Evaluation of total IGF-I assay methods using samples from Type I and Type II diabetic patients

Measurements of circulating insulin-like growth factor-I (IGF-I) levels are an important part of many studies on growth and development. Circulating IGF-I levels are growth hormone (GH) dependent and are also impacted by age, gender, nutritional status and disease. Moreover, IGF-I is the main pharmacodynamic marker of GH activity. The majority of circulating IGF-I is associated with high affinity insulin-like growth factor-binding proteins (IGFBPs), making accurate and precise measurements of total IGF-I concentrations in biological matrices technically challenging. Many total IGF-I assay methods combine an immunoassay with a sample preparation method aimed at removing IGFBPs. However, not all sample preparation methods efficiently remove all IGFBPs or BP fragments (BPFs), and there is currently no reference method for IGF-I measurement against which these IGF-I assays can be calibrated. We have evaluated a number of IGF-I immunoassays and sample preparation methods using plasma samples from normal donors and from donors with Type I and Type II diabetes mellitus. In order to eliminate the variability between assays due to differences in assay standardization, we used the same preparation of highly pure, fully characterized IGF-I as the standard for all assays. We found that the data produced by many of the IGF-I assay methods showed good agreement when IGF-I levels in samples from normal individuals were measured. However, we found that these agreements were quite poor when IGF-I levels in samples from diabetics were measured. This was true of methods that claimed to physically separate IGFBPs from IGF-I either by acid/ethanol extraction or by acid chromatography. Several methods have recently been developed that physically separate IGF-I from IGFBPs followed by a chemical displacer to displace any residual BPs or BPFs from IGF-I. We found that the data generated by these displacement methods showed good agreement when assaying samples from diabetic as well as normal donors. There is considerable discussion in the literature as to whether individuals with diabetes have normal circulating levels of IGF-I. Many of the published studies are based on assays that may not accurately measure IGF-I levels due to problems with assay standardization and/or with assay methodology. Displacement methods may enable us to more accurately measure IGF-I levels in diabetes.