Oral administration of growth hormone (GH)-releasing peptide stimulates GH secretion in normal men

Oral Administration of Growth Hormone (GH) ReleasingPeptide-Mimetic MK-677 Stimulates the GH/Insulin-LikeGrowth Factor-I Axis in Selected GH-Deficient Adults*

IAN M. CHAPMAN†, ORA H. PESCOVITZ, GAIL MURPHY, THERESA TREEP,KRISTINE A. CERCHIO, DAVID KRUPA, BARRY GERTZ, WILLIAM J. POLVINO,EMILY H. SKILES, SUZAN S. PEZZOLI, AND MICHAEL O. THORNER

Division of Endocrinology and Metabolism, Department of Medicine, University of Virginia,Charlottesville, Virginia 22908 (I.M.C., E.H.S., S.S.P., M.O.T.); Indiana University, Indianapolis,Indiana 46202 (O.H.P., T.T.); and Merck Research Laboratories, Rahway, New Jersey 07065 (G.M.,K.A.C., D.K., B.G., W.J.P.)

ABSTRACTTo determine the effect of the GH releasing peptide (GHRP)-mi-

metic, MK-677, on the GH/insulin-like growth factor-I (IGF-I) axis inselected GH-deficient adults, we studied nine severely GH-deficientmen [peak serum GH concentration in response to insulin-inducedhypoglycemia of 1.2 6 1.5 mg/L, mean 6 SD (range 0.02–4.79)], age17–34 yr, height 168 6 1.5 cm, body mass index 22.6 6 3.3 kg/m2, whohad been treated for GH deficiency with GH during childhood. In adouble-blind rising-dose design, subjects received once daily oraldoses of 10 or 50 mg MK-677 or placebo for 4 days over two treatmentperiods separated by at least 28 days. Four subjects received placeboand 10 mg/day MK-677 in a cross-over fashion in periods 1 and 2. Fivesubjects received 10 mg and then 50 mg/day MK-677 in a sequential,rising-dose fashion in periods 1 and 2, respectively. Blood was col-lected every 20 min for 24 h before treatment and at the end of eachperiod for GH measurement using an ultrasensitive assay. The drugwas generally well tolerated, with no significant changes from base-line in circulating concentrations of cortisol, PRL, and thyroid hor-mones. Serum IGF-I and 24-h mean GH concentrations increased inall subjects after treatment with both 10 and 50 mg/day MK-677 vs.baseline. After treatment with 10 mg MK-677, IGF-I concentrations

increased 52 6 20% (65 6 6 to 99 6 9 mg/L, geometric mean 6intrasubject SE, P # 0.05 vs. baseline), and 24 h mean GH concen-trations increased 79 6 19% (0.14 6 0.01 to 0.26 6 0.02 mg/L, P # 0.05vs. baseline). Following treatment with 50 mg MK-677, IGF-I con-centrations increased 79 6 9% (84 6 3 to 150 6 6 mg/L, P # 0.05 vs.baseline) and 24-h mean GH concentrations increased 82 6 29%(0.21 6 0.02 to 0.39 6 0.04 mg/L, P # 0.05 vs. baseline), respectively.Serum IGF binding protein-3 concentrations increased with both 10mg (1.2 6 0.1 to 1.7 6 0.1 mg/L, P # 0.05) and 50 mg MK-677 (1.7 60.1 to 2.2 6 0.2 mg/L, P # 0.05). The GH response to MK-677 wasgreater in subjects who were the least GH/IGF-I deficient at baseline;by linear regression analysis the increase in 24-h mean GH concen-tration was positively related to both baseline 24-h mean GH con-centration (r 5 0.81, P 5 0.009) and baseline IGF-I (r 5 0.79, P 5 0.01)for 10 mg MK-677. IGF-I responses were not significantly related toany baseline measurement. Fasting and postprandial insulin andpostprandial glucose increased significantly after MK-677 treatment,and the clinical significance of these changes will need to be assessedin longer term studies. Oral administration of such GHRP-mimeticcompounds may have a role in the treatment of GH deficiency ofchildhood onset. (J Clin Endocrinol Metab 82: 3455–3463, 1997)

GH IS secreted by the somatotrophes of the anteriorpituitary gland in multiple pulses each day. It acts

both directly and via a stimulatory effect on the productionand action of insulin-like growth factor-I (IGF-I) to stimulatelinear growth before epiphysial fusion and to exert a numberof metabolic effects throughout life. Childhood GH defi-

ciency is an important cause of short stature. The effects ofGH deficiency at any age include decreased muscle mass andstrength (1, 2), increased fat mass (1, 2), and unfavorablealterations in blood lipid concentrations (1, 3, 4) that mayhasten the development of vascular disease (5) and lead toincreased mortality from cardiovascular disease (6).

