0021-972X/92/7404-0898$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society
Vol. 74, No. 4 Printed
in U.S.A.
Dose-Response Study of Biosynthetic Human Growth Hormone (GH) in GH-Deficient Children: Effects on Auxological and Biochemical Parameters* S. M. P. F. DE MUINCK KEIZER-SCHRAMA, B. RIKKEN, H. J. WYNNE, A. C. S. HOKKEN-KOELEGA, J. M. WIT, A. BOT, S. L. S. DROP, AND DUTCH GROWTH HORMONE WORKING GROUP? Department of Pediatrics, Division of Endocrinology, Medical School (S.M.P.F.D.M.K.-S., A.C.S.H,-K., S.L.S.D.), and the Department of Endocrinology, Growth, and Reproduction (A.B.), Erasmus University, Rotterdam; and Medical School (B. R., J.M. W) and the Center for Biostatistics (H.J. W.), University of Utrecht, Utrecht, The Netherlands
ABSTRACT. A multicenter dose-response study evaluated the effect of two different doses of biosynthetic GH on auxological and biochemical parameters in 38 prepubertal children with GH deficiency (GHD). Twenty-one were newly diagnosed, while 17 transfer patients had been on GH treatment for at least 1 yr before the study. New and transfer patients alike were treated with either 2 or 4 IU GH/m’. day SC. At evaluation all new patients had completed 1 yr of treatment, while transfers had completed 2 yr of treatment under study. In the new patients both doses resulted in a significant increase in height velocity (HV) and height SD score (SDS), with comparable bone maturation. After correction for the severity of GHD, the increase in HV SDS was significantly greater with 4 IU than with 2 IU (P < 0.01). In the transfer patients HV, height SDS, and predicted adult height only increased significantly with 4 IU (P < 0.05). Bone maturation was comparable for the two doses. There was a significant correlation between first year growth response and
GH dose. In the new patients, the plasma insulin-like growth factor-I (IGF-I) concentration increased significantly without a significant difference between dosage groups. There was a positive correlation between growth response and increment of plasma IGF-I SDS. In new and transfer patients alike, above normal plasma IGF-I levels were observed, particularly with 4 IU. Hemoglobin-Al remained constant with both GH doses in both groups, while cholesterol and LDL levels tended to decrease. In the new patients, the mean apolipoprotein-A, level was lower than the control value after 1 yr on 4 IU GH. Treatment with 4 IU GH/m*.day led to a greater growth response than a dose of 2 IU in newly diagnosed as well as previously treated GHD patients. Bone maturation was comparable for both doses. No adverse effects were observed with the higher GH dose, but the long term effects on IGF-I and lipid metabolism need further attention. (J Clin Endocrinol Metab 74: 898-905,1992)
G
H THERAPY is known to increase the growth rate of children with GH deficiency (GHD). However, long term follow-up studies have revealed disappointing final height results (1,2). Apart from early diagnosis and treatment, optimization of dosage and modes of GH administration might increase final height (1, 2). Some investigators have found a quantitative relationship between GH dosage and growth response (3-7), but this is questioned by others (8). In only one of these studies (7)
was GH administered on a daily basis. Patients treated with daily SCinjections appeared to have a greater growth response than those receiving im injections three times per week (9, 10). Injecting 2 IU GH/m2+day SC, roughly equal to 0.06 III/kg. day, Sandahl Christiansen et al. (11) achieved near-normal nocturnal plasma patterns in GHD children. Consequently, the treatment regimen currently used in the Netherlands is 2 IU GH/m2.day SC. The unlimited availability of biosynthetic human GH (b-hGH) now allows for a wider dose range, which might increase the growth response. However, it may also cause unwanted side-effects. Of particular concern are the effects of higher GH dosage on body proportions and bone maturation as well as on glucose and lipid metabolism (12, 13). In 1987, we initiated a multicenter study comparing the effect of the standard dosage of 2 IU/m2.day with that of 4 IU/m2 -day in terms of growth, plasma insulin-like growth factor-I (IGF-I), serum lipids, and other biochemical parameters.
Received May 6, 1991. Address requests for reprints to: S. M. P. F. de Muinck KeizerSchrama, M.D., Ph.D., Division of Endocrinology, Sophia Children’s Hospital, P.O. Box 70029,300O LL Rotterdam, The Netherlands. * Presented in part at the 29th Annual Meeting of the European Society of Pediatric Endocrinology (ESPE), Vienna, Austria, September 1990. This work was supported by Novo-Nordisk A/S, Denmark. t The other members of the Dutch Growth Hormone Working Group who participated in this study are H. A. Delemarre-v.d.Waal (Amsterdam), W. Oostdijk (Leiden), H. M. Reeser (The Hague), and G. B. A. Stoelinga (Nijmegen). 898
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 July 2015. at 04:31 For personal use only. No other uses without permission. . All rights reserved.
