ESPE Abstracts (2014) 82 FC7.3

Gene Expression Networks Associated with Changes in Serum Markers of Metabolism and Growth in GH-Treated Children with GH Deficiency

Adam Stevensa, Chiara De Leonibusa, Pierre Chatelainb & Peter Claytona


aRoyal Manchester Children’s Hospital, Manchester, UK; bDépartement de Pédiatrie, Université Claude Bernard, Lyon, France


Introduction: Growth promoting effects of GH occur in parallel with its impact on insulin sensitivity and lipid metabolism; underlying biological networks that link these actions are not defined. Our objective was to identify gene expression (GE) networks linking growth with metabolic responses in GH-treated children with GHD.

Methods/design: Pre-pubertal children with GH Deficiency GHD (n=125) were enrolled from the PREDICT short-term (NCT00256126) and long-term follow-up prospective studies (NCT00699855). Whole blood GE was determined prior to treatment with GH. Associations were identified by partial correlation between height velocity after 1 year (HV1) and serum biomarker levels (change over 1 month of treatment in IGF1, insulin (Ins), HOMA-IR and triglycerides). GE was correlated with biomarkers using rank regression, and overlap in GE profiles between these markers was used to build network clusters from which a priority list of biological functions was generated (Moduland algorithm1 and hypergeometric test).

Results: Correlations between the following parameters (HV1 or 1-month change in biomarker levels) were observed: i) HV1 with IGF1; ii) IGF1 with Ins and HOMA-IR; iii) triglycerides with Ins and HOMA-IR (all P<0.05). The GE profile related to each correlation was defined. The GE common to HV1/IGF1 to IGF1/Ins/HOMA-IR included seven genes: C1orf21, SPTBN1, IQCH, LINC00667, SFSWAP, SLC39A8, and UGGT2 (all P<0.05). C1orf21, SPTBN1, and IQCH have been associated with adult height, IQCH with age at menarche, and all except LINC00667 with diabetes and obesity. Two (C1orf21 (associated with cell proliferation) and SPTBN1 (associated with organization of organelles)) were also represented in the GE overlap between triglycerides/Ins/HOMA-IR (P<0.05). GE network clusters indicated that cell cycle and adipogenesis pathways were the closest correlated functions (P<0.001).

Conclusions: This study identified distinct biological pathways and potential genomic markers that link growth and metabolic responses to r-hGH therapy in GHD children.

Article tools

My recent searches

No recent searches.