There is a general consensus that children with short stat-ure and GH deficiency should be treated to maximize theirgrowth potential. Increasingly, adult GH deficiency is alsobeing treated. Until now, the treatment for GH deficiency hasconsisted of GH injections, usually administered daily. Al-though effective, GH treatment has a number of drawbacks,including the need for parenteral administration and highcost. These have provided an incentive to develop GH secre-tagogues that are effective when taken orally.

GH releasing peptide (GHRP-6) is a synthetic hexapeptidethat stimulates GH secretion (7, 8). It appears to act directlyon the pituitary (9–11) and also on the hypothalamus (12, 13).Oral administration of GHRP-6 is able to stimulate GH se-cretion, but its bioavailability is much greater after parenteraladministration (14). Recently, a number of other compounds

Received March 18, 1997. Revision received June 12, 1997. AcceptedJune 23, 1997.

Address all correspondence and requests for reprints to: Michael O.Thorner, Department of Medicine, Box 466, University of VirginiaHealth Sciences Center, Charlottesville, Virginia 22908. E-mail:[email protected].

* This work was supported by a grant from Merck Research Labo-ratories and in part by grants from the NIH (DK-32632 to M.O.T.,R.R.-00847 to the General Clinical Research Center and ComputerizedData Management and Analysis Systems Laboratory at the Universityof Virginia and MO1 R.R.-00750 to Indiana University General ClinicalResearch Center), and a National Science Foundation Center for Bio-logical Timing Grant DIR89–20162 (to M.O.T.).

† Supported in part by a C.R.B. Blackburn Overseas Traveling Fel-lowship of the Royal Australasian College of Physicians, and a MarkJolley Fellowship of the South Australian Postgraduate Medical Edu-cation Association.

0021-972X/97/$03.00/0 Vol. 82, No. 10Journal of Clinical Endocrinology and Metabolism Printed in U.S.A.Copyright © 1997 by The Endocrine Society

3455

have been developed that mimic the GH stimulatory actionsof GHRP and have greater oral bioavailability and durationof action. Intravenous administration of one of these com-pounds, the nonpeptide L-692,429, has been shown to stim-ulate GH secretion when given acutely to healthy young andolder adults (15, 16) and as 12- and 24-h continuous infusionsto older adults (17).

The present study was designed to determine the effect ofshort-term oral administration of another of these com-pounds, the spiropiperidine MK-677, on the GH/IGF-I axisin selected adults with GH deficiency. Because it acts byincreasing somatotrophe secretion of GH, MK-677 would notbe expected to increase circulating GH in individuals withabsent pituitaries or severely damaged somatotrophes.Therefore, we studied GH- deficient young adults who hadbeen diagnosed with GH deficiency during childhood andwho had not had pituitary or hypothalamic tumor, surgery,or radiotherapy. GH deficiency in such subjects is often id-iopathic, and in the majority of cases, the deficiency persistsinto adulthood. The GH deficiency is thought to be causedby a functional deficiency of GH stimulatory signals to thepituitary (18). We hypothesized that MK-677 treatmentwould be well tolerated and increase circulating GH andIGF-I concentrations in adults with this condition.

Subjects and Methods

The study was approved by the Human InvestigationCommittees of the participating centers. Each subject gavewritten, informed consent before enrollment in the study.

