| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Articles |
Department of Pediatrics, University of Washington (C.P.), Seattle, Washington 98105; the Departments of Pediatrics and Pharmacology, University of Missouri (G.K.), and the Section of Pediatric Clinical Pharmacology and Experimental Therapeutics, Children’s Mercy Hospital (G.K.), Kansas City, Missouri 64108; Wyeth-Ayerst Research (D.F.), Princeton, New Jersey 08543; and the Department of Medicine, Tulane University School of Medicine (C.B.), New Orleans, Louisiana 70112
Address all correspondence and requests for reprints to: Catherine Pihoker, M.D., Division of Pediatric Endocrinology, Children’s Hospital and Medical Center, P.O. Box 5371, CH-92, Seattle, Washington 98105.
Abstract
Administration of GH-releasing peptide-2 (GHRP-2) represents a
potential mode of therapy for children of short stature with inadequate
secretion of GH. Requisite information to determine the dosing route
and frequency for GHRP-2 consists of the pharmacokinetics (PK) and
pharmacodynamics (PD) for this compound, neither of which have been
previously evaluated in children. The purpose of this study was to
characterize the PK and PD of GHRP-2 in children with short stature.
Ten prepubertal children (nine boys and one girl; 7.7 ± 2.4 yr
old) received a single 1 µg/kg iv dose of GHRP-2 over 1 min, followed
by repeated (n = 9) blood sampling over 2 h. GHRP-2 and GH
were quantitated by specific RIA methods. PK parameters were calculated
from curve fitting of GHRP-2 and GH vs. time data.
Posttreatment plasma GH concentrations (normalized for pretreatment
values) were used as the effect measurement. PD parameters were
generated using the sigmoid Emax model. Disposition of
GHRP-2 best fit a biexponential function. GHRP-2 PK parameters
(mean ± SD) were: 
= 13.4 ± 9.7
h-1, ß = 1.3 ± 0.3 h-1,
t1/2ß = 0.55 ± 0.14 h,
AUC0
= 2.02 ± 1.37 ng/mL·h, Cmax
= 7.4 ± 3.8 ng/mL, plasma clearance = 0.66 ± 0.32
L/h·kg, and apparent volume of distribution = 0.32 ± 0.14
L/kg. PK parameters for GH were: appearance rate constant =
5.9 ± 3.1h-1, elimination t1/2 =
0.37 ± 0.15 h, lag time = 0.05 ± 0.01 h,
Cmax = 50.7 ± 17.2 ng/mL, Tmax =
0.42 ± 0.16 h, and AUC0 = 47.9 ±
26.1 ng/mL·h. PD parameters for GHRP-2 were: Ke0 =
1.13 ± 0.94 h-1, 
= 13.15 ± 9.44,
E0 = 6.63 ± 4.86 ng/mL (GH), Emax =
67.5 ± 23.5 ng/mL (GH), and EC50 = 1.09 ± 0.59
ng/mL. We concluded that 1) GHRP-2 produced a predictable and
significant (i.e. compared to pretreatment values)
increase in plasma GH concentrations; 2) the PK-PD link model enabled
quantitative assessment of GHRP-2 modulation of serum GH levels; and 3)
definition of the EC50 for GHRP-2 will enable PD and PK
evaluations of extravascular dosing regimens for children.
THE GH-RELEASING peptides (GHRPs) are a family of molecules that stimulate GH secretion. Originally discovered while searching for a GnRH antagonist, these peptides are structurally related to Met-enkephalin (1, 2, 3). Structural modifications have been made to make the GHRPs more effective, selective GH secretagogues. The most potent GHRP to date in humans is GHRP-2 (3, 4). The pharmacology of GHRPs in man has been characterized and well described previously (5, 6, 7, 8, 9).
