Best Practice & Research Clinical Endocrinology & Metabolism xxx (2015) 1e14
Contents lists available at ScienceDirect
Best Practice & Research Clinical Endocrinology & Metabolism journal homepage: www.elsevier.com/locate/beem
2
Hormone replacement therapy in children: The use of growth hormone and IGF-I €ffle, Prof. Dr. med., MD Roland Pfa University Children's Hospital Leipzig, Liebigstr. 20a, 0413 Leipzig, Germany
a r t i c l e i n f o Article history: Available online xxx Keywords: recombinant human growth hormone recombinant human IGF-I growth hormone deficiency small for gestational age (SGA) Turner's syndrome treatment efficacy longterm side effects
Recombinant human GH (rhGH) has been available since 1985. This article gives an overview, what has been achieved over the past 30 years in respect to optimization of rhGH treatment for the individual child with GH deficiency and what are the safety issues concerned with this treatment. In the last twenty years significant scientific progress has been made in the diagnosis of GH deficiency, the genetic disorders that are associated with pituitary GH deficiency and the genetics that influence growth in general. On the other hand rhGH is not only used in states of GH deficiency but also various conditions without a proven GH deficiency by classical standards. Clinical studies that investigated both the genetics of growth and the individual responses to rhGH therapy in these patient populations were able to refine our concept about the physiology of normal growth. In most patients under rhGH treatment there is a considerable short-term effect, however the overall gain in growth obtained by a long-term treatment until final height still remains a matter of debate in some of the conditions treated. Also first studies on the long-term safety risks of rhGH treatment have raised the question whether this treatment is similarly safe for all the patient groups eligible for such a treatment. Therefore even in the face of a longstanding safety record of this drug replacement therapy the discussion about the right cost and risk to benefit ratio is continuing. Consequently there is still a need for carefully conducted long-term studies that use modern anthropometric, genetic, and laboratory techniques in order to
Abbreviations: EMA, European Medical Agency; FDA, United States Federal Drug Association; GHD, growth hormone deficiency; GHIS, growth hormone insensitivity syndrome; GHR, growth hormone receptor; IGFBP-3, IGF binding protein 3; ISS, idiopathic short stature; PIGFD, primary IGF-I deficiency; PTPN11, protein tyrosine phosphatase, nonreceptor type 11; rhGH, recombinant human GH; rhIGF-I, recombinant human IGF-I; SDS, SD score; SGA, small for gestational age. E-mail address: rpfaeffl[email protected]. http://dx.doi.org/10.1016/j.beem.2015.04.009 1521-690X/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Pf€affle R, Hormone replacement therapy in children: The use of growth hormone and IGF-I, Best Practice & Research Clinical Endocrinology & Metabolism (2015), http:// dx.doi.org/10.1016/j.beem.2015.04.009
€ ffle / Best Practice & Research Clinical Endocrinology & Metabolism xxx (2015) 1e14 R. Pfa
2
provide the necessary information for clinicians to select the patients that will benefit best from this valuable treatment without any long term risk. © 2015 Elsevier Ltd. All rights reserved.
