This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bakker, B. Articles by Rosenfeld, R. G. Search for Related Content PubMed PubMed Citation Articles by Bakker, B. Articles by Rosenfeld, R. G. Pubmed/NCBI databases Substance via MeSH Medline Plus Health Information Growth Disorders Related Collections Neuroendocrinology and Pituitary Pediatric Endocrinology

Height Velocity Targets from the National Cooperative Growth Study for First-Year Growth Hormone Responses in Short Children

Genentech, Inc. (B.B., B.L.), South San Francisco, California 94080; Consultant (J.F.), Santa Monica, California 90403; Saint Barnabas Medical Center (H.A.), Livingston, New Jersey 07039; and The Lucile Packard Foundation for Children’s Health (R.G.R.), Palo Alto, California 94301

Address all correspondence and requests for reprints to: Bert Bakker, Genentech, Inc., 1 DNA Way, mailstop 454 B, South San Francisco, California 94080. E-mail: bakker.bert{at}gene.com.

	Abstract

Top Abstract Introduction Patients and Methods Results Discussion References

 
Context: Although GH has been used to treat short stature in GH deficiency (GHD) and other conditions for more than 40 yr, criteria for satisfactorily defining targets for GH responsiveness have never been developed.

Objective: The objective of this study was to present the first-year growth expressed as height velocity (HV) for prepubertal boys and girls with idiopathic GHD, organic GHD, idiopathic short stature, or Turner syndrome from Genentech’s National Cooperative Growth Study to derive age-specific targets for GH responsiveness for each etiology and gender.

Design and Population: Using data from the National Cooperative Growth Study, we constructed curves of response to GH during the first year of treatment with standard daily doses in naive-to-treatment prepubertal children with idiopathic GHD (2323 males, 842 females), organic GHD (582 males, 387 females), idiopathic short stature (1392 males, 465 females), or Turner syndrome (1367 females).

Main Outcome Measure: For each category, mean pretreatment and mean ±1 and ±2 SD for the first-year HV on GH were assessed. Mean and mean ± 1 SD for HV were plotted vs. age at baseline (initiation of GH treatment) and compared with mean pretreatment HV.

Results: HV plots for each category as a factor of age at baseline are presented. Mean – 2 SD HV plots approximated the pretreatment HV.

Conclusion: Using baseline age- and gender-specific targets will assist clinicians in assessing a patient’s first-year growth response. We propose that HV below the mean – 1 SD on these plots be considered a "poor" response. These curves may be used to identify patients who may benefit from GH dose adjustment, to assess compliance issues, or to challenge the original diagnosis.

	Introduction

Top Abstract Introduction Patients and Methods Results Discussion References

 
Growth failure is difficult to define. Although there are a number of systems for the classification of growth disorders, they are problematic because diagnostic categories have not always been clearly defined and frequently overlap (1).

A clear definition of growth response after intervention with therapies such as GH is also lacking. Although GH has been used for treating short stature in GH deficiency (GHD) and other conditions for more than 40 yr, criteria for defining satisfactory GH response targets have never been developed. The range of GH response is large; differences can be attributed to diagnosis, age, GH dose, parental height (Ht), compliance, intercurrent illness, other (endocrine) therapies, and still poorly defined molecular and biochemical factors that may include the structure and concentration of GH receptors, the robustness of thepostreceptor signaling cascade, IGF-I transcriptional and translational efficiency, and epiphyseal responsiveness to GH, IGF-I, and other factors.

Several attempts have been made to define and predict growth response. More than 20 yr ago, Frasier et al. (2) published a dose-response curve for human GH (hGH) based on observed responses. Dose responsiveness has also been looked at in more recent studies using higher GH dosages (3, 4, 5). In addition to dose-response models and reports, various prediction models were developed (6, 7, 8). Of these, the models described by Ranke et al. based on data from the Kabi International Growth Study registry are best known. Ranke’s first mathematical model predicted the growth response of prepubertal children with idiopathic GHD (IGHD) to exogenous recombinant hGH (7). Other prediction models for different etiologies, such as Turner syndrome (TS) and small for gestational age followed, suggesting that the models would help identify the factors determining the observed variability of GH sensitivity (9, 10). Such models of predicting the growth response to hGH do not completely account for the variability observed in GH responsiveness; mostly, they predict around 50% and, at best, 70% of the variability (9), and clearly evidence-based criteria for better defining GH responsiveness are needed. To develop such evidence-based reference data and to aid clinicians in evaluating GH responsiveness, we used data from the Genentech National Cooperative Growth Study (NCGS) to construct plots of growth response during the first year of treatment with standard daily GH doses in prepubertal children, aged 2–14 yr at onset of GH treatment. Target growth responses for the purposes of this paper were defined as height velocity (HV) in the first year of GH treatment; change in Ht SD score (SDS) and HV SDS in the first year of GH treatment were also examined to assess the relative merits of their use vs. HV as the target.