DOSE-RESPONSE
Subjects
GH TRIAL
and Methods
Patients
Two separate groups of prepubertal GHD patients were studied, 21 new patients enrolled upon presentation in the course of the study and 17 transfer patients. The latter group had been receiving b-hGH treatment for at least 1 yr before the study, of which at least 6 months were with a fixed dosage (12 IU/m* sweek, SC).The diagnosis of GHD was based on a yearly height velocity (HV) below the 25th percentile (P25) for chronological age (CA) before the age of 12 yr for boys and 10 yr for girls (14), while for older children the HV had to be below P25 for bone age (BA), and maximum plasma GH peaks had to be under 10 pg/L in at least 2 standard pharmacological provocation tests (arginine, clonidine, L-dopa/propranolol, and glucagon). The entry requirements were 1) documented GHD of organic or idiopathic origin, including isolated GHD as well as multiple pituitary hormone deficiencies; 2) BA under 12 yr for boys and under 10 yr for girls, determined according to the procedure of Tanner and Whitehouse, 20 bones (TW2-20) (15); 3) normal serum T4 levels; and 4) no signs of puberty. Excluded from the study were 1) patients with growth retardation due to diabetes mellitus, inborn error of metabolism, primary bone disease, disorders of cardiopulmonary, gastrointestinal, genitourinary, or nervous tracts, or serious suspicion of psychosocial dwarfism; 2) children receiving any kind of drug that might interfere with GH therapy, with the exception of replacement doses of T4 and/or hydrocortisone in multiple hormone deficiencies; and 3) patients with insufficient growth response to previous GH therapy. Each group was equally divided into 2 subsets for treatment with either 2 IU (subset A) or 4 IU (subset B) authentic b-hGH (Norditropin, Novo-Nordisk, Gentofte, Denmark)/m’, injected SCdaily at bedtime. Assignment to either subset was at random. The transfer patients, who enrolled en bloc at the start of the study, were paired according to CA and BA. Table 1 gives the
IN GHD CHILDREN
baseline clinical data for all 38 patients. The protocol was approved by the ethical committees of all six participating centers, and all parents gave their written informed consent for the study. Growth
evaluation
At the start of the study and subsequently every 3 months, the patients underwent a physical examination, including measurements of height, sitting height (SH), weight, and blood pressure. Height and SH were measured with Harpenden equipment, four times in succession by two investigators for each center and always at the same time of day. The mean values were used for further analysis. Weight was expressed as a percentage of the median weight for height (14). Systolic and diastolic blood pressures were measured in a sitting position with a Dynamap four times in succession with lo-min rest periods in between. The mean of the last three measurements was taken for analysis. BA was determined by one investigator (A.B.) immediately before the start of the study and subsequently every 6 months, using the TW2-20 and TW2-RUS method (15). TW2-20 was used for all calculations, and TW2-RUS was used for adult height prediction (16). Height was expressed as the SD score (SDS) for CA and BA using the most recent Dutch national growth reference values (17). The ratio of SH/subischial leg length (SLL) was expressed as SDS for BA, using the reference values from the Zurich Longitudinal Growth Study (18). The predicted adult height (PAH) was expressed as SDS according to the 1980 Dutch national reference values (17). The midparental height (MPH) SDS was calculated on the basis of the 1965 Dutch reference values (19). PAH in relation to MPH was based on the difference between these two parameters, expressed as SDS. Height velocity (HV) was expressed as centimeters per yr. HV SDS was based on the HV reference values of the childhood component from the infancy-childhood-puberty model designed by Karlberg (20). The growth response in the first year of study
TABLE 1. Baseline clinical data for new and transfer patients per subset (A, 2 IU GH/m’. day; B, 4 IU GH/m’. day) New patients A
Transfer patients B
10
No. of patients
8M/2F
Sexratio CA W’ BA (TW2-20; yr)’ Isolated GHD (no. of patients)
6.8 (1.5-11.6) 5.4 (0.7-9.4)
8 Idiopathic 1 Organic MPHD (no. of patients) 1 Idiopathic Organic Maximum GH values (no. of patients) 4 c3.5 fig/L 6 3.5-10 fig/L Midparental ht SDS” -1.13 (-1.72 to 0.20) Yr of GH therapy” a Values are the mean, with the range in parentheses.