Introduction GH for replacement therapy in GH deficient children until 1985 was extracted from cadaveric pituitary glands. However, at this time it was first recognized that by injecting pituitary extracts there was a possibility of transmitting Creutzfeldt-Jakob disease. Although the risk was statistically small, there were no means of testing the extract for the presence of the infecting agent and the disorder had such a devastating clinical outcome that the approval for using extracted human pituitary GH was withdrawn worldwide within a matter of weeks. This swift decision may also have been influenced by the upcoming approval of recombinant human GH (rhGH) as a treatment alternative. After the introduction of recombinant human insulin only months before, rhGH was the second biosynthetic molecule that was introduced into clinical medicine as a therapeutic agent. Because the quantities of rhGH for treatment thereafter seemed theoretically unlimited many physicians started clinical studies to optimize treatment strategies in respect to dosing and timing and to evaluate the use rhGH in children with short stature but without classical GHD. In the following years rhGH was approved for the treatment of various forms of short stature, where it has proven to improve growth and final height and/or psychosocial development of the patients (see Tables 1 and 2). Diagnosis of GHD Growth hormone secretion from the pituitary in children is pulsatile in nature, is mainly happening during the night and highly variable. A 24-h GH secretion profile in a child is only remotely overlapping on a day-to-day basis, because GH secretion is influenced by numerous factors on different levels of the hypothalamic-pituitary GH axis, therefore measuring a GH serum level even on a specific time point does not have any value in establishing the diagnosis Growth hormone deficiency (GHD). On the other hand GH is a hormone of “mass production”. The pituitary produces and secretes this hormone on a mg scale every day. The GH receptor (GHR) is not only expressed in the growth plate of the bone, that accounts for the longitudinal growth of the body, but it is expressed in many tissues throughout the organism, where GH secretion evokes numerous biochemical responses that can be used as surrogate markers for GH action (IGF-I and IGF binding proteins). However, none of these markers strongly correlate with the cumulative GH secretion as they are influenced by a number of
Table 1 Chronology of approvals of rhGH for treatment. Year of approval
Diagnosis
1985 1993 1996 2000 2001
GHD Chronic renal failure Turner Syndrome PradereWilli Syndrome Small for Gestational Age without catch-up growth Idiopathic short stature (ISS) Short stature homeobox containing gene (Shox) deficiency Noonan Syndrome
2003 2006 2007
Recommended dose according to approving institution USA (FDA) (mg/kg/day)
Japan (PMDA) mg/kg/week
Europe (EMA) (mg/kg/day)
23e43 50 47e67 34 67e69
0.175 0.175e0.35 0.35 0.245 0.23e0.47
25e35 45e50 45e50 35 35
43e67 50
Not approved Not approved
Not approved 50
66
Not approved
Not approved
Please cite this article in press as: Pf€affle R, Hormone replacement therapy in children: The use of growth hormone and IGF-I, Best Practice & Research Clinical Endocrinology & Metabolism (2015), http:// dx.doi.org/10.1016/j.beem.2015.04.009
€ ffle / Best Practice & Research Clinical Endocrinology & Metabolism xxx (2015) 1e14 R. Pfa
3
other conditions that dictate their production (i.e general health status, nutrition, liver and kidney function etc.). Despite these limitations IGF-I is generally used to screen for GHD, however, both IGF-I immunoassays and GH-immunoassays have also technical limitations in that there is considerable inter-assay variability, although some of the reasons for this variability in results are different [1,2]. To better standardize IGF-I measurements it is recommended to calibrate all assays against a new International Reference Agent for IGF-I (02/254) [3]. Similar efforts have been taken to better standardize the GH immunoassays. Over the past 10 years the minimal cut-off value for serum GH reached in GH stimulation tests has been lowered from the initially set 10 ng/ml to appr. 8 ng/ml due to the use of new biosynthetically produced calibrators for the assays [4]. However, these adaptations also depend on the individual assay used, therefore some collaborative effort has been taken to “harmonize” the measurements serum GH by the different assays used in clinical practice [5]. The use of different assays, however, is not the only reason for the difficulties encountered when establishing the diagnosis of GHD [6,7]. All agents used for provocative testing (Insulin, Arginine, Glucagon, Dopamine etc.) will evoke variable responses of GH secretion [4,8,9], therefore theoretically different cut-off values should be established for each of the stimuli used. Another issue that is discussed intensely by Pediatric endocrinologists is whether to or not to use sex hormone priming in prepubertal children undergoing GH testing, as it is well established that after exposure to sex hormones the peak GH level is increased [10,11], Considering the published data in regard to this issue priming of children suspected of GHD should be undertaken in adolescents with pubertal delay in girls from 11 years upward an boys from 13 years upward who do not show any signs of pubertal development [12,13]. Recently a discussion also has arisen to what extend the BMI influences GH response to provocative testing. Thereby obese children show diminished GH responses [14,15]. In adults this phenomenon has been previously described [16,17] and has found its consideration in the cut-off levels of adult GHtesting [18]. The genetics of normal growth and of the response to rhGH Genetic variability of adult height Genome-wide association studies have been used increasingly to study the genetic influence on height. As twin studies have shown in the past that app. 80% of adult height variation is genetically determined. In the initial studies in several hundred to thousand individuals height variation seemed to be associated with around 20 to 50 loci within the human genome [19], more recent meta analyses suggest that the number of genes controlling human height is much higher (several hundreds) and the effect size of every gene is in average less than 1 mm [20] (see Table 2). It is interesting to note that so far the number of genes identified in GWAS analyses on height which play a direct or indirect role within the GH-IGF-I axis is very limited. It thus is unlikely, that this approach will help predicting individual height in the near future. It also suggests that genetic variants of genes that have clinically significant effects on GH secretion are rare events within a population. Nevertheless this approach holds important information on the basic processes involved in growth regulation.