A major potential clinical use for GHRP-2 is stimulation of GH secretion when endogenous secretion is inadequate. In many children with short stature and poor growth rates, the problem appears to be insufficient GH secretion, not an inability to produce GH. Evidence for this includes demonstration of subnormal 12- and 24-h GH pulsatile secretion profiles and a robust GH response when GHRH or GHRP-2 is administered to these children (10, 11, 12, 13, 14). Thus, this patient population may potentially benefit from treatment with a GH secretagogue such as GHRP-2.
A feature of the GHRP-2 that makes it an attractive treatment modality resides with its potential for administration by noninvasive methods. Previous investigations have demonstrated release of endogenous GH after the administration of GHRP-2 via the oral, sc, and intranasal routes as well as iv (3, 15, 16, 17). This is a possible advantage over GH or GHRH, both of which must be administered parenterally. However, before large scale studies of the safety and efficacy of GHRP-2 can be conducted in children with GH deficiency, the pharmacokinetics (PK) and pharmacodynamics (PD) of this agent in pediatric patients must first be characterized. Therefore, we examined the PK and PD of iv GHRP-2 in children of short stature who were undergoing evaluation for GH deficiency.
Subjects and Methods
This study was of open design, with no blinding of the subjects
or investigators. Ten short prepubertal children comprised the study
population. These children were all at least 2 SD below the
mean height for age, with mean of -3.03 ± 0.16 SD.
None of the children had evidence of chronic disease, syndromic
disorder, or skeletal dysplasia. The single female patient had a normal
karyotype. All children had a slow growth rate, delayed bone age, and
low serum insulin-like growth factor I levels (Table 1
). Responses to GH stimulation tests,
using standard secretagogues (e.g. arginine, insulin, or
L-dopa) were variable, with five patients whose maximal GH
response was more than 10 µg/L and five patients whose GH response
was less than 10 µg/L. Six of the children (including the five whose
maximal GH response was >10 µg/L) had 12-h, overnight GH secretion
studies. The mean GH concentration was low in each patient; the group
mean was 2.3 µg/L. Magnetic resonance imaging of the head was
performed in each child, and no intracranial lesions were observed.
Subjects were recruited by informed parental consent and, where
appropriate (i.e. age 
7 yr), by patient assent. The
protocol was approved by the human research advisory committee of the
University of Arkansas for Medical Sciences.
|
Quantitation of GH from each serum specimen was performed using a commercial polyclonal RIA (Corning-Nichols Institute, San Juan Capistrano, CA). All GH samples from a given subject were run in the same assay, thereby minimizing the effect of interassay variability on pharmacokinetic profiles. Intra- and interassay coefficients of variation of this RIA method for GH were 3% and 8%, respectively. GHRP-2 serum concentrations were determined by a RIA developed and validated by Wyeth-Ayerst Research (Princeton, NJ), which used a polyclonal antibody provided by Dr. C. Y. Bowers. The calibration curve had a range of 62.5–2000 pg/mL. The interassay coefficients of variation and relative errors ranged from 5.8–8.4% and -4.4 to -7.4%, respectively. The GHRP-2 standards, quality control samples, and study samples were analyzed in triplicate, and the mean values were used in performing calculations.
Individual serum concentration vs. time data for both GH and GHRP-2 were evaluated in each subject by use of the Siphar/Base software package (SIPHAR, version 4.0, SIMED, Creteil-Cedex, France). Initial polyexponential parameter estimates were generated with a peeling algorithm (18). Final parameter estimates were obtained from curve fitting of individual datasets using a nonlinear, weighted, least squares algorithm, with the weight set as the reciprocal of the calculated plasma concentration (19). Compartment model selection was made after application of the Akaike information criterion (20). Additionally, goodness of fit for the serum concentration vs. time data was evaluated by assessment of the variance-covariance matrixes and the coefficients of variation for each polyexponential parameter (e.g. coefficients and exponents) calculated from a given model. Finally, model-dependent pharmacokinetic parameters were calculated for both GH and GHRP-2 according to previously described methods (21).
