Collagen metabolites in the prediction of response to GH therapy in short children P Tapanainen, M Knip, L Risteli 1 , L Kemppainen, M-L Ka ¨a ¨r 2 and J Risteli 1 Departments of Pediatrics and 1 Medical Biochemistry and Clinical Chemistry, University of Oulu, Oulu, Finland and 2 Central Hospital of Vaasa, Vaasa, Finland (Correspondence should be addressed to P Tapanainen, Department of Pediatrics, University of Oulu, FIN-90220 Oulu, Finland) Abstract To evaluate the role of collagen metabolites in the prediction of the response to GH treatment we measured the serum concentrations of the C-terminal propeptide of type I procollagen (PICP) and the N-terminal propeptide of type III procollagen (PIIINP) with specific RIAs in 35 short children (16 boys) before and after 5 days, 5 weeks and 3 months of GH therapy. The mean ageof the children was 10.3 years (range 1.9–16.4 years) and the bone age ranged from 1.2 to 12.5 years (mean 7.6 years). The initial mean relative height (RHI) was ¹3.6 SDS (range ¹6.6 to ¹2.4 S.D.). Nineteen children were found to have GH deficiency (GHD; peak GH responses in two pharmacological tests <10 mg/l), while the remaining 16 were considered to have undefined short stature (USS). The children were treated with recombinant human GH (0.1 U/kg given subcutaneously at bedtime 6–7 times/week). The increases in RHI over the first 6 and 12 months of therapy were used as response measures. There was already a significant increase (P < 0:001) in both the serum PICP and PIIINP levels at 5 days, and the concentrations continued to rise up to 3 months, PICP levels rising less than the PIIINP levels. In the whole group the RHI over 6 months correlated most strongly with the absolute PICP concentrations at 3 months (r s ¼ 0.59; P < 0:05), while the absolute PIIINP concentrations at 3 months showed the strongest relation to the one year RHI (r s ¼ 0.69; P < 0:001). In the GHD group the 6 month RHI was most strongly related to the absolute PICP concentration at 3 months (r s ¼ 0.59; P < 0:05). In the USS group the absolute PICP concentrations at 3 months correlated most strongly with the one year RHI (r s ¼ 0.82; P < 0:01). Significant correlations were also observed between the absolute PIIINP levels at 3 months and the 6 month RHI (r s ¼ 0.60; P < 0:05) and 12 month RHI (r s ¼ 0.76; P < 0:01) in this group. These results show that GH therapy results in an unequivocal increase in circulating concentrations of PICP and PIIINP. The serum PICP and PIIINP concentrations may be of value in the prediction of the long-term response to GH therapy. European Journal of Endocrinology 137 621–625 Introduction The increased availability of biosynthetic human growth hormone (GH) and the possibility of using GH for treatment of a variety of growth disorders has increased the need for biochemical predictors of the response to growth-stimulating therapy. Present knowl- edge indicates that the GH-induced insulin-like growth factor-I (IGF-I) response may be a useful predictor of response to therapy in children with undefined short stature (USS) but not in GH-deficient (GHD) children (1, 2). A preliminary report suggests, however, that pretreatment serum IGF-I concentrations correlate inversely with growth response to GH in patients with hypopituitarism, if the IGF-I levels are measured with appropriate methodology (3). Serum concentrations of the N-terminal propeptide of type III collagen (PIIINP) have been found to be related to growth rate in healthy children (4). We and others have previously shown that the measurement of serum PIIINP, particularly in children receiving exogenous GH, may be a potential early indicator of the growth response to therapy and may provide useful additional biochemical information during growth (5–7). Serum concentrations of the C- terminal propeptide of type I collagen (PICP) have also been reported to be abnormally low in GHD children (8, 9), and there is some evidence that serum PICP could be more useful in assessing growth than serum PIIINP (8, 10). To evaluate the collagen metabolites in the prediction of the response to GH treatment, we measured the serum concentrations of PIIINP and PICP with specific RIAs in short children treated with GH. Materials and methods Study subjects Thirty-five children, 16 boys and 19 girls, aged 1.9–16.4 years, were studied. All had been referred to the European Journal of Endocrinology (1997) 137 621–625 ISSN 0804-4643  1997 Society of the European Journal of Endocrinology