The NCGS was initiated in 1985 after the withdrawal of pituitary-derived GH because of Creutzfeldt-Jakob disease and in response to a request by the Food and Drug Administration to monitor the efficacy and safety of recombinant hGH. Since that time, the NCGS registry has grown to become the largest and longest-standing repository of GH-related data in North America, containing demographic and outcomes data from more than 54,000 children with growth-related disorders and encompassing more than 180,000 patient years of experience (11). The NCGS provides an opportunity to evaluate the treatment trends over time across several childhood growth disorders.

Here, we use growth data for the first year of GH treatment from 4297 prepubertal boys and 3061 girls with IGHD, organic GHD (OGHD), idiopathic short stature (ISS), or TS to develop age-, sex-, and diagnosis-specific growth targets of responsiveness.

	Patients and Methods

Top Abstract Introduction Patients and Methods Results Discussion References

 
Using data from the NCGS, we studied first-year growth response to GH in four common growth disorders. The first-year HV was determined using the time point at which follow-up Ht was measured that was closest to 1 yr after the initiation of GH therapy provided that the time point was within 90 d of 1 yr. The HV was computed as an annualized HV (centimeters per year). Dose was computed as the average dose during the time over which the HV was computed. We selected previously untreated subjects with Ht SDS less than –2 at onset of GH treatment and with ages between 2 and 14 yr at onset of GH treatment. Baseline is defined as the age at initiation of GH treatment with a daily or six times weekly injection schedule. Patients were prepubertal at the end of the first year of GH treatment, i.e. boys with testes volume more than 3 ml, girls with Tanner breast stage of at least 2, and patients with Tanner pubic hair stage of at least 2 were excluded. If the testes volume for a boy at least 13 yr old was unknown or if the breast stage for a girl at least 11 yr old was unknown, the patient was excluded from analysis. From among these subjects, we further selected subjects who had been identified by the NCGS investigators as having IGHD, OGHD, ISS, or TS.

This research was conducted in accordance with the guidelines in The Declaration of Helsinki and was formally approved by the appropriate institutional review committees or its equivalent. This approval and informed consent were obtained from all subjects.

For each gender and etiology, mean pretreatment and mean ± 1 SD for first-year HV on GH were plotted against subject age at onset of GH treatment. Like all the data, the pretreatment HV data are cross-sectional. A smooth curve was fitted as a function of baseline age through the HVs of individual subjects using nonparametric loess regression (12), which generates a smooth curve defining the mean response at each age. The PROC LOESS procedure in SAS (SAS Institute, Cary, NC) was used to perform these analyses.

The absolute values of the residuals from this first fit were computed. Next, these residuals were fit vs. age, again using SAS loess. Last, the SD was taken as the square root of /2 times this second fit (using the relationship between the mean of a half-normal distribution with the SD of the corresponding normal distribution).

Ht SDS were computed using the method of Kuczmarski et al. (13) from the US Centers for Disease Control 2000 growth charts, and HV SD scores were computed using data from Tanner et al. (14). The same statistical curve-fitting methodology was used to obtain the curves for the change in Ht and HV SDS.

	Results

Top Abstract Introduction Patients and Methods Results Discussion References

 
Sample sizes for fitting curves were as shown in Table 1. The 10th, 50th, and 90th percentiles for dose during the first year are shown in Table 2. Doses were generally close to 0.30 mg/kg·wk, except for TS, in which 0.35 mg/kg·wk was typical.

View larger version (31K): [in this window] [in a new window]   FIG. 1. Points represent first-year growth responses to daily GH expressed as HV at age of treatment onset (x-axis) in naive, prepubertal IGHD males.

View larger version (25K): [in this window] [in a new window]   FIG. 2. First-year growth responses to daily GH expressed as HV at age of treatment onset (x-axis) in naive, prepubertal IGHD males. Data given for mean and mean ± 1 and – 2 SD.

View larger version (43K): [in this window] [in a new window]   FIG. 3. First-year growth responses to daily GH expressed as HV at age of treatment onset (x-axis) in naive, prepubertal females and males with IGHD (A, B), OGHD (C, D), and ISS (E, F), and females with TS (G). Data given for mean and mean ± 1 SD.