8&F 6.9 (1.5-13.8) 5.6 (0.8-11.6) 8 2 1 4 7 -0.78 (-2.90 to 1.31)
A
B
9
8
6M/3F
5M/3F
9.9 (2.8-14.9) 7.6 (2.7-10.8)
9.5 (6.1-14.9) 7.6 (4.5-11.2)
5
2
3 1
6
8 1 -0.18 (-2.17 to 1.29) 4.0 (1.4-8.8)
8 0.05 (-1.50 to 1.88) 5.1 (1.3-10.5)
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 July 2015. at 04:31 For personal use only. No other uses without permission. . All rights reserved.
900
DE MUINCK
KEIZER-SCHRAMA
(A HV SDS) was determined by subtracting the HV SDS in the preceding year from the HV SDS in the first year of the study. Laboratory
analyses
At baseline, 3 and 6 months, and subsequently every 6 months, a number of hematological, biochemical, and immunological variables were determined. Biochemical determinations included levels of urea nitrogen, creatinine, alkaline phosphatase, ALAT, ASAT, albumin, glycated hemoglobin (HbAl), Td, IGF-I, and serum lipids. Urine was tested for glucose, protein, blood, and sediment. Total stable HbAl was measured by agar gel electrophoresis (Corning Medical Ltd., Medfield, MA) and quantitated by means of scanning densitometry (21). After acid chromatography, plasma IGF-I was determined by a specific RIA (22). To adjust measured IGF-I for age and sex, SDS were calculated after log transformation up to the age of 14.5 yr for boys and 13.5 yr for girls. After that age, mean plasma IGF-I levels decrease, and the use of IGF-I SDS beyond these limits would, therefore, lead to unacceptable bias. To calculate the increment in IGF-I SDS during the first year (A IGF-I SDS), the baseline IGF-I SDS was substracted from the mean of IGF-I SDS determined at 3,6, and 12 months (average IGF-I SDS). Sampling for serum lipids was restricted to baseline and 12 and 24 months. Levels of cholesterol, low density lipoprotein (LDL), apolipoprotein-Al (APO-Al), and APO-B were measured in the new patients, while only cholesterol and LDL concentrations were determined for the transfer group. Serum total cholesterol was measured with an automated enzymatic method, using the Boehringer Mannheim cholesteroloxydaseparaaminophenazone reagent kit (Indianapolis, IN) (23). LDL cholesterol was measured by the same method after separation of the relevant lipoprotein fractions and precipitation with polyvinylsulphate (Boehringer Mannheim). APO-A, and APO-B were assayed by an automated immunoturbidimetric method (Kone Diagnostics, Espoo, Finland). Lipid analysis was subject to the quality assessment program of the WHO Regional Lipid Reference Center (Prague, Czechoslovakia). The lipid profiles were compared with those of healthy Dutch children available at the laboratory from a previous study involving 757 boys and 801 girls wih a mean (*SD) age of 8.4 f 0.7 yr (24). Immunological screening concerned anti-GH antibodies. These were measured semiquantitatively by determining the percent binding of 5000 cpm [‘251]hGH in 50 PL serum before and after the addition of 25 PU b-hGH after overnight incubation at 2 C in 400 PL phosphate buffer. Separation of free and bound hormone was achieved with 1 mL charcoal-Dextran250 (4%-l%). A binding percentage of more than 15% was taken as evidence for the presence of anti-GH antibodies. Scatchard plots were made to determine the binding capacities of these sera. Statistical
analyses
In the auxological evaluation, the Wilcoxon signed rank test was used for differences within each dosage group, and the Mann-Wittney U test was used for differences between dosage groups. In the biochemical evaluation, repeated measures analysis of variance was used to test the significance of changes
JCE & M .1992 Vol74.No4
ET AL.
with time and dosage and/or interaction of time with dosage. For plasma IGF-I values, this analysis was performed after logarithmic transformation to obtain homogeneity of variance. To study the effects of a number of auxological and laboratory variables on the first year increment in HV SDS (A HV SDS), a stepwise multiple linear regression (MLR) analysis was performed separately for each study group, with A HV SDS as the dependent variable. The variables were sex, age, bone age (TW2-20), bone age retardation, and weight as percentage of the median weight for height, all at the start of the study; HV SDS in the year before the study (HV SDS 0), maximum plasma GH peak after pharmacological stimulation (GHP), IGF-I SDS at the start of the study (IGF SDS 0), isolated GHD or multiple pituitary hormone deficiencies (MPHD), idiopathic or organic cause of GHD, years of GH treatment before study (transfers only), HV SDS 1 yr before any GH therapy was started (transfers only), and dose of GH administered during study (2 or 4 IU/m’ . day). In view of the small numbers, MLR analyses did not include the two new patients with MPHD or the one transfer patient with an organic cause of GHD. Results Regarding the two study groups separately, the clinical data for each subset appeared comparable, except for the type of GHD in the transfer patients (Table 1). At evaluation, the new patients had all completed 1 yr of treatment, and the transfer patients had completed 2 yr of treatment under study. Table 2 gives the auxological results for both study groups, once again showing comparable prestudy data for the subsets.
