Background

Antibodies to posttranslationally modified insulin (oxPTM‐INS‐Ab) are a novel biomarker of type 1 diabetes (T1D). Here, we evaluated whether oxPTM‐INS‐Ab can improve T1D prediction in children with positive standard islet autoantibodies (AAB).

Methods

We evaluated sensitivity, specificity, accuracy, and risk for progression to T1D associated with oxPTM‐INS‐Ab and the standard islet AAB that include insulin (IAA), GAD (GADA), and tyrosine phosphatase 2 (IA‐2A) in a cohort of islet AAB‐positive (AAB+) children from the general population (median follow‐up 8.8 years).

Results

oxPTM‐INS‐Ab was the most sensitive and specific autoantibody biomarker (74% sensitivity, 91% specificity), followed by IA‐2A (71% sensitivity, 91% specificity). GADA and IAA showed lower sensitivity (65% and 50%, respectively) and specificity (66% and 68%, respectively). Accuracy (AUC of ROC) of oxPTM‐INS‐Ab was higher than GADA and IAA (P = 0.003 and P = 0.017, respectively), and similar to IA‐2A (P = 0.896). oxPTM‐INS‐Ab and IA‐2A were more effective than IAA for detecting progr‐T1D when used as second‐line biomarker in GADA+ children. Risk for diabetes was higher (P = 0.03) among multiple AAB+ who were also oxPTM‐INS‐Ab+ compared with those who were oxPTM‐INS‐Ab–. Importantly, when replacing IAA with oxPTM‐INS‐Ab, diabetes risk increased to 100% in children with oxPTM‐INS‐Ab+ in combination with GADA+ and IA‐2A+, compared with 84.37% in those with IAA+, GADA+, and IA‐2A+ (P = 0.04).

Conclusions

Antibodies to oxidized insulin (oxPTM‐INS‐Ab), compared with IAA which measure autoantibodies to native insulin, improve T1D risk assessment and prediction accuracy in AAB+ children.