View larger version (18K): [in this window] [in a new window]   FIG. 4. First-year growth responses to daily GH expressed as change in Ht SDS at age of treatment onset (x-axis) in naive, prepubertal IGHD males.

View larger version (23K): [in this window] [in a new window]   FIG. 5. First-year growth responses to daily GH expressed as HV SDS at age of treatment onset (x-axis) in naive, prepubertal IGHD males.

Figure 3 shows the mean ± 1 SD HV during the first year of GH therapy, as well as the pretreatment mean HV for females and males with IGHD (Fig. 3, A and B), OGHD (Fig. 3, C and D), ISS (Fig. 3, E and F), and females with TS (Fig. 3G). The first-year mean HV was higher at younger ages. The SD of the HV varied only somewhat with age at baseline, with more variability for females than males.

Using the same male IGHD patients as in Fig. 2, Fig. 4 shows the same mean and mean ± 1 SD curves for change in Ht SDS, and Fig. 5 does the same for HV SDS. Compared with the HV, there is greater variation in means and SDs with respect to baseline age, especially for the HV SDS.

	Discussion

Top Abstract Introduction Patients and Methods Results Discussion References

 
Clinical experience shows there is considerable variability in response to therapy with GH in children with short stature, at least with a standard weight-based dosing schedule (15, 16). We still do not completely understand why children with GHD, for whom therapy is directed at replacing absent GH, should exhibit such a wide range in response, which persists even when we take into account such variables as statural deficiency before therapy, bone age, midparental Ht, and duration of treatment (17). For children with IGHD, variability in growth response reflects, at least in part, the well-documented inaccuracies inherent in the diagnosis. For children with OGHD, variability in response certainly reflects the impact of radiotherapy, chemotherapy, and chronic illness. In TS, variability of response is commonly observed and inadequately understood and could reflect differences in the degree of target tissue SHOX (short stature homeobox) deficiency inherent in a heterozygous genetic condition (18). Finally, with ISS, it is commonly accepted that this diagnosis encompasses a heterogeneous group of disorders, and a spectrum of responses to GH therapy should not be surprising. Recognition of the wide range of clinical responsiveness to GH should serve to challenge the traditional, weight-based GH dosing that has been in use for more than 40 yr (19). The evidence-based criteria we present here take the next step in helping to define growth responses to GH in several childhood disorders.

First-year responses to GH in four different groups of patients were derived from the very large postmarketing surveillance database NCGS. The first point to note, as seen in Fig. 2 for IGHD males, is that the mean pretreatment HV curve approximates the curve for the first-year mean – 2 SD for HV for all of the conditions evaluated. This indicates that children with HV at the mean – 2 SD during the first year of GH therapy are not exhibiting the expected improvement in growth velocity. If this is observed, it prompts a number of questions, such as the correctness of the diagnosis, compliance issues, etc. We will address these issues toward the end of Discussion.

As seen in Fig. 3, mean HV varies similarly across etiologies with age at baseline. Hence, HV needs to be interpreted in the context of age at baseline (as well as of etiology of growth failure and gender). The similarity of patterns across etiologies as well as HVs over the baseline age range emphasizes the importance of age at initiation of treatment.

As seen in Fig. 4, the mean and SD of the change in Ht SDS both differ substantially with age. Hence, the interpretation of the change in Ht SDS during the first year of treatment, as shown, and subsequently (data not shown) is highly dependent on age (as well as etiology and gender). Interpretation of the pretreatment change in Ht SDS in males with IGHD is also highly dependent on age, as shown in Fig. 4 by the dashed curve for the mean annualized change in Ht SDS before treatment began. Generally, these children were decreasing in Ht SDS before treatment, but the rate of decrease varied substantially with age. Similar results are seen for females and for other etiologies of short stature. Thus, a 1-yr increase in Ht SDS of 0.5 may be acceptable in patients greater than 10 yr of age but would be considered poor in younger children. The larger mean and SD at younger ages is the result of the fact that the SD of Ht is smaller at younger ages. For example, an increase in Ht during the first treatment year of 10 cm at a young age results in a larger change in Ht SDS than the same 10 cm at an older age. This observation on change in Ht SDS has to be taken into consideration when one compares two cohorts in which the age at onset of GH treatment differs considerably.

As seen in Fig. 5, HV SDS depends even more on age at baseline than the change in Ht SDS. Hence, the need is even greater for interpreting the change in HV in a child during treatment in the context of the general experience of children of the same etiology, age, and gender. The major problem with HV SDS as a primary efficacy endpoint is that it does not take into consideration the disturbances in the timing of puberty that are often seen in the etiologies discussed in this article.

