Development of a type 2 diabetes risk model from a panel of serum biomarkers from the Inter99 cohort

Abstract

Objective: The purpose of this study was to develop a model for assessing the 5-year risk of developing type 2 diabetes from a panel of 64 circulating candidate biomarkers.

Research design and methods: Subjects were selected from the Inter99 cohort, a longitudinal population-based study of approximately 6,600 Danes in a nested case-control design with the primary outcome of 5-year conversion to type 2 diabetes. Nondiabetic subjects, aged >or=39 years, with BMI >or=25 kg/m(2) at baseline were selected. Baseline fasting serum samples from 160 individuals who developed type 2 diabetes and from 472 who did not were tested. An ultrasensitive immunoassay was used to measure of 58 candidate biomarkers in multiple diabetes-associated pathways, along with six routine clinical variables. Statistical learning methods and permutation testing were used to select the most informative biomarkers. Risk model performance was estimated using a validated bootstrap bias-correction procedure.

Results: A model using six biomarkers (adiponectin, C-reactive protein, ferritin, interleukin-2 receptor A, glucose, and insulin) was developed for assessing an individual's 5-year risk of developing type 2 diabetes. This model has a bootstrap-estimated area under the curve of 0.76, which is greater than that for A1C, fasting plasma glucose, fasting serum insulin, BMI, sex-adjusted waist circumference, a model using fasting glucose and insulin, and a noninvasive clinical model.

Conclusions: A model incorporating six circulating biomarkers provides an objective and quantitative estimate of the 5-year risk of developing type 2 diabetes, performs better than single risk indicators and a noninvasive clinical model, and provides better stratification than fasting plasma glucose alone.

Figure 1

Performance and validation of a model to assess risk of 5-year incidence of type 2 diabetes in the Inter99 cohort. Shown are ROC curves for a model that uses the levels of six biomarkers (fasting serum ADIPOQ, CRP, insulin, FTH1, and IL-2RA and fasting plasma glucose) that was developed using the entire dataset (all 632 converters and nonconverters, blue line) and validated using a bootstrap resampling approach (red dashed line).

Figure 2

ROC analyses for 11 methods of assessing 5-year risk for type 2 diabetes. DRS, diabetes risk score developed in the present study; HOMA-IR, (fasting serum insulin × fasting plasma glucose)/22.5; noninvasive clinical model, a noninvasive clinical algorithm using age, BMI, waist circumference, and family history in a first-degree relative); OGTT, 2-h oral glucose tolerance test. ***P > 0 < 0.001; **P = 0.001 to < 0.01.

Figure 3

Performance of the DRS and FPG in the at-risk Inter99 subpopulation defined by BMI ≥25 kg/m² and age ≥39 years. The green, yellow, and pink regions correspond to the low-, medium-, and high-risk strata, respectively. The results from the study were adjusted using Bayes' law to reflect the observed 5-year incidence of 5.7% among the 3,032 at-risk individuals in Inter99 (A). On the left axis, absolute risk is indicated, and relative risk is shown on the right axis. ——, Relationship between risk and DRS prediction; · · · · ·, mean upper and lower 95% CIs on the risk, as estimated from the SEM of the individual risk predictions in the study; ▴, deciles of the adjusted study population. The mean observed fraction that converted is plotted versus mean DRS. Details of the development of this risk curve are presented in online Appendix C. Stratification of the at-risk Inter99 subpopulation by fasting plasma glucose status (B) and by DRS risk stratum (C). NFG, normal fasting glucose (≤100 mg/dl); IFG, impaired fasting glucose (>100 mg/dl).