Nine men, age 17–34 yr with idiopathic GH deficiency ofchildhood onset were studied. All had been treated with GHduring childhood, but had not been treated with GH or anyGH secretagogue for at least 6 months before taking part inthis study. None had a past history of pituitary or hypotha-lamic tumor, surgery, or radiotherapy. Persistent GH defi-ciency was confirmed by a reduced GH response to insulin-induced hypoglycemia; on a prestudy insulin tolerance testall subjects had a peak serum GH concentration of ,5 mg/Lwith at least one glucose concentration #45 mg/dL after0.1–0.15 U/kg iv regular insulin. (The insulin tolerance testfor four of the subjects enrolled was discontinued after 30min because of severe hypoglycemic symptoms.) All subjects

were generally in good health on the basis of medical history,physical examination, and laboratory screening. Subjects hadnormal urinalyses, electrocardiograms, and chest x-rays, andnormal serum concentrations of biochemical indices of renal,hepatic, and hematological function, thyroid function stud-ies, PRL, and testosterone, except where stated in Table 1.Subjects were not taking any medications on a regular basis,apart from replacement doses of thyroid hormone, glucocor-ticoids, and testosterone, which were stable for at least 6months prestudy, and occasional (documented) acetamino-phen or ibuprofen. One of the subjects was on adequatetestosterone replacement for hypogonadism throughout thestudy. Based on plasma testosterone concentrations at thetime of screening, six of the nine subjects were hypogonadalduring this study (Table 1). Two of these six subjects werecurrently being treated with monthly testosterone injections(treatment begun at least 4 yr before the study). Their lowtestosterone levels at screening may reflect an inadequatereplacement regimen and/or measurement of testosteronejust before testosterone injection. One subject had previouslybeen treated with testosterone during adult life, but hadstopped it at least 3 months before the study. The other threehad not been treated with testosterone as adults. None of thesubjects started testosterone or any other regular medicationduring the study. Prestudy subject characteristics are sum-marized in Table 1. Four potential subjects who had beentreated with GH during childhood for a presumptive diag-nosis of GH deficiency were excluded from the study becauseof a peak GH response to hypoglycemia greater than 7 mg/L.

Study design

Subjects were entered into a randomized, double-blind, placebo-con-trolled study in which they received the study drug (MK-677 or placebo)orally each day for 4 days during each of two study periods, separatedby a 28- to 150-day washout period (Fig. 1). There were two studygroups: group I (n 5 4) received placebo and 10 mg MK-677 in across-over fashion in periods 1 and 2; Group II (n 5 5) received 10 mgMK-677 in the first study period, then 50 mg MK-677 in the second studyperiod.

Study period 1

Subjects were admitted to the Clinical Research Center in the evening.An intravenous cannula was inserted into an arm vein for subsequent

TABLE 1. Prestudy subject characteristics

Patientno.

Age(yr)

Height(cm)

BMI(kg/m2)

Total duration ofGH treatment

(ages)

IGF-I(mg/L)a

Peak GHresponse to

insulin (mg/L)

Testosterone(ng/dL)b Other diagnoses/other medications

1 34 151 19.3 5 yr (16–21) 90.5 1.05 93 Hypogonadal2 22 161 21.6 11 yr (6–18) 103.8 0.57 36 Hypogonadal3 23 168 26.5 10 yr (6–17) 50.0 3.06 454 Hypothyroid/L-thyroxine4 26 173 24.1 7 yr (14–22) 89.8 0.2c 490 Panhypopituitarism/hydrocortisone, L-thyroxine, testosterone5 21 162 19.5 10 yr (8–19) 34.5 0.11c 40 Hypogonadal, scoliosis, mild mental retardation/

testosterone6 21 159 21.5 10 yr (8–19) 23.8 0.11c 59 Hypogonadal, scoliosis, mild mental retardation/

testosterone7 17 183 19.6 14 yr (2–17) 151.0 4.8 229 Partial unilateral sensorineural deafness8 22 174 23.8 6 yr (12–19) 34.8 0.03c 505 None9 17 178 28.5 10 yr (5–16) 135.0 0.79 56 Hypogonadal, previous cryptorchidisma Normal range, 202–456 mg/L.b Normal range, 300–1000 ng/dL.c Insulin tolerance test discontinued after 30 min because of severe hypoglycemic symptoms.