Growth eualuation New patients
HV and HV SDS increased substantially with both doses in the first year of treatment. Except for 1 patient of subset A, the HV increment amounted to more than 2 cm. The mean increase in HV and HV SDS was 2.5 cm and 2.5 SDS higher with 4 IU than with 2 IU/m2, but the difference was statistically nonsignificant. Table 3 gives the results of the MLR analysis with A HV SDS as the dependent variable. The following variables were preselected as promising predictors: GHP, IGF SDS 0, and HV SDS 0. Dose was always the key variable, because the objective of the analysis was to search for variables that modified the effect of GH dosage on A HV SDS. Equations 5 and 6 represent models of 3 variables combining the independent significant effect of DOSE, GHP, or IGF SDS 0 and that of HV SDS 0 on the growth response. Taking into account the maximum plasma GH peak after pharmacological stimulation or pretreatment IGF-I level combined with pretreatment HV SDS, the effect of GH dosage on the growth response was statistically significant (P < 0.01). In other words, GH dosage and severity of GHD were the determining factors. Residual analysis of all 21 patients revealed that the 2 MPHD patients conformed with the others regarding Eq
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 July 2015. at 04:31 For personal use only. No other uses without permission. . All rights reserved.
DOSE-RESPONSE TABLE 2. Effect of either for CA and BA, and PAH
2 IU (A) or 4 IU (B) GH/m’.day SDS minus MPH SDS
GH TRIAL on mean
(&SD)
IN GHD HV, HV
CHILDREN
SDS, advance
901
in bone maturation
New patients
Transfer Change from prestudy, A vs. B (P value)
(ni
9)
(nf
(A BA/A
CA),
height
SDS
patients
(n !lO)
(n =“ll)
8)
5.5 (2.2) 11.0 (3.0)”
5.3 (2.2) 13.3 (3.9)”
NS
8.2 (2.2) 7.4 (1.6) 7.2 (1.7)
-1.71 (1.85) 5.54 (3.18)*
-1.58 (1.70) 8.18 (2.92)”
NS
3.92 (2.70) 3.47 (1.58) 3.19 (2.28)
5.00 (2.23) 8.14 (2.34)’ 6.39 (1.96)d
0.7 (0.4) 1.3 (0.4)
0.7 (0.2) 1.1 (0.3)
0.8 (0.4) 1.1 (0.4) 1.0 (0.3)
1.1 (0.5) 1.2 (0.5) 1.2 (0.6)
Change from prestudy, A vs. B (P value)
HV (cm/yr) Prestudy Yr 1 Yr2’ HV SDS Prestudy Yr 1 Yr 2’ A BA/A CA (yr/yr) Prestudy Yr 1 Yr 2’ Ht SDS (CA) Prestudy Yr 1 Yr2” Ht SDS (BA) Prestudy Yr 1 Yr 2’ PAH SDS - MPH Prestudy Yr 1 Yr2
SDS
-3.60 -2.44
(0.88) (0.53)O
-3.26 -1.70
(1.52) (1.35)”
NS
-1.68 -1.01
(1.04) (1.06)*
-1.33 -0.11
(1.24) (1.00)”
co.05
(n = 8) -1.78 (1.51) -0.84 (1.33)’
NS
(n = 6) -2.07 (0.87) -1.13 (0.88)d
8.9 (1.6) 11.0 (1.8)* 9.6 (1.4)d
-2.18 -1.94 -1.49
(0.90) (1.27) (1.48)
-2.08 -1.20 -0.17
(1.89) (2.12)b (2.49)d
-0.26 0.00 0.07 (n -0.68 -0.60 -0.76
(1.04) (0.96) (0.84) = 8) (1.25) (0.96) (0.63)
-0.24 0.46 1.08 (n -1.59
(1.49) (1.48)d (1.56) = 8) (1.38)
-0.96 (1.37) b -0.15
(1.69)b
Vol. 74, No. 4 Printed
in U.S.A.