These first-year GH-treatment HV curves (as well as the curves for change in Ht SDS and HV SDS) should be viewed as conservative for a number of reasons.

For example, there is an unknown amount of noncompliance in all groups. There are also diagnostic challenges such as distinguishing between some forms of IGHD (when GHD is isolated and not part of idiopathic multiple pituitary hormone deficiency) and ISS that make the IGHD group more heterogeneous. These issues could inflate the SDs for each of the response endpoints and lower the means for the IGHD group. Thus, the computed mean – 1 SD and mean – 2 SD curves could be lower than what the curves would be without these problems, especially for IGHD. Hence, if a child’s on-treatment first year HV is below the on-treatment mean – 1 SD curve, then effort should be made to explore possible causes of poor response to GH treatment. As mentioned previously, we suggest using the mean – 1 SD instead of, for example, the mean – 2 SD because, in all cases (as can be observed in Fig. 2), the mean – 2 SD is very close to the pretreatment HV curve and would hardly be considered an improvement.

These curves can serve as a practical tool for clinicians to use when comparing the first-year HV of specific patients to the HV of a large population in the NCGS database. If the HV in the first year of treatment is considered "substandard," as defined here, one should reconsider whether 1) the diagnosis is correct, 2) there is comorbidity, such as undiagnosed underlying illness, 3) compliance is adequate, or 4) undetermined factors are affecting the clinical response to GH, such as defects of GH and/or IGF sensitivity. Results for HV are notably more homogeneous with age than change in Ht SDS or HV SDS. Based on the analyses presented, we suggest that the clinician use HV as the primary efficacy endpoint for evaluating treatment for short stature based on age at baseline, gender, and etiology.

In summary, the use of age at baseline-, gender-, and etiology-specific first-year growth response curves offer the clinician a benchmark against which to assess the progress of an individual patient and can contribute to evidence-based decision making to maximize the efficacy of GH treatment.

	Acknowledgments

 
We thank Joan Jacobs for the critical reading of this manuscript and her contributions to the discussion. Genentech, Inc., provided writing assistance for this manuscript.

	Footnotes

 
Disclosure Statement: B.B. and B.L. are employees of Genentech and hold stock with Genentech. J.F. received consulting fees from Genentech, Tercica, Fibrogen, DiObex, Theravance, and Acumen Sciences. H.A. and R.G.R. received consulting fees from Genentech.

First Published Online November 13, 2007

Abbreviations: GHD, GH deficiency; hGH, human GH; Ht, height; HV, height velocity; IGHD, idiopathic GH deficiency; ISS, idiopathic short stature; NCGS, National Cooperative Growth Study; OGHD, organic GH deficiency; TS, Turner syndrome.

Received July 17, 2007.

Accepted November 5, 2007.

	References

Top Abstract Introduction Patients and Methods Results Discussion References

 

