Session: FP20-Growth: Clinical Trials & Observational Studies
Room 122 (Moscone Center)
Poster Board SUN-629
OBJECTIVES: To investigate the association between markers of Ins sensitivity (change in fasting glucose and Ins over 1mth) and markers of growth (change in IGF-I over 1mth and height velocity [HV] over 1yr) in relation to baseline gene expression.
METHODS: Prepubertal children with GHD (n=125) were enrolled from the PREDICT (NCT00256126) and PREDICT long-term follow-up (NCT00699855) studies. Whole blood gene expression was determined at baseline (Affymetrix Human Genome U133 Plus 2.0 Array), changes in biomarkers (Glucose, Ins, IGF-I) were assessed after 1mth on r-hGH (median dose 33µg/kg/d), and HV measured after 1yr. First, correlations between changes in these clinical markers were assessed adjusted for age, gender, and body mass index. Second, associations between basal gene expression and changes in clinical markers were assessed using rank regression. Third, genes common to pairs of related markers (glucose+Ins; Ins+IGF-I; IGF-I+HV) were subjected to a network analysis based on all known interacting partners for these genes (generating an ‘interactome’ model [Biogrid]). Fourth, overlap of these interactomes between the marker pairs was examined and highly connected networks identified (ClusterOne algorithm). Finally, pathway associations within these networks were determined (hypergeometric test).
RESULTS: (1) Changes in glucose + Ins were correlated (p<0.05). Changes in IGF-I correlated with Ins (p<0.001) but not glucose (p=0.34). HV at 1yr correlated with 1mth changes in IGF-I (p<0.002) but not with glucose or Ins. (2) Changes in clinical markers were associated with 891 (Ins), 1844 (Gluc), 3583 (IGF-I), and 2140 (HV) genes. (3) 72 (glucose+Ins), 101 (Ins+IGF-I), and 317 (IGF-I+HV) genes were common to each correlated pair. (4) The top pathways associated with the networks derived from these comparisons included adipocyte differentiation (p<1.5x10-34) and cell-cycle regulation (p<3.6x10-11).
CONCLUSIONS: We have used a network biology approach based on baseline gene expression in children with GHD to determine common pathways that underlie glucose, insulin, IGF-I, and growth responses to r-hGH. These data could be used to generate individual gene expression profiles associated with both the efficacy and safety of r-hGH.
Disclosure: AS: Speaker, Merck Serono. CD: Research Funding, NovoNordisk ESPE Research Fellowship. BD: Employee, Merck Serono. MDR: Investigator, Merck Serono. JR: Consultant, Merck Serono. PC: Speaker, Merck Serono, Investigator, Merck Serono, Consultant, Merck Serono. PEC: Speaker, Merck Serono, Investigator, Merck Serono. Nothing to Disclose: GA
*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm
See more of: Abstracts - Orals, Featured Poster Presentations, and Posters