Dose-Response Study of Biosynthetic Human Growth Hormone (GH) in GH-Deficient Children: Effects on Auxological and Biochemical Parameters* S. M. P. F. DE MUINCK KEIZER-SCHRAMA, B. RIKKEN, H. J. WYNNE, A. C. S. HOKKEN-KOELEGA, J. M. WIT, A. BOT, S. L. S. DROP, AND DUTCH GROWTH HORMONE WORKING GROUP? Department of Pediatrics, Division of Endocrinology, Medical School (S.M.P.F.D.M.K.-S., A.C.S.H,-K., S.L.S.D.), and the Department of Endocrinology, Growth, and Reproduction (A.B.), Erasmus University, Rotterdam; and Medical School (B. R., J.M. W) and the Center for Biostatistics (H.J. W.), University of Utrecht, Utrecht, The Netherlands
ABSTRACT. A multicenter dose-response study evaluated the effect of two different doses of biosynthetic GH on auxological and biochemical parameters in 38 prepubertal children with GH deficiency (GHD). Twenty-one were newly diagnosed, while 17 transfer patients had been on GH treatment for at least 1 yr before the study. New and transfer patients alike were treated with either 2 or 4 IU GH/m’. day SC. At evaluation all new patients had completed 1 yr of treatment, while transfers had completed 2 yr of treatment under study. In the new patients both doses resulted in a significant increase in height velocity (HV) and height SD score (SDS), with comparable bone maturation. After correction for the severity of GHD, the increase in HV SDS was significantly greater with 4 IU than with 2 IU (P < 0.01). In the transfer patients HV, height SDS, and predicted adult height only increased significantly with 4 IU (P < 0.05). Bone maturation was comparable for the two doses. There was a significant correlation between first year growth response and
GH dose. In the new patients, the plasma insulin-like growth factor-I (IGF-I) concentration increased significantly without a significant difference between dosage groups. There was a positive correlation between growth response and increment of plasma IGF-I SDS. In new and transfer patients alike, above normal plasma IGF-I levels were observed, particularly with 4 IU. Hemoglobin-Al remained constant with both GH doses in both groups, while cholesterol and LDL levels tended to decrease. In the new patients, the mean apolipoprotein-A, level was lower than the control value after 1 yr on 4 IU GH. Treatment with 4 IU GH/m*.day led to a greater growth response than a dose of 2 IU in newly diagnosed as well as previously treated GHD patients. Bone maturation was comparable for both doses. No adverse effects were observed with the higher GH dose, but the long term effects on IGF-I and lipid metabolism need further attention. (J Clin Endocrinol Metab 74: 898-905,1992)
G
H THERAPY is known to increase the growth rate of children with GH deficiency (GHD). However, long term follow-up studies have revealed disappointing final height results (1,2). Apart from early diagnosis and treatment, optimization of dosage and modes of GH administration might increase final height (1, 2). Some investigators have found a quantitative relationship between GH dosage and growth response (3-7), but this is questioned by others (8). In only one of these studies (7)
was GH administered on a daily basis. Patients treated with daily SCinjections appeared to have a greater growth response than those receiving im injections three times per week (9, 10). Injecting 2 IU GH/m2+day SC, roughly equal to 0.06 III/kg. day, Sandahl Christiansen et al. (11) achieved near-normal nocturnal plasma patterns in GHD children. Consequently, the treatment regimen currently used in the Netherlands is 2 IU GH/m2.day SC. The unlimited availability of biosynthetic human GH (b-hGH) now allows for a wider dose range, which might increase the growth response. However, it may also cause unwanted side-effects. Of particular concern are the effects of higher GH dosage on body proportions and bone maturation as well as on glucose and lipid metabolism (12, 13). In 1987, we initiated a multicenter study comparing the effect of the standard dosage of 2 IU/m2.day with that of 4 IU/m2 -day in terms of growth, plasma insulin-like growth factor-I (IGF-I), serum lipids, and other biochemical parameters.
Received May 6, 1991. Address requests for reprints to: S. M. P. F. de Muinck KeizerSchrama, M.D., Ph.D., Division of Endocrinology, Sophia Children’s Hospital, P.O. Box 70029,300O LL Rotterdam, The Netherlands. * Presented in part at the 29th Annual Meeting of the European Society of Pediatric Endocrinology (ESPE), Vienna, Austria, September 1990. This work was supported by Novo-Nordisk A/S, Denmark. t The other members of the Dutch Growth Hormone Working Group who participated in this study are H. A. Delemarre-v.d.Waal (Amsterdam), W. Oostdijk (Leiden), H. M. Reeser (The Hague), and G. B. A. Stoelinga (Nijmegen). 898
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 July 2015. at 04:31 For personal use only. No other uses without permission. . All rights reserved.
