Metabolic effects and long-term safety of childhood growth hormone treatment

Abstract

The overall aim of this thesis was to investigate the metabolic effects and long-term safety of childhood growth hormone (GH) treatment. In order to achieve this aim, the different projects consist of a two-part clinical trial on metabolic features linked to GH physiology and GH treatment as well as two large population-based cohort studies with focus on the long-term cardiovascular and cancer risks in previously GH-treated patients.

In study I, the metabolic profile of 35 prepubertal children of short stature, between 7 and 10 years of age, with stimulated peak GH levels in the lower normal range (7-14 µg/L) was compared to 12 age- and sex-matched control children of normal height and weight. The groups were compared using blood samples of fasting glucose and insulin, HbA1c, insulin-like growth factor I (IGF-I), insulin sensitivity using both homeostasis model assessment of insulin resistance (HOMA-IR) and frequently sampled intravenous glucose tolerance test (FSIVGTT), dual-energy x-ray absorptiometry (DEXA), microdialysis and stable isotope examination of glucose production and lipolysis. Few differences between the groups were found but the subgroup of children with the lowest GH peak levels demonstrated lower fasting insulin levels and signs of increased insulin sensitivity. In study II, the 35 short children from study I were subsequently randomized to three different doses of recombinant human GH (rhGH) treatment; low dose (11 µg/kg/d), standard dose (33 µg/kg/d) or high dose (100 µg/kg/d), and followed for two years. The doses were blinded to both patients and the study investigators. The metabolic effects of the different treatment doses were analyzed by the same methods as in study I and a clear dose-dependent metabolic effect could be demonstrated, in particular for the high dose group regarding fasting insulin and different measures of insulin sensitivity.

In study III, the long-term cardiovascular morbidity in childhood rhGH-treated Swedish patients between 1985 and 2010, due to isolated GH deficiency (GHD), small for gestational age (SGA) or idiopathic short stature (ISS), was investigated. Data on cardiovascular outcomes and important covariates were gathered for a total of 3,408 patients and 50,036 randomly selected controls matched on sex, age and county. Time to first cardiovascular event was analyzed by Cox proportional-hazard regression models and the study showed increased adjusted hazard ratios for the patients compared to the controls. In study IV, the long-term cancer incidence and mortality in a large meta-cohort of approximately 24,000 previously childhood rhGH-treated patients from eight European countries were investigated. The results did not support an overall carcinogenic effect of rhGH treatment but the significant trend of increased cancer mortality risk in relation to rhGH dose in patients with previous cancer and the indication of possible effects on bone cancer, bladder cancer and Hodgkin’s lymphoma requires further vigilance.