were not different between patients with or without residual insulin secretion (at onset: 103+/-42 vs 70.3+/-8. 5ng/ml respectively; at 6 months: 79.7+/-25.8ng/ml vs 81.6+/-15ng/ml respectively). In contrast, trypsin levels were significantly lower in every group of diabetic patients. Results were expressed as means +/- S.E.M. and groups compared by Student's t-test (P<0.05). CONCLUSIONS: We conclude that serum reg protein level cannot be used as a marker for the progression of the diabetogenic process in type I diabetes.