A 1-year open randomized controlled multicentre trial was carried out on 90 patients with recent onset (< 4 weeks) insulin-dependent diabetes (IDDM) to compare the effect of nicotinamide (NCT) with the combination NCT and low dose cyclosporin (CyA) on clinical remission and optimization of metabolic control during the first year from diagnosis. Three groups of patients were randomly assigned to receive for 12 months either NCT 25 mg kg-1 day-1 (n = 30) or NCT 25 mg kg-1 day-1 + CyA 5 mg kg-1 day-1 (n = 30), the latter adjusted to maintain 12 whole blood trough levels of 83 nmol l-1; a third group of patients (n = 30) receiving insulin only acted as a control group for spontaneous remission and metabolic control. Clinical remission (i.e. suspension of insulin therapy with normal metabolic parameters for more than 2 weeks according to the International Diabetes Immunotherapy Group) was achieved at 3 months in 6/30 NCT treated patients and in 1/30 NCT + CyA treated patient (p = 0.05); no remission was observed in control patients. At 6 months the number of patients achieving remission in each group was 4/29, 3/27, and 1/29, respectively (p = NS). One year after diagnosis 4/27 NCT treated, 2/25 NCT+CyA treated but 0/28 of the control patients were in remission (NCT vs control p = 0.05). Clinical remission lasted longer (7 +/- 3 SD months) in NCT treated patients than in NCT+CyA treated or control patients (p < 0.02).(ABSTRACT TRUNCATED AT 250 WORDS)