Increase in remission rate in newly diagnosed type I diabetic subjects treated with azathioprine.

Azathioprine (2 mg/kg) was given, in addition to routine insulin treatment, to alternate patients presenting with recent-onset type I diabetes. Treated (N = 13) and untreated (N = 11) patients did not differ significantly at diagnosis with respect to age, duration of symptoms, body weight, blood glucose, hemoglobin A1c, or presence of ketosis. Eight patients were treated for 12 mo, three elected to stop treatment at 6 mo, and treatment was stopped in two because of side effects. Seven treated patients had a remission compared with one untreated patient. At 12 mo these seven patients were distinguished by significantly higher basal and glucagon-stimulated levels of C-peptide (1.98 +/- 0.52 and 3.88 +/- 0.34 micrograms/L, respectively) compared with the other six treated patients (0.93 +/- 0.52 and 1.32 +/- 0.85 microgram/L, respectively), and by the persistence of islet cell cytoplasmic antibodies. Remissions were not sustained in the 1-2 yr after treatment, although relapsed patients required less insulin for control. These results corroborate those from nonrandomized trials using cyclosporine and suggest that protracted treatment with nonspecific immunosuppressive drugs may be necessary to avert insulin dependence.