Table 2 Genome wide association studies. Sample size
Variance explained
No. SNPs significant
Effect size (mm)
Reference
25,000 130,000
3e5 % 10%
47 180
2e4 1e2
250,000
16%
697
0.05 mg/kg/day) doses of rhGH (SMR ¼ 2.73; CI 1.13e6.62). However, the overall number of patients, who were treated with such doses was small. In a second analysis this group reported an increased incidence of hemorrhagic brain insults with a standard incidence ration (SIR) of 1.5 (CI 0.7e2.7) [87]. These results were worrisome, however, another meta-analysis of 2800 patients from Sweden, Belgium and the Netherlands could not find an increased mortality [42]. These studies also illustrate that both the morbidity and the mortality risks in patients with short stature due to SGA are slightly different in comparison to patients with GHD or idiopathic short stature and that comparing their overall risks with the general population may be problematic. IGF-I treatment Even in comparison to Growth Hormone deficiency (GHD) GH insensitivity syndrome (GHIS) is a very rare cause for short stature in children. GHIS is caused either by primary IGF-I deficiency (PIGFD) due to GH resistance caused by mutations in the GH receptor (GHR formerly Laron's syndrome), or by mutations in on of the second messengers of the GHR like Stat5b, [88]. Additionally GH resistance can be found in patients with IGF-I gene deletions or mutations. As the molecular mechanisms cannot be identified in all patients with PIGFD criteria were proposed to classify this disorder on clinical grounds:
Severe short stature (height < 3 SD) Normal (or elevated) GH secretion Low IGF-I and IGFBP-3 serum levels (
Contents lists available at ScienceDirect
Best Practice & Research Clinical Endocrinology & Metabolism journal homepage: www.elsevier.com/locate/beem
2
Hormone replacement therapy in children: The use of growth hormone and IGF-I €ffle, Prof. Dr. med., MD Roland Pfa University Children's Hospital Leipzig, Liebigstr. 20a, 0413 Leipzig, Germany
a r t i c l e i n f o Article history: Available online xxx Keywords: recombinant human growth hormone recombinant human IGF-I growth hormone deficiency small for gestational age (SGA) Turner's syndrome treatment efficacy longterm side effects
Recombinant human GH (rhGH) has been available since 1985. This article gives an overview, what has been achieved over the past 30 years in respect to optimization of rhGH treatment for the individual child with GH deficiency and what are the safety issues concerned with this treatment. In the last twenty years significant scientific progress has been made in the diagnosis of GH deficiency, the genetic disorders that are associated with pituitary GH deficiency and the genetics that influence growth in general. On the other hand rhGH is not only used in states of GH deficiency but also various conditions without a proven GH deficiency by classical standards. Clinical studies that investigated both the genetics of growth and the individual responses to rhGH therapy in these patient populations were able to refine our concept about the physiology of normal growth. In most patients under rhGH treatment there is a considerable short-term effect, however the overall gain in growth obtained by a long-term treatment until final height still remains a matter of debate in some of the conditions treated. Also first studies on the long-term safety risks of rhGH treatment have raised the question whether this treatment is similarly safe for all the patient groups eligible for such a treatment. Therefore even in the face of a longstanding safety record of this drug replacement therapy the discussion about the right cost and risk to benefit ratio is continuing. Consequently there is still a need for carefully conducted long-term studies that use modern anthropometric, genetic, and laboratory techniques in order to
Abbreviations: EMA, European Medical Agency; FDA, United States Federal Drug Association; GHD, growth hormone deficiency; GHIS, growth hormone insensitivity syndrome; GHR, growth hormone receptor; IGFBP-3, IGF binding protein 3; ISS, idiopathic short stature; PIGFD, primary IGF-I deficiency; PTPN11, protein tyrosine phosphatase, nonreceptor type 11; rhGH, recombinant human GH; rhIGF-I, recombinant human IGF-I; SDS, SD score; SGA, small for gestational age. E-mail address: rpfaeffl[email protected]. http://dx.doi.org/10.1016/j.beem.2015.04.009 1521-690X/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Pf€affle R, Hormone replacement therapy in children: The use of growth hormone and IGF-I, Best Practice & Research Clinical Endocrinology & Metabolism (2015), http:// dx.doi.org/10.1016/j.beem.2015.04.009
€ ffle / Best Practice & Research Clinical Endocrinology & Metabolism xxx (2015) 1e14 R. Pfa
2
provide the necessary information for clinicians to select the patients that will benefit best from this valuable treatment without any long term risk. © 2015 Elsevier Ltd. All rights reserved.
