The effects of a freeze-thaw cycle and pre-analytical storage temperature on the stability of insulin-like growth factor-I and pro-collagen type III N-terminal propeptide concentrations: Implications for the detection of growth hormone misuse in athletes

A method based on two serum biomarkers – insulin‐like growth factor‐I (IGF‐I) and pro‐collagen type III N‐terminal propeptide (P‐III‐NP) – has been devised to detect growth hormone (GH) misuse. The aims of this study were to determine the stability of IGF‐I and P‐III‐NP concentrations in serum stored at ?20°C and to establish the effects of one freeze‐thaw cycle. Blood was collected from 20 healthy volunteers. Serum aliquots were analyzed after storage for one day at 4°C and one day, one week, five weeks, and three months at ?20°C. IGF‐I and P‐III‐NP results were combined to calculate a GH‐2000 discriminant function score for each volunteer. Inter‐assay precision was determined by analysing one quality control sample at each time‐point. A single freeze‐thaw cycle, storage of serum at 4°C for one day and at ?20°C for up to three months had no significant effect on IGF‐I or P‐III‐NP concentration. Intra‐sample variability for IGF‐I was 6.8% (Immunotech assay) and 12.9% (DSL assay). Intra‐sample variability for P‐III‐NP was 10.9% (Cisbio assay) and 13.7% (Orion assay). When IGF‐I and P‐III‐NP results were combined, intra‐sample variability of the GH‐2000 score expressed as a standard deviation varied between 0.31 and 0.50 depending on the assay combination used. Variability in IGF‐I and P‐‐III‐NP results of stored samples is largely determined by the characteristics of the assays. A single freeze‐thaw cycle, storage of serum at 4°C for one day or at ?20°C for up to 3 months does not result in a significant change in GH‐2000 score. Copyright © 2012 John Wiley & Sons, Ltd.