1. Rosenfeld RG, Cohen P 2002 Disorders of growth hormone/insulin-like growth factor secretion and action. In: Sperling MA, ed. Pediatric endocrinology. 2nd ed. Philadelphia: Saunders; 211–288.
2. Frasier SD, Costin G, Lippe BM, Aceto Jr T, Bunger PF 1981 A dose-response curve for human growth hormone. J Clin Endocrinol Metab 53:1213–1217[Abstract/Free Full Text]
3. Cohen P, Bright GM, Rogol AD, Kappelgaard AM, Rosenfeld RG 2002 Effects of dose and gender on the growth and growth factor response to GH in GH-deficient children: implications for efficacy and safety. J Clin Endocrinol Metab 87:90–98[Abstract/Free Full Text]
4. van Pareren YK, de Muinck Keizer-Schrama SM, Stijnen T, Sas TC, Jansen M, Otten BJ, Hoorweg-Nijman JJ, Vulsma T, Stokvis-Brantsma WH, Rouwe CW, Reeser HM, Gerver WJ, Gosen JJ, Rongen-Westerlaken C, Drop SL 2003 Final height in girls with turner syndrome after long-term growth hormone treatment in three dosages and low dose estrogens. J Clin Endocrinol Metab 88:1119–1125[Abstract/Free Full Text]
5. Wit JM, Rekers-Mombarg LT, Cutler GB, Crowe B, Beck TJ, Roberts K, Gill A, Chaussain JL, Frisch H, Yturriaga R, Attanasio AF 2005 Growth hormone (GH) treatment to final height in children with idiopathic short stature: evidence for a dose effect. J Pediatr 146:45–53[CrossRef][Medline]
6. de Ridder MA, Stijnen T, Hokken-Koelega AC 2006 Prediction of adult height in growth hormone treated children with growth hormone deficiency. J Clin Endocrinol Metab 92:925–931[CrossRef][Medline]
7. Ranke MB, Lindberg A, Chatelain P, Wilton P, Cutfield W, Albertsson-Wikland K, Price DA 1999 Derivation and validation of a mathematical model for predicting the response to exogenous recombinant human growth hormone (GH) in prepubertal children with idiopathic GH deficiency. KIGS International Board. Kabi Pharmacia International Growth Study. J Clin Endocrinol Metab 84:1174–1183[Abstract/Free Full Text]
8. Wikland KA, Kristrom B, Rosberg S, Svensson B, Nierop AF 2000 Validated multivariate models predicting the growth response to GH treatment in individual short children with a broad range in GH secretion capacities. Pediatr Res 48:475–484[Medline]
9. Ranke MB, Lindberg A, Chatelain P, Wilton P, Cutfield W, Albertsson-Wikland K, Price DA 2000 Prediction of long-term response to recombinant human growth hormone in Turner syndrome: development and validation of mathematical models. KIGS International Board. Kabi International Growth Study. J Clin Endocrinol Metab 85:4212–4218[Abstract/Free Full Text]
10. Ranke MB, Lindberg A, Cowell CT, Wikland KA, Reiter EO, Wilton P, Price DA 2003 Prediction of response to growth hormone treatment in short children born small for gestational age: analysis of data from KIGS (Pharmacia International Growth Database). J Clin Endocrinol Metab 88:125–131[Abstract/Free Full Text]
11. Kemp SF 2005 Growth hormone treatment of idiopathic short stature: history and demographic data from the NCGS. Growth Horm IGF Res 15(Suppl A):S9–S12
12. Cleveland W, Grosse E 1991 Computational methods for local regression. Stat Comput 1:47–52[CrossRef]
13. Kuczmarski R, Ogden C, Guo S 2002 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11:1–90
14. Tanner JM, Whitehouse RH, Takaishi M 1966 Standards from birth to maturity for height, weight, height velocity, and weight : British children, 1965. I. Arch Dis Child 41:454–471[Free Full Text]
15. Blethen SL, Baptista J, Kuntze J, Foley T, LaFranchi S, Johanson A 1997 Adult height in growth hormone (GH)-deficient children treated with biosynthetic GH. The Genentech Growth Study Group. J Clin Endocrinol Metab 82:418–420[Abstract/Free Full Text]
16. Carel JC, Ecosse E, Nicolino M, Tauber M, Leger J, Cabrol S, Bastie-Sigeac I, Chaussain JL, Coste J 2002 Adult height after long term treatment with recombinant growth hormone for idiopathic isolated growth hormone deficiency : observational follow up study of the French population based registry. BMJ 325:70
17. Rosenfeld RG 2006 Editorial: the pharmacogenomics of human growth. J Clin Endocrinol Metab 91:795–796[Free Full Text]
18. Blaschke RJ, Topfer C, Marchini A, Steinbeisser H, Janssen JW, Rappold GA 2003 Transcriptional and translational regulation of the Leri-Weill and turner syndrome homeobox gene SHOX. J Biol Chem 278:47820–47826[Abstract/Free Full Text]
19. Cohen P, Rogol AD, Howard CP, Bright GM, Kappelgaard AM, Rosenfeld RG 2007 IGF-based dosing of GH therapy in children: a randomized controlled study. J Clin Endocrinol Metab 92:2480–2486[Abstract/Free Full Text]

This article has been cited by other articles:

				P. Cohen, A. D. Rogol, C. L. Deal, P. Saenger, E. O. Reiter, J. L. Ross, S. D. Chernausek, M. O. Savage, J. M. Wit, and on behalf of the 2007 ISS Consensus Workshop parti Consensus Statement on the Diagnosis and Treatment of Children with Idiopathic Short Stature: A Summary of the Growth Hormone Research Society, the Lawson Wilkins Pediatric Endocrine Society, and the European Society for Paediatric Endocrinology Workshop J. Clin. Endocrinol. Metab., November 1, 2008; 93(11): 4210 - 4217. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bakker, B. Articles by Rosenfeld, R. G. Search for Related Content PubMed PubMed Citation Articles by Bakker, B. Articles by Rosenfeld, R. G. Pubmed/NCBI databases Substance via MeSH Medline Plus Health Information Growth Disorders Related Collections Neuroendocrinology and Pituitary Pediatric Endocrinology