Introduction GH for replacement therapy in GH deficient children until 1985 was extracted from cadaveric pituitary glands. However, at this time it was first recognized that by injecting pituitary extracts there was a possibility of transmitting Creutzfeldt-Jakob disease. Although the risk was statistically small, there were no means of testing the extract for the presence of the infecting agent and the disorder had such a devastating clinical outcome that the approval for using extracted human pituitary GH was withdrawn worldwide within a matter of weeks. This swift decision may also have been influenced by the upcoming approval of recombinant human GH (rhGH) as a treatment alternative. After the introduction of recombinant human insulin only months before, rhGH was the second biosynthetic molecule that was introduced into clinical medicine as a therapeutic agent. Because the quantities of rhGH for treatment thereafter seemed theoretically unlimited many physicians started clinical studies to optimize treatment strategies in respect to dosing and timing and to evaluate the use rhGH in children with short stature but without classical GHD. In the following years rhGH was approved for the treatment of various forms of short stature, where it has proven to improve growth and final height and/or psychosocial development of the patients (see Tables 1 and 2). Diagnosis of GHD Growth hormone secretion from the pituitary in children is pulsatile in nature, is mainly happening during the night and highly variable. A 24-h GH secretion profile in a child is only remotely overlapping on a day-to-day basis, because GH secretion is influenced by numerous factors on different levels of the hypothalamic-pituitary GH axis, therefore measuring a GH serum level even on a specific time point does not have any value in establishing the diagnosis Growth hormone deficiency (GHD). On the other hand GH is a hormone of “mass production”. The pituitary produces and secretes this hormone on a mg scale every day. The GH receptor (GHR) is not only expressed in the growth plate of the bone, that accounts for the longitudinal growth of the body, but it is expressed in many tissues throughout the organism, where GH secretion evokes numerous biochemical responses that can be used as surrogate markers for GH action (IGF-I and IGF binding proteins). However, none of these markers strongly correlate with the cumulative GH secretion as they are influenced by a number of
Table 1 Chronology of approvals of rhGH for treatment. Year of approval
Diagnosis
1985 1993 1996 2000 2001
GHD Chronic renal failure Turner Syndrome PradereWilli Syndrome Small for Gestational Age without catch-up growth Idiopathic short stature (ISS) Short stature homeobox containing gene (Shox) deficiency Noonan Syndrome
2003 2006 2007
Recommended dose according to approving institution USA (FDA) (mg/kg/day)
Japan (PMDA) mg/kg/week
Europe (EMA) (mg/kg/day)
23e43 50 47e67 34 67e69
0.175 0.175e0.35 0.35 0.245 0.23e0.47
25e35 45e50 45e50 35 35
43e67 50
Not approved Not approved
Not approved 50
66
Not approved
Not approved
Please cite this article in press as: Pf€affle R, Hormone replacement therapy in children: The use of growth hormone and IGF-I, Best Practice & Research Clinical Endocrinology & Metabolism (2015), http:// dx.doi.org/10.1016/j.beem.2015.04.009
€ ffle / Best Practice & Research Clinical Endocrinology & Metabolism xxx (2015) 1e14 R. Pfa
3
other conditions that dictate their production (i.e general health status, nutrition, liver and kidney function etc.). Despite these limitations IGF-I is generally used to screen for GHD, however, both IGF-I immunoassays and GH-immunoassays have also technical limitations in that there is considerable inter-assay variability, although some of the reasons for this variability in results are different [1,2]. To better standardize IGF-I measurements it is recommended to calibrate all assays against a new International Reference Agent for IGF-I (02/254) [3]. Similar efforts have been taken to better standardize the GH immunoassays. Over the past 10 years the minimal cut-off value for serum GH reached in GH stimulation tests has been lowered from the initially set 10 ng/ml to appr. 8 ng/ml due to the use of new biosynthetically produced calibrators for the