Characteristics of children with the best and poorest first- and second-year growth during rhGH therapy: data from 25 years of the Genentech national cooperative growth study (NCGS)

Background Models assessing characteristics contributing to response to recombinant human growth hormone (rhGH) response rarely address growth extremes in both years 1 and 2 or examine how children track from year to year. Using National Cooperative Growth Study (NCGS) data, we determined characteristics contributing to responsiveness to rhGH and the pattern of change from years 1 to 2. Patients and methods Height velocity standard deviation score (HV SDS) for 2 years for prepubertal children with idiopathic GH deficiency (IGHD) (n = 1899) and idiopathic short stature (ISS) (n = 1186) treated with similar doses for two years were computed. Group 1 = HV SDS < −1; 2 = HV SDS −1 to +1; 3 = HV SDS > +1. Results For IGHD , mean age was 7.5 years and similar in all groups. Year 1 HV SDS was associated with greater body mass index (BMI) SDS, lower pre-treatment HV, baseline height SDS, greater target height SDS minus height SDS, and lower maximum stimulated GH ( P <0.0001). Year 2, 172/271 (73%) in group 1 moved to either group 2 (n = 156) or 3 (n = 16). Year 2 HV SDS was associated with greater year 1 HV SDS ( r = 0.045, P <0.0001), greater BMI SDS, taller parents and lower peak GH. For ISS , year 1 HV SDS was associated with greater BMI SDS and lower pre-treatment HV ( P ≤0.0001). 109/169 (64%) in group 1 moved to group 2 (n = 90) or group 3 (n = 19). Greater year 2 HV SDS was related to year 1 HV SDS ( r = 0.27, P <0.0001). Conclusion For IGHD, multiple characteristics contributed to best first-year response but for ISS, best first-year HV SDS was associated only with BMI SDS and inversely with pre-treatment HV. For both GHD and ISS, year 1 HV SDS was not a strong enough predictor of year 2 HV SDS to use first-year HV alone to determine GH continuation.