The pharmacodynamics of GHRP-2 were examined by normalization of the
posttreatment serum GH values to the mean of two pretreatment
measurements to obtain the discrete value of 
GH. For the
pharmacodynamic analysis, the GH values were used as a surrogate
biomarker for the pharmacodynamic effect exerted by GHRP-2 in each
subject. Visual inspection of the serum concentration vs.
time data for both GHRP-2 and GH (Fig. 1)
and also of the GH vs. time values (data not shown)
revealed a discordance in the serum concentration profiles compatible
with an apparent equilibration delay between the serum concentrations
of GHRP-2 and its pharmacological effect, as reflected by serum GH
concentrations. Thus, the data suggested that a
pharmacokinetic-pharmacodynamic model could be constructed that links
the concentration of drug in the central (i.e. serum)
compartment with the concentration of drug in an effect compartment
(22). Accordingly, concentration vs. effect profiles were
then generated from the GH values for each subject by use of the
sigmoid maximal effect model described by Holford and Sheiner (22).
Pharmacodynamic parameter estimates [maximal effect
(Emax) serum concentration associated with 50%
response as measured by post-treatment increase in serum concentration
(EC50), and sigmoidicity constant ()] were then
calculated using an extended least squares algorithm with variance
assumed to be constant (i.e. a homoscedastic model). The
pharmacodynamic analysis was conducted using programs contained within
the Siphar/D software package (SIPHAR, version 4.0, SIMED).
|
= 0.05. Results
Ten short children received a single iv dose of GHRP-2 and
completed the entire study. All subjects tolerated the infusion of
GHRP-2 and all study-related procedures without apparent adverse
effects. As illustrated in Fig. 1
, the mean (±SD) serum
GHRP-2 concentration vs. time curve revealed a multiphasic
relationship that was best fit using a biexponential function. The
elimination and distribution rate constants (i.e. and
ß, respectively) from the curve fit of the mean data were 1.32 and
10.58 h-1, respectively, which yielded the following
generalized equation that best described the mean serum vs.
time concentration profile for GHRP-2: Cptx = 1.51 x
e-1.32 x tx + 13.11 x e-10.58 x
tx, where Cptx represents the plasma concentration
of GHRP-2 at any given time point (i.e. tx). The excursion
of serum concentrations ranged from 7.4 ± 3.8 to 0.15 ±
0.14 ng/mL over a period of 0.083–2.0 h, respectively, after
administration of the GHRP-2 dose. A curve fit of the mean serum GHRP-2
vs. time data for all subjects revealed a rapid distribution
() phase with a mean distribution half-life
(t1/2) of 0.06 h, followed by a longer
terminal elimination half-life (t1/2ß) with a mean
value of 0.52 h.
The polyexponential parameter estimates for GHRP-2 that resulted from
the curve fits of the serum concentration vs. time data for
each subject were used to calculate relevant pharmacokinetic
parameters, which are contained and summarized in Table 2. In each instance, a biexponential
relationship provided the best fit of the serum concentration data.
|
) also revealed a multiphasic relationship that was best
fit using a biexponential function. The appearance and elimination rate
constants (i.e. Kapp and Kel,
respectively) from the curve fit of the mean serum concentration
vs. time data were 4.97 and 1.75 h-1,
respectively, which yielded the following generalized equation that
best described the mean serum concentration vs. time
concentration profile for GH: Cptx = 131.62 x
e-1.75 x tx + 158.83 x e-4.97 x
tx, where Cptx represents the plasma concentration
of GH at any given time point (i.e. tx). The apparent peak
serum GH concentration from these data was 44.8 ± 19.0 ng/mL.
Serum GH concentrations declined in an apparent monoexponential fashion
after the attainment of the peak concentration (i.e.
Tmax) at approximately 0.5 h and declined to 4.8
± 7.8 ng/mL 2 h after GHRP-2 administration. The curve fit of
these data (Fig. 1) revealed a mean first order appearance rate
constant (Kapp) of 4.97 h-1 with a
corresponding mean appearance t1/2 of 0.14 h, and a
mean apparent elimination rate constant (Ke) of 0.39
h-1 with a corresponding elimination t1/2 of
0.4 h.
