Insulin growth factor-based dosing of growth hormone therapy in children: a randomized, controlled study

Abstract

Context: Weight-based dosing of GH is the standard of care for short children, although IGF-I is thought to be the main mediator of GH actions on growth.

Objective: The objective of the study was to test whether IGF-I levels achieved during GH therapy are determinants of the growth responses to GH treatment.

Design: This was a 2-yr, open-label, randomized, IGF-I concentration-controlled trial. Prepubertal short children [n = 172, mean age 7.53 yr, mean height sd score (HT-SDS) -2.64] with low IGF-I levels (mean IGF-I SDS -3.56) were randomized to receive one of two GH dose-titration arms in which GH dosage was titrated to achieve an IGF-I SDS at the mean [IGF((low)) group, n = 70] or the upper limit of the normal range [+2 SDS, IGF((high)) group, n = 68] or to a comparison group of conventional GH dose of 40 microg/kg/d (n = 34).

Setting: The study was conducted in a multicenter, outpatient setting.

Primary outcome measure: Change in HT-SDS over 2 yr was measured.

Results: One hundred forty-seven patients completed the trial. Target IGF-I levels were achieved in the dose-titration arms within 6-9 months. The changes in HT-SDS were +1.0, +1.1, and +1.6 for conventional, IGF((low)), and IGF((high)), respectively, with IGF((high)) showing significantly greater linear growth response (P < 0.001, compared with the other two groups). The IGF((high)) arm required higher doses (>2.5 times) than the IGF((low)) arm, and these GH doses were highly variable (20-346 microg/kg/d). Multivariate analyses suggested that the rise in the IGF-I SDS significantly impacted height outcome along with the GH dose and the pretreatment peak-stimulated GH level.

Conclusion: IGF-I-based GH dosing is clinically feasible and allows maintaining serum IGF-I concentrations within the desired target range. Titrating the GH dose to achieve higher IGF-I targets results in improved growth responses, although at higher average GH doses.