3456 CHAPMAN ET AL. JCE & M • 1997Vol 82 • No 10

blood sampling, and subjects spent the night in the Research Center toequilibrate to sleep conditions (day 22). At 2240 h the following nightsubjects received an oral dose of single-blind placebo (day 21). At 2240 hfor the next four nights they received an oral dose of double-blind studydrug (MK-677 or placebo). Each dose was taken with 150 mL water, andlights were turned off immediately following dosing. Blood samples forGH measurement were collected through the indwelling cannula at20-min intervals for 24 h after the single-blind placebo dose on the firstnight (2240 h day 21) and for 8 h after the first administration of studydrug at 2240 h (day 1). Samples were again collected for 24 h after thefourth day of study drug administration.

All overnight blood sampling was performed from outside the sub-ject’s room through long tubing to minimize disturbance of the subjects.Additional blood samples were drawn at designated time points formeasurement of PRL, cortisol, thyroid hormones, IGF-I, and IGF-I bind-ing protein-3 (IGFBP-3). Two 24-h urine collections were saved formeasurement of urinary free cortisol and creatinine starting at the timeof single-blind placebo dosing (day 21) and at the start of the fourth doseof study drug administration (day 4). Vital signs (heart rate and bloodpressure) were measured and an electrocardiogram was performed 9 hpostdose each morning, and a physical examination and routine labo-ratory tests were performed before discharge after completion of thetreatment period. A physical examination, an electrocardiogram, andcollection of a fasting blood sample for laboratory testing were per-formed 5–7 days after completion of each period.

During all admissions, subjects were required to stay awake until justafter dosing at 2240 h, at which time lights were turned out. Alcoholconsumption was not permitted, and regular research center diets wereconsumed by all subjects, with the following exceptions: an eveningsnack was served at approximately 2000 h on sampling days, and sub-jects then fasted until breakfast the following morning; and to assesspeak glucose and insulin concentrations, subjects consumed 8 oz (237mL) Sustacal Plus (Mead Johnson Nutritionals, Evansville, IN) insteadof breakfast on two occasions. This liquid nutrition supplement [con-taining 360 cal made up of 14 g fat (98 cal), 45 g carbohydrate (180 cal),and 14 g protein (56 cal)] was given the next morning 10 h after the day21 placebo dose and 10 h after the fourth dose of study drug. On theseoccasions, blood samples for glucose and insulin measurement wereobtained immediately before the Sustacal and 0.5, 1, 2, and 4 h followingadministration.

Subjects were required to remain in the research unit on samplingdays (days 22, 21, 1, and 4). On days 2 and 3 they were permitted toleave the unit as outpatients and self-administer the study drug at home.

Study period 2

The same procedures as study period 1 were performed, with theseexceptions: the dose of treatment drug was different (see above and Fig.

1), and after the equilibration night, the single-blind day 21 placeboadministration was omitted.

Analytic methods

Assays. Serum GH concentrations were measured in duplicate by achemiluminescence assay (Nichols Institute Diagnostics, San JuanCapistrano, CA) modified to enhance sensitivity as previously described(19). All GH assays were performed in the same laboratory, and allsamples from a single subject were run in the same assay. The sensitivityof the assay was 0.002 mg/L, and the measured GH concentrations in allsamples were above this detection limit. The intraassay coefficients ofvariation were 10.1% at 0.03 mg/L, 8.1% at 0.3 mg/L, and 15.0% at 6.8mg/L. Interassay coefficients of variation were 7.4% at 0.03 mg/L, 18.5%at 0.3 mg/L, and 8.8% at 6.8 mg/L. Cortisol, PRL, IGF-I, IGFBP-3, thyroidfunction, urine free cortisol, glucose, and insulin assays were performedby Endocrine Sciences Laboratories (Calabasas Hills, CA). Routine lab-oratory analyses were performed at the respective study sites, andtestosterone was measured at the University of Virginia Medicine Clin-ical Laboratory.

Analysis of pulsatile GH release. The GH concentration profiles were an-alyzed by the cluster peak detection program version 6.0 (20). Thethreshold parameters used (test peak 5 1, test nadir 5 1, t statistic 5 1)had a sensitivity of 75% for detection of GH concentration pulses anda positive predictive accuracy of 93% determined in a validation studyemploying computer simulations of 30 24-h GH series at 20-min inter-vals, in which the exact locations of the secretory episodes and conse-quent GH concentration peaks were known.

