Validation of serum IGF-I as a biomarker to monitor the bioactivity of exogenous growth hormone agonists and antagonists in rabbits

Abstract

The development of new growth hormone (GH) agonists and growth hormone antagonists (GHAs) requires animal models for pre-clinical testing. Ideally, the effects of treatment are monitored using the same pharmacodynamic marker that is later used in clinical practice. However, intact rodents are of limited value for this purpose because serum IGF-I, the most sensitive pharmacodynamic marker for the action of GH in humans, shows no response to treatment with recombinant human GH and there is little evidence for the effects of GHAs, except when administered at very high doses or when overexpressed. As an alternative, more suitable model, we explored pharmacodynamic markers of GH action in intact rabbits. We performed the first validation of an IGF-I assay for the analysis of rabbit serum and tested precision, sensitivity, linearity and recovery using an automated human IGF-I assay (IDS-iSYS). Furthermore, IGF-I was measured in rabbits of different strains, age groups and sexes, and we monitored IGF-I response to treatment with recombinant human GH or the GHA Pegvisomant. For a subset of samples, we used LC-MS/MS to measure IGF-I, and quantitative western ligand blot to analyze IGF-binding proteins (IGFBPs). Although recovery of recombinant rabbit IGF-I was only 50% in the human IGF-I assay, our results show that the sensitivity, precision (1.7-3.3% coefficient of variation) and linearity (90.4-105.6%) were excellent in rabbit samples. As expected, sex, age and genetic background were major determinants of IGF-I concentration in rabbits. IGF-I and IGFBP-2 levels increased after single and multiple injections of recombinant human GH (IGF-I: 286±22 versus 434±26 ng/ml; P<0.01) and were highly correlated (P<0.0001). Treatment with the GHA lowered IGF-I levels from the fourth injection onwards (P<0.01). In summary, we demonstrated that the IDS-iSYS IGF-I immunoassay can be used in rabbits. Similar to rodents, rabbits display variations in IGF-I depending on sex, age and genetic background. Unlike in rodents, the IGF-I response to treatment with recombinant human GH or a GHA closely mimics the pharmacodynamics seen in humans, suggesting that rabbits are a suitable new model to test human GH agonists and antagonists.