Objective: A number of long-term research studies are in progress to evaluate the effects of treatment with GH on growth and final height in children with short stature but no demonstrable abnormality of GH secretion. Such treatment is invasive, expensive and carries some risk to the child. An early indication of growth response would allow restriction of treatment to those children most likely to benefit, but anthropometric measurements are relatively subjective, insensitive and imprecise. The aim of this study was to evaluate bone alkaline phosphatase, procollagen Type I C-terminal propeptide, procollagen Type III N-terminal propeptide and the cross-linked carboxy-terminal telopeptide of Type I collagen as early biochemical predictors of height velocity response to growth-promoting treatments in short normal children.
Design: A prospective intervention study, partially placebo controlled on a double blind basis.
Patients: Fifty healthy children with familial short stature or constitutional delay in growth and puberty (8 girls, 42 boys, ages 5.5-16.5 years and all either prepubertal (45) or in very early puberty (5 boys) at the start of treatment) were treated with placebo (6), GH alone (32), GH plus oxandrolone (8) or GH plus testosterone (4).
Measurements: Bone alkaline phosphatase and the collagen markers were measured at the start of treatment and 3 months later. Height velocity was calculated at the start of treatment and again after one year.
Results: Pre-treatment biochemical marker concentrations did not predict height velocity response after one year. Increments in all markers after 3 months were significantly correlated with height velocity increments after one year of treatment, the highest correlations being observed for bone alkaline phosphatase (r = 0.67, P < 0.0001) and procollagen Type III N-terminal propeptide (r = 0.57, P < 0.0001). Highly significant correlations (P < 0.0001) were also observed between bone alkaline phosphatase and procollagen Type I C-terminal propeptide (r = 0.55) and between procollagen Type III N-terminal propeptide and the cross-linked carboxy-terminal telopeptide of Type I collagen (r = 0.62). Multiple linear regression with stepwise selection of variables identified bone alkaline phosphatase and procollagen Type III N-terminal propeptide as the only two independent variables that contributed significantly to the prediction of height velocity response after one year (analysis of variance, P < 0.0001). Together they predicted 59% of the variability in height velocity response after a year.
Conclusions: The best early predictors of height velocity response were bone alkaline phosphatase (a protein found in hypertrophic chondrocytes in the epiphyseal growth plate, in calcifying matrix vesicles and in mature osteoblasts) and procollagen Type III N-terminal propeptide, a marker of interstitial fibril biosynthesis in soft tissues. Using these markers, GH treatment could be targeted to those children most likely to benefit in the medium term.