Statistical methods

Data for the 10 mg MK-677 dose in the two study groups werecombined (n 5 9). Absolute and percentage changes from baseline werecalculated. Baseline GH values were those at the beginning of period 1;all other baseline values were those at the beginning of the treatmentperiod under analysis. t tests were calculated to compare values aftertreatment with MK-677 and placebo with baseline levels. ANOVA wasused to obtain a pooled estimate of within-subject standard deviation sothat the t tests comparing levels following placebo or 10 mg or 50 mgMK-677 with baseline levels could be calculated with the same estimateof variability. For some parameters, it was necessary to transform datato the natural log scale to satisfy basic statistical assumptions, in whichcase results are reported as geometric mean 6 an appropriately back-transformed estimate of within-subject se. Otherwise, results are re-ported as mean 6 se. The design of this study, which was chosen in lightof expected difficulties recruiting GH-deficient adults and safety con-cerns common to initial studies of a new drug, prevented a definitivestatistical assessment of differences between treatments. The relation-ship between measures was determined by univariate linear regressionanalysis. A P value of #0.05 was considered statistically significant.

ResultsResponse to treatment

Oral treatment with both 10 mg and 50 mg/day MK-677was associated with statistically significant increases in cir-culating concentrations of GH, IGF-I, and IGFBP-3, whereasplacebo treatment was without a significant effect on theseparameters (Table 2).

GHAll subjects had baseline 24-h mean GH concentrations

below the age-matched normal range (geometric mean 1.16,range 0.67–2.35 mg/L) in this assay. Twenty-four hour meanGH concentrations were higher than baseline concentrationsin all subjects after treatment with both doses of MK-677 (Fig.2), and there were similar stimulatory effects by the twodoses (Table 2). Representative individual 24-h GH concen-tration profiles are shown in Fig. 3, and 24-h mean concen-

FIG. 1. Study design. Randomized, double-blind, placebo-controlled,rising-dose study. Placebo, 10 mg/day or 50 mg/day MK-677 admin-istered orally (with 150 mL water) once a day at approximately2240 h. Sustacal Plus is a liquid nutrition supplement containing 360cal made up of 14 g fat (98 cal), 45 g carbohydrate (180 cal), and 14 gprotein (56 cal).

MK-677 STIMULATES GH/IGF-I IN GHD ADULTS 3457

tration profiles are shown in Fig. 4. The 24-h mean GHconcentration (mg/L) rose 79 6 19% (range 28–219) frombaseline (0.14 6 0.01 to 0.26 6 0.02, P # 0.05) after four dosesof 10 mg MK-677 and increased 82 6 29% (3 to 165) frombaseline (0.21 6 0.02 to 0.39 6 0.04, P # 0.05) after four dosesof 50 mg MK-677. However, 24-h mean GH concentrationsremained below the age-adjusted normal range in all subjectsafter 4 days treatment. Eight hour mean GH concentrations(mg/L) were significantly greater after the first than thefourth dose of both 10 mg (0.56 6 0.07 vs 0.35 6 0.04, P # 0.05)and 50 mg MK-677 (0.99 6 0.16 vs 0.54 6 0.09, P # 0.05), andwere significantly greater than baseline at both times.

Linear regression analysis indicated that the absoluteincrease in mean GH concentrations in response to MK-677treatment was greater in subjects who were the least GH/IGF-I deficient at baseline. The increase in 24-h mean GHconcentration after 10 mg MK-677 (calculated as the dif-ference between the 24-h mean GH concentration afterfour doses of MK-677 vs. baseline) was positively relatedto baseline measurements of both 24-h mean GH, with acorrelation coefficient of r 5 0.81 (P 5 0.009) and IGF-Iconcentration of r 5 0.79 (P 5 0.012), and the correlationwith the peak GH concentration after insulin-induced hy-poglycemia approached statistical significance with r 50.62 (P 5 0.075). The correlation between the change in24-h mean GH concentration following 50 mg MK-677 andbaseline measurements of 24-h mean GH had a correlationcoefficient of r 5 0.56 (P 5 0.3) and with IGF-I of r 5 0.71(P 5 0.18), and peak GH concentration following insulin-induced hypoglycemia of r 5 0.83 (P 5 0.08). The baselineserum testosterone concentration did not correlate with

baseline 24-h mean GH or IGF-I concentrations or to theincrease in GH concentrations produced by either dose ofMK-677.