assays [4]. However, these adaptations also depend on the individual assay used, therefore some collaborative effort has been taken to “harmonize” the measurements serum GH by the different assays used in clinical practice [5]. The use of different assays, however, is not the only reason for the difficulties encountered when establishing the diagnosis of GHD [6,7]. All agents used for provocative testing (Insulin, Arginine, Glucagon, Dopamine etc.) will evoke variable responses of GH secretion [4,8,9], therefore theoretically different cut-off values should be established for each of the stimuli used. Another issue that is discussed intensely by Pediatric endocrinologists is whether to or not to use sex hormone priming in prepubertal children undergoing GH testing, as it is well established that after exposure to sex hormones the peak GH level is increased [10,11], Considering the published data in regard to this issue priming of children suspected of GHD should be undertaken in adolescents with pubertal delay in girls from 11 years upward an boys from 13 years upward who do not show any signs of pubertal development [12,13]. Recently a discussion also has arisen to what extend the BMI influences GH response to provocative testing. Thereby obese children show diminished GH responses [14,15]. In adults this phenomenon has been previously described [16,17] and has found its consideration in the cut-off levels of adult GHtesting [18]. The genetics of normal growth and of the response to rhGH Genetic variability of adult height Genome-wide association studies have been used increasingly to study the genetic influence on height. As twin studies have shown in the past that app. 80% of adult height variation is genetically determined. In the initial studies in several hundred to thousand individuals height variation seemed to be associated with around 20 to 50 loci within the human genome [19], more recent meta analyses suggest that the number of genes controlling human height is much higher (several hundreds) and the effect size of every gene is in average less than 1 mm [20] (see Table 2). It is interesting to note that so far the number of genes identified in GWAS analyses on height which play a direct or indirect role within the GH-IGF-I axis is very limited. It thus is unlikely, that this approach will help predicting individual height in the near future. It also suggests that genetic variants of genes that have clinically significant effects on GH secretion are rare events within a population. Nevertheless this approach holds important information on the basic processes involved in growth regulation.
Table 2 Genome wide association studies. Sample size
Variance explained
No. SNPs significant
Effect size (mm)
Reference
25,000 130,000
3e5 % 10%
47 180
2e4 1e2
250,000
16%
697
0.05 mg/kg/day) doses of rhGH (SMR ¼ 2.73; CI 1.13e6.62). However, the overall number of patients, who were treated with such doses was small. In a second analysis this group reported an increased incidence of hemorrhagic brain insults with a standard incidence ration (SIR) of 1.5 (CI 0.7e2.7) [87]. These results were worrisome, however, another meta-analysis of 2800 patients from Sweden, Belgium and the Netherlands could not find an increased mortality [42]. These studies also illustrate that both the morbidity and the mortality risks in patients with short stature due to SGA are slightly different in comparison to patients with GHD or idiopathic short stature and that comparing their overall risks with the general population may be problematic. IGF-I treatment Even in comparison to Growth Hormone deficiency (GHD) GH insensitivity syndrome (GHIS) is a very rare cause for short stature in children. GHIS is caused either by primary IGF-I deficiency (PIGFD) due to GH resistance caused by mutations in the GH receptor (GHR formerly Laron's syndrome), or by mutations in on of the second messengers of the GHR like Stat5b, [88]. Additionally GH resistance can be found in patients with IGF-I gene deletions or mutations. As the molecular mechanisms cannot be identified in all patients with PIGFD criteria were proposed to classify this disorder on clinical grounds:
Severe short stature (height < 3 SD) Normal (or elevated) GH secretion Low IGF-I and IGFBP-3 serum levels (