Pharmacokinetic parameters for GH were calculated using the
polyexponential parameters resulting from the curve fits of serum
concentration vs. time data in each subject and are
contained and summarized in Table 3. As
was true for the analysis of GHRP-2 data, a biexponential relationship
provided the best fit of the serum GH concentration vs. time
data for each subject (Fig. 1). Three of the 10 subjects did not have
quantifiable serum GH concentrations at the 2 h posttreatment
collection point. This did not influence the determination of the
pharmacokinetic parameters for these subjects, as each of them had a
sufficient number of serum concentration vs. time points in
the elimination phase to produce a reliable estimate of the apparent
terminal elimination rate constant.
|
and 3, respectively) for apparent developmental dependence
through attempts to correlate them with patient age revealed no
statistically significant linear or nonlinear relationships. These same
findings were apparent when the pharmacodynamic parameters for GHRP-2
(Table 4) were examined.
|
GHt) data
produced a counterclockwise hysteresis loop, which was consistent with
an equilibration delay between the appearance of GHRP-2 in serum and
its pharmacological effect (i.e. the stimulation of GH
release and its appearance in serum). This particular pharmacodynamic
relationship was present for all 10 subjects. Consequently, application
of the sigmoid Emax model enabled calculation of the
pharmacodynamic parameters for GHRP-2 using GH as the surrogate marker
of its pharmacological activity. These parameters are summarized in
Table 4. Discussion
The pharmacokinetics of GHRP-2 found in our cohort of pediatric patients are similar to those previously reported in healthy adult volunteers after iv administration of the peptide (3). A comparison of the maximum GH response observed after GHRP-2 administration between these two studies revealed similarities in both the magnitude (i.e. mean values = 44 µg/L in children vs. 55 µg/L in adults) and time of maximal response (i.e. average values = 45–60 min for both). The GH responses observed after iv or sc GHRP-2 are also similar to those previously reported after the parenteral administration of GHRP-6, GHRP-1, or GHRH (3, 4, 23, 24).
To our knowledge, our data represent not only the first report of
GHRP-2 pharmacokinetics in pediatric patients, but also the first
pharmacodynamic assessment of this peptide. Comparison of the serum
concentration vs. time profiles for both GHRP-2 and GH in
our subjects reveals an equilibration delay in the attainment of peak
GH response, a period that we believe corresponds to the time course of
GHRP-2 action. This assertion is supported in part by the consistent
observation of an equilibration delay between the serum concentrations
of GHRP-2 vs. effect (i.e. GHt)
curves, reflected by the production of a counterclockwise hysteresis
and our success in using the sigmoid Emax model
to effectively determine the pharmacodynamic parameters for GHRP-2. As
previously reported by Holdford and Sheiner (22), the successful
application of this pharmacodynamic model suggests both linearity and
predictability in the drug concentration vs. effect
relationship. Given the fact that GH is a proximate biological marker
of GHRP (and presumably, GHRP-2) activity (23, 24), our assumptions
entailed in the pharmacodynamic analysis of our data appear valid and
reflective of the expected pharmacological response of GHRP-2.