The results of cluster analysis of 24-h GH concentrationprofiles are summarized in Table 2. The increase in GHconcentrations produced by MK-677 treatment was becauseof an increase in GH pulse, amplitude, and interpeak valley

FIG. 2. Left panel, Individual 24-h mean GH responses (mg/L) atbaseline (n 5 9) and at end of day 4 of 10 mg/day (n 5 9) and 50 mg/day(n 5 5) oral MK-677. Note that 24-h mean GH in age-matched normalmen is 1.16 mg/L (range 0.67–2.35) in modified chemiluminescenceGH assay. Right panel, Percentage change from baseline (geometricmean 6 SE) in 24-h mean GH after placebo (n 5 4), 10 mg/day (n 59), and 50 mg/day (n 5 5) oral MK-677. *, P # 0.05 vs. baseline.

TABLE 2. Analysis of 24-h GH concentration profiles and serum IGF-I and IGFBP-3 concentrations at baseline and after four doses oforal treatment with MK-677

Parameter Treatment(per day) N

Geometric mean 6 intrasubject SE

Baseline Day 4 % Change % Change range

24-h mean GH (mg/L)a Placebo 4 0.09 6 0.01 0.04 6 0.01c 249 6 10 262 to 2410 mg 9 0.14 6 0.01 0.26 6 0.02c 79 6 19 28 to 21950 mg 5 0.21 6 0.02 0.39 6 0.04c 82 6 29 3 to 165

Mean GH (mg/L) (0–8 h postdose) Placebo 4 0.11 6 0.02 0.06 6 0.01 243 6 15 255 to 21310 mg 9 0.18 6 0.02 0.35 6 0.04c 96 6 33 5 to 28250 mg 5 0.27 6 0.04 0.54 6 0.09c 100 6 46 24 to 248

GH peak number/24 hb Placebo 4 11 6 1.2 10 6 1.2 29 6 15 230 to 3010 mg 9 10 6 0.6 11 6 0.6 13 6 9 225 to 8050 mg 5 8 6 1.1 11 6 1.1 29 6 21 217 to 133

GH peak height (mg/L)b Placebo 4 0.13 6 0.02 0.07 6 0.01c 244 6 10 261 to 2310 mg 9 0.21 6 0.01 0.38 6 0.02c 79 6 19 21 to 18750 mg 5 0.32 6 0.03 0.49 6 0.05c 55 6 24 27 to 114

GH interpeak nadir (mg/L)b Placebo 4 0.06 6 0.01 0.03 6 0.01c 249 6 9 264 to 2510 mg 9 0.09 6 0.01 0.18 6 0.01c 93 6 22 38 to 24750 mg 5 0.13 6 0.01 0.25 6 0.03c 100 6 33 34 to 165

GH interpeak valley mean (mg/L)b Placebo 4 0.07 6 0.01 0.04 6 0.01c 246 6 10 264 to 2910 mg 9 0.11 6 0.01 0.21 6 0.02c 94 6 21 46 to 23550 mg 5 0.15 6 0.02 0.29 6 0.03c 88 6 31 19 to 164

IGF-I (mg/L) Placebo 4 49 6 2.0 54 6 2.2 9 6 6.3 215 to 4010 mg 9 65 6 6.1 99 6 9.3c 52 6 20 7 to 22650 mg 5 84 6 3.1 150 6 5.5c 79 6 9.2 46 to 200

IGFBP-3 (mg/L) Placebo 4 0.82 6 0.07 0.95 6 0.08 16 6 14 218 to 5210 mg 9 1.16 6 0.10 1.67 6 0.14c 44 6 17 0 to 20950 mg 5 1.65 6 0.12 2.23 6 0.16c 35 6 14 22 to 55

a 24-h mean GH in age-matched normal men is 1.16 (mg/L, geometric mean), range 0.67–2.35 in modified chemiluminescence assay.b Attributes of pulsatile GH release were assessed by the cluster algorithm.c P # 0.05 vs. baseline.


and nadir GH concentrations and not because of increasedpulse frequency.

