The objective of this study was to determine if islet autoantibodies (IA) can develop de novo in youth ≥10 y. We also explored the risk of progression to T1D in youth seroconverting to IA after 10 y.

From the prospective DAISY birth cohort (n=2547), we excluded those who were ever previously IA positive, disenrolled prior to age 10, or had >24 month gap in annual IA testing, leaving 917 for analysis. Seroconversion was defined as IA+ by radiobinding assays for insulin, GAD, IA-2 or ZnT8 autoantibodies, confirmed within 3-6 months. The same autoantibodies were also measured by electrochemiluminescence (ECL) assays, to identify high-affinity IA.

IA developed after age 10 at a higher rate (p=0.049) in first-degree relatives than in general population youth with increased-risk HLA (Figure 1); there was no difference by sex or race/ethnicity. Of the 35 IA+ seroconverters, 6 progressed to T1D, during a median follow-up of 5.04 years (IQR: 3.56-8.45). This included 2 of 5 individuals who were multiple IA+ at seroconversion, with at least one IA confirmed as high-affinity by ECL. Among those with a single persistent IA+ at seroconversion, 4/9 youth with high-affinity IA (ECL+) progressed to T1D, compared to 0/21 of the youth with low-affinity IA (ECL-).

In conclusion, seroconversion continues after age 10. In our genetically at-risk cohort, all who progress to T1D had high-affinity IA confirmed by ECL.


B.I. Frohnert: None. F. Dong: None. K. Waugh: None. A. Steck: None. J.M. Norris: None. L. Yu: None. M. Rewers: None.


JDRF (5-ECR-2017-388-A-N); National Institutes of Health (R01DK32493)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at