Despite the apparent differences in serum GH pharmacokinetics reported after exogenous administration of the hormone (25) as opposed to the administration of GH secretagogues (26, 27, 28, 29, 30), both the pharmacokinetic and pharmacodynamic data from our study can be used to address the potential therapeutic efficacy of GHRP-2 in pediatric patients with GH insufficiency. First, the mean AUC for GH after the iv administration of a single 1 µg/kg dose of GHRP-2 (i.e. 50.7 ng/mL·h) was approximately 50% of the AUC at steady state (i.e. 114.2 ± 32.7 ng/mL·h) previously reported in a study of pediatric patients who received daily sc doses of 43 µg/kg methionyl GH (25). If one assumes linearity in the dose-response relationship for iv GHRP-2, administration of a single 2 µg/kg iv dose would be expected to produce an AUC for GH that would be virtually identical to that observed under steady state conditions after sc administration of the currently recommended daily doses of recombinant human GH (25), doses that have been shown to produce acceptable rates of linear growth in children who are GH deficient (30). Second, both the Cmax (mean, 50.7 ng/mL) and Emax values for GHRP-2 in our patient cohort (mean GH, 67.5 ng/mL) actually exceeded the average Cmax values for GH (37.6 ± 11.6 ng/mL; range, 17.6–49.5 ng/mL) after a single sc dose of 0.1 mg/kg methionyl GH to GH-deficient children (25). Finally, the EC50 for GHRP-2 in our study cohort (1.1 ± 0.6 ng/mL) was substantially less than the Cmax value (7.4 ± 3.8 ng/mL). This particular finding not only supports the adequacy of the 1 µg/kg iv dose of GHRP-2 in producing a desirable biological effect, but also suggests that extravascular administration of this peptide by a route that could be associated with up to a 50% reduction in bioavailability may still produce an acceptable increase in the serum GH concentration sufficient to initiate and sustain a desired growth response. This hypothesis is being tested by our group in dose-ranging studies of oral and intranasal GHRP-2 that are currently underway.
In conclusion, both the pharmacokinetics and pharmacodynamics of iv administered GHRP-2 in short children are predictable and reflective of the potential for therapeutic application of this peptide. The data produced in this investigation will enable the selection of GHRP-2 doses for future evaluation of their bioavailability, safety, tolerance, and efficacy in children.
Acknowledgments
The authors gratefully acknowledge the analytical support provided by Drs. Ronald Jordan and Boas Gonen from Wyeth Ayerst Research, and the clinical contributions of Nancy Lowery, R.N., C.C.R.C., in the conduct of this investigation.
Footnotes
1 This work was supported in part by a grant in aid from Wyeth Ayerst
Research (Philadelphia, PA) and Grant 1U01-HD-31313–04 (Network of
Pediatric Pharmacology Research Units; to G.L.K.) from the NICHHD
(Bethesda, MD). Presented in part at the Annual Meeting of the Southern
Society for Pediatric Research, New Orleans, LA, February 1,
1996. 
Received March 7, 1997.
Revised December 30, 1997.
Accepted January 12, 1998.
References
This article has been cited by other articles:
 |
|
 |
|
  J. D. Veldhuis, W. S. Evans, and C. Y. Bowers Impact of Estradiol Supplementation on Dual Peptidyl Drive of GH Secretion in Postmenopausal Women J. Clin. Endocrinol. Metab., February 1, 2002; 87(2): 859 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Anderson, N. Shah, W. S. Evans, J. T. Patrie, C. Y. Bowers, and J. D. Veldhuis Short-Term Estradiol Supplementation Augments Growth Hormone (GH) Secretory Responsiveness to Dose-Varying GH-Releasing Peptide Infusions in Healthy Postmenopausal Women J. Clin. Endocrinol. Metab., February 1, 2001; 86(2): 551 - 560. [Abstract] [Full Text]
|
|
|
|
|
|
N. Shah, W. S. Evans, C. Y. Bowers, and J. D. Veldhuis Oral Estradiol Administration Modulates Continuous Intravenous Growth Hormone (GH)-Releasing Peptide-2-Driven GH Secretion in Postmenopausal Women J. Clin. Endocrinol. Metab., August 1, 2000; 85(8): 2649 - 2659. [Abstract] [Full Text]
|
|
|
|
|
|
N. Shah, W. S. Evans, C. Y. Bowers, and J. D. Veldhuis Tripartite Neuroendocrine Activation of the Human Growth Hormone (GH) Axis in Women by Continuous 24-Hour GH-Releasing Peptide Infusion: Pulsatile, Entropic, and Nyctohemeral Mechanisms J. Clin. Endocrinol. Metab., June 1, 1999; 84(6): 2140 - 2150. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