DOSE-RESPONSE
Subjects
GH TRIAL
and Methods
Patients
Two separate groups of prepubertal GHD patients were studied, 21 new patients enrolled upon presentation in the course of the study and 17 transfer patients. The latter group had been receiving b-hGH treatment for at least 1 yr before the study, of which at least 6 months were with a fixed dosage (12 IU/m* sweek, SC).The diagnosis of GHD was based on a yearly height velocity (HV) below the 25th percentile (P25) for chronological age (CA) before the age of 12 yr for boys and 10 yr for girls (14), while for older children the HV had to be below P25 for bone age (BA), and maximum plasma GH peaks had to be under 10 pg/L in at least 2 standard pharmacological provocation tests (arginine, clonidine, L-dopa/propranolol, and glucagon). The entry requirements were 1) documented GHD of organic or idiopathic origin, including isolated GHD as well as multiple pituitary hormone deficiencies; 2) BA under 12 yr for boys and under 10 yr for girls, determined according to the procedure of Tanner and Whitehouse, 20 bones (TW2-20) (15); 3) normal serum T4 levels; and 4) no signs of puberty. Excluded from the study were 1) patients with growth retardation due to diabetes mellitus, inborn error of metabolism, primary bone disease, disorders of cardiopulmonary, gastrointestinal, genitourinary, or nervous tracts, or serious suspicion of psychosocial dwarfism; 2) children receiving any kind of drug that might interfere with GH therapy, with the exception of replacement doses of T4 and/or hydrocortisone in multiple hormone deficiencies; and 3) patients with insufficient growth response to previous GH therapy. Each group was equally divided into 2 subsets for treatment with either 2 IU (subset A) or 4 IU (subset B) authentic b-hGH (Norditropin, Novo-Nordisk, Gentofte, Denmark)/m’, injected SCdaily at bedtime. Assignment to either subset was at random. The transfer patients, who enrolled en bloc at the start of the study, were paired according to CA and BA. Table 1 gives the
IN GHD CHILDREN
baseline clinical data for all 38 patients. The protocol was approved by the ethical committees of all six participating centers, and all parents gave their written informed consent for the study. Growth
evaluation
At the start of the study and subsequently every 3 months, the patients underwent a physical examination, including measurements of height, sitting height (SH), weight, and blood pressure. Height and SH were measured with Harpenden equipment, four times in succession by two investigators for each center and always at the same time of day. The mean values were used for further analysis. Weight was expressed as a percentage of the median weight for height (14). Systolic and diastolic blood pressures were measured in a sitting position with a Dynamap four times in succession with lo-min rest periods in between. The mean of the last three measurements was taken for analysis. BA was determined by one investigator (A.B.) immediately before the start of the study and subsequently every 6 months, using the TW2-20 and TW2-RUS method (15). TW2-20 was used for all calculations, and TW2-RUS was used for adult height prediction (16). Height was expressed as the SD score (SDS) for CA and BA using the most recent Dutch national growth reference values (17). The ratio of SH/subischial leg length (SLL) was expressed as SDS for BA, using the reference values from the Zurich Longitudinal Growth Study (18). The predicted adult height (PAH) was expressed as SDS according to the 1980 Dutch national reference values (17). The midparental height (MPH) SDS was calculated on the basis of the 1965 Dutch reference values (19). PAH in relation to MPH was based on the difference between these two parameters, expressed as SDS. Height velocity (HV) was expressed as centimeters per yr. HV SDS was based on the HV reference values of the childhood component from the infancy-childhood-puberty model designed by Karlberg (20). The growth response in the first year of study
TABLE 1. Baseline clinical data for new and transfer patients per subset (A, 2 IU GH/m’. day; B, 4 IU GH/m’. day) New patients A
Transfer patients B
10
No. of patients
8M/2F
Sexratio CA W’ BA (TW2-20; yr)’ Isolated GHD (no. of patients)
6.8 (1.5-11.6) 5.4 (0.7-9.4)
8 Idiopathic 1 Organic MPHD (no. of patients) 1 Idiopathic Organic Maximum GH values (no. of patients) 4 c3.5 fig/L 6 3.5-10 fig/L Midparental ht SDS” -1.13 (-1.72 to 0.20) Yr of GH therapy” a Values are the mean, with the range in parentheses.
