Radioimmunoassay determination of insulin autoantibodies in NOD mice. Correlation with increased risk of progression to overt diabetes

In initial cross-sectional study, 29 female and 25 male nondiabetic weaned nonobese diabetic (NOD) mice of various ages (age range 30-300 days, mean 108 ± 10 days) and 11 unweaned NOD pups were evaluated for competitive insulin autoantibodies (CIAAs) with a fluid-phase radioassay. Eleven of 54 (20%) weaned NOD mice had CIAA levels above the range (>39 nU/ml) of 81 control mice. The group of NOD mice that progressed to diabetes had a significantly higher level of CIAAs than NOD mice that did not progress to diabetes (NOD mice progressing to diabetes: CIAA 63 ± 12 nU/ml; NOD mice not progressing to diabetes: CIAA 8 ± 4 nU/ml; P<.02). Seven of 11 (64%) NOD mice having CIAA concentrations exceeding the normal range progressed to diabetes, whereas only 4 of 43 (9%) NOD mice progressed to diabetes without detection of elevated CIAAs (Fisher's exact test, P<.0005). The relative risk of progressing to overt diabetes with CIAA levels >39 nU/ml was therefore 17 (P<.005), giving a positive predictive value of 64%, a negative predictive value of 91%, and an overall accuracy of 85%. None of 11 unweaned NOD pups had CIAA levels above the normal range (mean -9.4 ± 4.9 nU/ml). At 6 wk of age, 37% of female NOD mice were CIAA⁺, whereas none of the male animals exceeded the normal range at this age (38 ± 13 vs. 5 ± 6 nU/ml, P<.05). Results of testing two samples during the follow-up of NOD mice indicated that individual NOD mice positive or negative for CIAAs generally remain positive or negative, and CIAA concentrations often increased in initially positive animals. Fourteen NOD mice were included in a prospective study, and CIAAs were evaluated at 9-wk intervals over a follow-up of 8 mo. Four of these NOD mice had elevated CIAA levels, and 3 of 4 CIAA⁺ mice developed overt diabetes, whereas none of the NOD mice with CIAAs <39 nU/ml developed diabetes. These results were consistent with the results of our cross-sectional study. From these data, we concluded that 1) NOD mice express CIAAs to insulin before progressing to overt diabetes, 2) these antibodies appear after weaning, and 3) CIAA analysis can identify animals with increased or decreased risk of progressing to overt diabetes.