We studied 5-yr T1D risk associations using a series of risk factors in an autoantibody positive population of 6,020 in the TrialNet Pathway to Prevention study (PTP) study. We used data from 25 subgroups developed from 5 AUC glucose (<115;115-<130;130-<145;145-<160; ≥160 mg/dl) and 5 AUC C-peptide (<3.5;3.5-<5.0;5.0-<6.5;6.50-<8.0; ≥8.0 ng/ml) categories from baseline OGTT data. Subgroup boundaries were set to provide quantitative power for proper statistical analysis (range: 138 to 487) . We assessed 5-yr T1D risk associations with risk factors among the 25 subgroups using correlation coefficients and linear regression analysis (slope: 5-yr risk/risk factor prevalence) . The table shows the strongest correlation of 5-yr T1D risk with IA-2A (r=0.96) . This was consistent in split sample analyses, as were the linear regression indices (Sample 1 vs. Sample 2: r=0.96 vs. 0.93; intercept=-0.186 vs. -0.228; slope=0.014 vs. 0.015) . Using the regression indices for a cross-sample validation (i.e., predicting 5-yr risk for one sample using its IA-2A prevalence and the other sample's regression indices) , the predicted/actual 5-yr risk was 0.27/0.26 for Sample 1 and 0.27/0.23 for Sample 2.

In summary, in PTP subgroups, 5-yr T1D risk correlated strongly with IA-2A. Further, regression indices predicted 5-yr T1D risk in cross-sample validation, suggesting a tight linkage between active β-cell decline and IA-2A.


H.M.Ismail: n/a. D.D.Cuthbertson: None. M.A.Atkinson: None. M.J.Redondo: Advisory Panel; Provention Bio, Inc. J.Sosenko: None. Type 1 diabetes trialnet: n/a.



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 http://www.diabetesjournals.org/content/license.