8&F 6.9 (1.5-13.8) 5.6 (0.8-11.6) 8 2 1 4 7 -0.78 (-2.90 to 1.31)
A
B
9
8
6M/3F
5M/3F
9.9 (2.8-14.9) 7.6 (2.7-10.8)
9.5 (6.1-14.9) 7.6 (4.5-11.2)
5
2
3 1
6
8 1 -0.18 (-2.17 to 1.29) 4.0 (1.4-8.8)
8 0.05 (-1.50 to 1.88) 5.1 (1.3-10.5)
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 July 2015. at 04:31 For personal use only. No other uses without permission. . All rights reserved.
900
DE MUINCK
KEIZER-SCHRAMA
(A HV SDS) was determined by subtracting the HV SDS in the preceding year from the HV SDS in the first year of the study. Laboratory
analyses
At baseline, 3 and 6 months, and subsequently every 6 months, a number of hematological, biochemical, and immunological variables were determined. Biochemical determinations included levels of urea nitrogen, creatinine, alkaline phosphatase, ALAT, ASAT, albumin, glycated hemoglobin (HbAl), Td, IGF-I, and serum lipids. Urine was tested for glucose, protein, blood, and sediment. Total stable HbAl was measured by agar gel electrophoresis (Corning Medical Ltd., Medfield, MA) and quantitated by means of scanning densitometry (21). After acid chromatography, plasma IGF-I was determined by a specific RIA (22). To adjust measured IGF-I for age and sex, SDS were calculated after log transformation up to the age of 14.5 yr for boys and 13.5 yr for girls. After that age, mean plasma IGF-I levels decrease, and the use of IGF-I SDS beyond these limits would, therefore, lead to unacceptable bias. To calculate the increment in IGF-I SDS during the first year (A IGF-I SDS), the baseline IGF-I SDS was substracted from the mean of IGF-I SDS determined at 3,6, and 12 months (average IGF-I SDS). Sampling for serum lipids was restricted to baseline and 12 and 24 months. Levels of cholesterol, low density lipoprotein (LDL), apolipoprotein-Al (APO-Al), and APO-B were measured in the new patients, while only cholesterol and LDL concentrations were determined for the transfer group. Serum total cholesterol was measured with an automated enzymatic method, using the Boehringer Mannheim cholesteroloxydaseparaaminophenazone reagent kit (Indianapolis, IN) (23). LDL cholesterol was measured by the same method after separation of the relevant lipoprotein fractions and precipitation with polyvinylsulphate (Boehringer Mannheim). APO-A, and APO-B were assayed by an automated immunoturbidimetric method (Kone Diagnostics, Espoo, Finland). Lipid analysis was subject to the quality assessment program of the WHO Regional Lipid Reference Center (Prague, Czechoslovakia). The lipid profiles were compared with those of healthy Dutch children available at the laboratory from a previous study involving 757 boys and 801 girls wih a mean (*SD) age of 8.4 f 0.7 yr (24). Immunological screening concerned anti-GH antibodies. These were measured semiquantitatively by determining the percent binding of 5000 cpm [‘251]hGH in 50 PL serum before and after the addition of 25 PU b-hGH after overnight incubation at 2 C in 400 PL phosphate buffer. Separation of free and bound hormone was achieved with 1 mL charcoal-Dextran250 (4%-l%). A binding percentage of more than 15% was taken as evidence for the presence of anti-GH antibodies. Scatchard plots were made to determine the binding capacities of these sera. Statistical
analyses
In the auxological evaluation, the Wilcoxon signed rank test was used for differences within each dosage group, and the Mann-Wittney U test was used for differences between dosage groups. In the biochemical evaluation, repeated measures analysis of variance was used to test the significance of changes
JCE & M .1992 Vol74.No4
ET AL.
with time and dosage and/or interaction of time with dosage. For plasma IGF-I values, this analysis was performed after logarithmic transformation to obtain homogeneity of variance. To study the effects of a number of auxological and laboratory variables on the first year increment in HV SDS (A HV SDS), a stepwise multiple linear regression (MLR) analysis was performed separately for each study group, with A HV SDS as the dependent variable. The variables were sex, age, bone age (TW2-20), bone age retardation, and weight as percentage of the median weight for height, all at the start of the study; HV SDS in the year before the study (HV SDS 0), maximum plasma GH peak after pharmacological stimulation (GHP), IGF-I SDS at the start of the study (IGF SDS 0), isolated GHD or multiple pituitary hormone deficiencies (MPHD), idiopathic or organic cause of GHD, years of GH treatment before study (transfers only), HV SDS 1 yr before any GH therapy was started (transfers only), and dose of GH administered during study (2 or 4 IU/m’ . day). In view of the small numbers, MLR analyses did not include the two new patients with MPHD or the one transfer patient with an organic cause of GHD. Results Regarding the two study groups separately, the clinical data for each subset appeared comparable, except for the type of GHD in the transfer patients (Table 1). At evaluation, the new patients had all completed 1 yr of treatment, and the transfer patients had completed 2 yr of treatment under study. Table 2 gives the auxological results for both study groups, once again showing comparable prestudy data for the subsets.
