IGF-I During High-dose Growth Hormone Treatment of Children Born SGA

« Back to Volume 22, Issue 3, September 2006 - Table of Contents

Growth hormone (GH) treatment in short children born small for gestational age (SGA) results in a significant improvement of final height, as most children reach normal stature defined as height above –2SD. This treatment is accepted by health authorities in both the US and Europe. However, a previous study by these authors had shown that only a slight and non-significant difference in adult height was observed when 2 GH doses (1 mg/m²/d [0.033 mg/kg/d] and 2 mg/m²/d [0.067 mg/kg/d]) were compared.¹ Other reports have shown that the serum insulin-like growth factor (IGF)-I and IGF binding-protein (IGFBP)3 levels are particularly related to the GH dose. Therefore, this study aimed to document GH levels during an overnight profile and IGF-I / IGFBP3 levels before and after 6 months of treatment. This was performed in view of multiple epidemiological studies pointing at cancer risk in relation to high circulating GH and IGF-I levels.

Thirty-six prepubertal short children born SGA were stratified according to gender, and randomized into 2 groups according to dose of GH. The overnight GH profile after subcutaneous injection and IGF data were recorded before and after 6 months of treatment. Results were converted to SDS values when appropriate. Both groups had comparable baseline data. The growth response was significantly higher in the high-dose group after 6 months of treatment. This group had significantly higher GH levels overnight, whether considering area under the curve, mean GH, or maximal GH levels. IGF levels increased in the low-dose group from –1.6 to 0.2 SD, while the change in the high-dose group was from –1.6 to 1.5 SD. In the latter, 74% of the children had IGF-I levels in the highest quartile (>0.84 SD) and 37% had levels above +2SD, compared with only 19% and 6% respectively, in children treated with the lower dose of GH treatment.

The short SGA children given the high dose of GH had evidence of higher circulating GH and IGF-I levels. The IGF-I/IGFBP3 ratio was also more elevated, suggesting higher values of circulating free IGF-I. A higher stimulation of the growth control axis for 6 months produced a significant increase in height, +0.7 SD as compared to +0.5 in the low-dose group. Of interest is that no correlation could be found between the growth response and the increase of GH or IGF-I/IGFBP3 levels. The authors suggested that this may reflect a reduced GH/IGF-I receptor sensitivity, eventually related to the SGA condition, and they recommended monitoring IGF-I levels during GH treatment to ensure that these remain within the normal range for age.

van Dijk M, Mulder P, Houdijk M, et al. High serum levels of growth hormone (GH) and insulin-like growth factor-I (IGF-I) during high-dose GH treatment in short children born small for gestational age. J Clin Endocrinol Metab. 2006;91:1390–1396.

First Editor’s Comment

This is the first study comparing 2 doses of GH in short children with SGA and the effect on growth and IGF-I levels. The issue is important since there is a theoretical risk of cancer after prolonged exposure to higher circulating levels of IGF-I and IGF-I/IGFBP3 ratio. This has led to repeated recommendations for the evaluation of circulating IGF-I levels, at least yearly, during GH treatment. The present data document precisely the effect of the 2 most frequently prescribed doses of GH and provide unique data for an appropriate comparison at 6 months of treatment. However, the study does not elucidate the question whether the dose-related increase of IGF values remains the same at a later time, when the high-dose GH does not produce a higher growth rate anymore.

In any case, the authors challenge the recommended doses and focus on the efficacy of the medication. This paper should be read in relation to the previous study 1 by the same group showing a moderate, but not significant, increase in final height, when doubling the dose up to 0.067 mg/kg/d. It was shown that height gain was dependent on fewer amount of doses over the long-term than over the short-term. Therefore, the economical aspects of long-term administration of higher doses of GH should be considered. Interestingly, the present data again confirmed that during the first 6 months of treatment, there is a significant GH-dose effect allowing better and faster catch-up growth.

Even if a different approach is chosen by individualizing GH treatment to optimize height gain, one may still expect difficulties in adjusting GH dose when taking into account multiple factors such as differences between initial and later treatment periods, dose-related effects on IGF-I, individual susceptibility, and poor correlation between height gain and changes in IGF-I score. It will be of interest to determine whether long-term GH dose adjustments will cope with the observed changes of IGF-I levels and the need for maintaining them within safe limits.

Raphaël Rappaport, MD

Second Editor’s Comment

Although most SGA children show catch-up growth and achieve normal height during the first 2 years of life, approximately 10% to 15% of them remain short with a height below –2 SDS. Recent studies have demonstrated that GH treatment of short children born SGA results in the normalization of height during childhood. Van Pareren demonstrated that long-term treatment of short SGA children with a low dose of GH (1 mg/m 2/d, 0.033 mg/kg/d) was as effective in attaining a normal final height as the treatment with a high dose of GH (2 mg/m 2/d, 0.067 mg/kg/d).¹ In this study van Dijk et al showed that most SGA patients receiving a high dose of GH treatment had high GH levels for most of the day and IGF-I levels and IGF-I/IGFBP3 ratios in the upper quintile. In recent years, concern regarding the detrimental effects of persistent high serum GH and IGF-I levels has been expressed in various studies. Of particular importance are the reports of an increased cancer risk (ie, breast, prostate, and colon cancer) in patients with IGF-I levels in the upper tertile to quintile, more so if accompanied by low IGFBP3 levels.^2-4 Recent studies have recommended beginning GH treatment of short SGA children at an early age.^2-4 Thus, GH and IGF-I levels may be elevated in many of these patients for a good part of childhood and adolescence, possibly placing them at an increased risk for complications later in their lives. The long-term deleterious effects of GH treatment in SGA children remain unknown. However, the use of an initially lower GH dose, which can then be individually adjusted and the monitoring of IGF-I and IGFBP3 during GH therapy, in an attempt to maintain IGF-I concentrations in the upper half of the age-adjusted reference range, is strongly recommended.

Roberto Lanes, MD

References - (linked to )

Last Updated: 04/30/2008