Nicotinamide (NCT) has been shown to be effective in preventing the onset of type 1 diabetes mellitus (IDDM) in mice with non-obese diabetic (NOD) and beta cell damage, mediated by the diabetogenic agents including streptozotocin. NCT therapy in man has been shown to have a beneficial effect on the remission phase of IDDM, and its use is safe. In this open pilot trial we therefore studied the effect of oral NCT administration on insulin secretion rate and islet-cell antibody (ICA) titres in IDDM high risk subjects. NCT (25 mg/10 kg bw) was administered in 6/13 high risk patients identified by a family screening programme. Those subjects tested after eight months without treatment showed a decreasing secretion in comparison to onset baseline (56,1 +/- 37.8 versus 35,5 +/- 12.2), whereas the treated subjects showed an increasing insulin secretion after treatment (26 +/- 10 versus 50.2 +/- 26.6), in spite of ICA persistence. Statistical analysis shows an increased insulin secretion in the treated group versus the untreated group (chi 2 = 3.899, P = 0.048). No side-effects were observed. We conclude that NCT may repair beta-cell function in high risk subjects too if damage is not too severe; furthermore, the effect seems not to be mediated by an immune mechanism.