Growth eualuation New patients
HV and HV SDS increased substantially with both doses in the first year of treatment. Except for 1 patient of subset A, the HV increment amounted to more than 2 cm. The mean increase in HV and HV SDS was 2.5 cm and 2.5 SDS higher with 4 IU than with 2 IU/m2, but the difference was statistically nonsignificant. Table 3 gives the results of the MLR analysis with A HV SDS as the dependent variable. The following variables were preselected as promising predictors: GHP, IGF SDS 0, and HV SDS 0. Dose was always the key variable, because the objective of the analysis was to search for variables that modified the effect of GH dosage on A HV SDS. Equations 5 and 6 represent models of 3 variables combining the independent significant effect of DOSE, GHP, or IGF SDS 0 and that of HV SDS 0 on the growth response. Taking into account the maximum plasma GH peak after pharmacological stimulation or pretreatment IGF-I level combined with pretreatment HV SDS, the effect of GH dosage on the growth response was statistically significant (P < 0.01). In other words, GH dosage and severity of GHD were the determining factors. Residual analysis of all 21 patients revealed that the 2 MPHD patients conformed with the others regarding Eq
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 30 July 2015. at 04:31 For personal use only. No other uses without permission. . All rights reserved.
DOSE-RESPONSE TABLE 2. Effect of either for CA and BA, and PAH
2 IU (A) or 4 IU (B) GH/m’.day SDS minus MPH SDS
GH TRIAL on mean
(&SD)
IN GHD HV, HV
CHILDREN
SDS, advance
901
in bone maturation
New patients
Transfer Change from prestudy, A vs. B (P value)
(ni
9)
(nf
(A BA/A
CA),
height
SDS
patients
(n !lO)
(n =“ll)
8)
5.5 (2.2) 11.0 (3.0)”
5.3 (2.2) 13.3 (3.9)”
NS
8.2 (2.2) 7.4 (1.6) 7.2 (1.7)
-1.71 (1.85) 5.54 (3.18)*
-1.58 (1.70) 8.18 (2.92)”
NS
3.92 (2.70) 3.47 (1.58) 3.19 (2.28)
5.00 (2.23) 8.14 (2.34)’ 6.39 (1.96)d
0.7 (0.4) 1.3 (0.4)
0.7 (0.2) 1.1 (0.3)
0.8 (0.4) 1.1 (0.4) 1.0 (0.3)
1.1 (0.5) 1.2 (0.5) 1.2 (0.6)
Change from prestudy, A vs. B (P value)
HV (cm/yr) Prestudy Yr 1 Yr2’ HV SDS Prestudy Yr 1 Yr 2’ A BA/A CA (yr/yr) Prestudy Yr 1 Yr 2’ Ht SDS (CA) Prestudy Yr 1 Yr2” Ht SDS (BA) Prestudy Yr 1 Yr 2’ PAH SDS - MPH Prestudy Yr 1 Yr2
SDS
-3.60 -2.44
(0.88) (0.53)O
-3.26 -1.70
(1.52) (1.35)”
NS
-1.68 -1.01
(1.04) (1.06)*
-1.33 -0.11
(1.24) (1.00)”
co.05
(n = 8) -1.78 (1.51) -0.84 (1.33)’
NS
(n = 6) -2.07 (0.87) -1.13 (0.88)d
8.9 (1.6) 11.0 (1.8)* 9.6 (1.4)d
-2.18 -1.94 -1.49
(0.90) (1.27) (1.48)
-2.08 -1.20 -0.17
(1.89) (2.12)b (2.49)d
-0.26 0.00 0.07 (n -0.68 -0.60 -0.76
(1.04) (0.96) (0.84) = 8) (1.25) (0.96) (0.63)
-0.24 0.46 1.08 (n -1.59
(1.49) (1.48)d (1.56) = 8) (1.38)
-0.96 (1.37) b -0.15
(1.69)b
