Previous studies suggest that after 6 years of discordance, identical twin pairs rarely become concordant for type I diabetes. With up to 39 years of follow-up from the onset of diabetes in the index twin, we determined how many discordant twins have evidence of β-cell autoimmunity and how many develop overt diabetes. We longitudinally followed 23 pairs of identical twins (or triplets) that were selected from a total group of 30 pairs because they were discordant for type I diabetes when first ascertained. Seven developed diabetes after 3, 3, 7, 8, 9, 31 and 36 years of discordance. By survival analysis, the concordance after 10 years from the onset of diabetes in the index twin was estimated as 23% (95% confidence interval, 5–40%), increasing to 38% (95% confidence interval, 8–69%) after 31 years. Among 16 twins remaining nondiabetic at last follow-up (8–39 years of discordance), 12 were assessed with serial intravenous glucose tolerance tests and a total of 407 measurements by radioassay of antibodies against three defined autoantigens (glutamic acid decarboxylase, insulin, and the recently cloned molecule ICA512). Two-thirds (8 of 12) had evidence of β-cell autoimmunity (persistently positive autoantibody levels) and/or first-phase insulin release < the 1st percentile of control subjects. In summary, identical twins may develop diabetes after a prolonged period of discordance and ∼ two-thirds of long-term discordant twins have evidence of persistent β-cell autoimmunity and/or β-cell damage. The concordance for β-cell autoimmunity, therefore, is much higher than for overt diabetes. This suggests that additional environmental or non-Mendelian genetic factors or time are required for the development of type I diabetes.

This content is only available via PDF.