IGF-IAll subjects had baseline IGF-I concentrations below the

age-adjusted normal range for the assay (202–456 mg/L),further supporting the diagnosis of GH/IGF-I deficiency.IGF-I concentrations increased in all subjects after four dosesof 10 mg or 50 mg MK-677 (Fig. 5), and the mean increasesin IGF-I concentrations were statistically significant withboth doses (Table 2). IGF-I concentrations increased into theage-adjusted normal range after drug treatment in two sub-jects treated with 50 mg/day MK-677. The IGF-I concentra-tion changes after MK-677 treatment were not significantlyrelated to any baseline measurement (all P values .0.2).

IGFBP-3IGFBP-3 concentrations increased in all subjects after treat-

ment with 10 and 50 mg MK-677, and the mean concentra-tions on day 4 were significantly greater than baseline withboth doses (Table 2, Fig. 6).

Insulin, glucose and other hormones. Fasting and post-Sustacalserum glucose and insulin concentrations are shown in Table3. After 4 days of treatment with 50 mg, but not 10 mgMK-677, fasting insulin concentrations were significantlyhigher than at baseline (19.5 6 3.2 vs. 8.3 6 1.4 mU/ml, P #0.05). Fasting serum glucose concentrations were not signif-icantly affected by either dose of MK-677. Post-Sustacal glu-

cose and insulin, measured as both the peak concentrationand area under the curve (AUC), were significantly higherthan baseline after four doses of 10 mg MK-677, and therewere nonsignificant increases in both measures of glucose

FIG. 3. Representative individual 24-h serum GH (mg/L) profiles fortwo GH-deficient subjects at baseline (E) and at end of day 4 of 10mg/day (f) and 50 mg/day (Œ) oral MK-677. Twenty four-hour meanGH and IGF-I data are shown for reference, and patient numberscorrespond to those in Table 1. Note different Y axes.

FIG. 4. Upper panel, Twenty four-hour serum GH profiles (mg/L,arithmetic mean 6 SE) at baseline (E) and at end of day 4 of 10 mg/day(f, n 5 9) oral MK-677. Data for 10-mg dose are combined (see Fig.1). Lower panel, Twenty four-hour serum GH profiles (arithmeticmean 6 SE) at baseline (E) and at end of day 4 of 10 mg/day (f) and50 mg/day (Œ) oral MK-677 doses for those five subjects who receivedboth doses (see Fig. 1). Note different Y axes.

FIG. 5. Left panel, Individual serum IGF-I concentrations (mg/L) atbaseline and on day 4 of 10 mg/day (n 5 9) and 50 mg/day (n 5 5) oralMK-677. Shaded area represents lower portion of normal range. #,Indicates a value at end of day 3 of dosing. Right panel, Percentagechange from baseline (geometric mean 6 SE) in serum IGF-I concen-trations after placebo (n 5 4), 10 mg/day (n 5 9), and 50 mg/day (n 55) MK-677. *, P # 0.05 vs. baseline.


and insulin after treatment with 50 mg MK-677. Post-Sustacalinsulin concentrations were also significantly different frombaseline after placebo treatment. Neither placebo nor MK-677 treatment resulted in significant changes from baselinein circulating concentrations of serum T4, T3, TSH (data not

shown), cortisol, PRL, or 24-h urinary free cortisol levels(Table 3).

Adverse experiences

MK-677 treatment was generally well tolerated and nosymptoms developed that were definitely attributed to studydrug. There were no serious adverse events during thisstudy, and no subjects were discontinued or had treatmentinterrupted because of an adverse experience.

Five out of nine subjects who were treated with 10 mgMK-677 had clinical adverse experiences that the investigatorconsidered to be possibly drug related. These included oneepisode each of headache and diarrhea and three occurrencesof dry skin. The occurrences of dry skin did not requiretreatment or medical consult. Except for the diarrhea, whichwas moderate, these adverse experiences were rated by theinvestigator as mild.

Two out of five subjects who received 50 mg MK-677 hadclinical adverse experiences. One subject had night sweatsand another subject had numbness in the ulnar nerve dis-tribution area of the right hand that lasted one day. Bothadverse experiences were rated as mild.

One subject treated with 10 mg MK-677 demonstrated anincreased serum aspartate amino transferase (98 U/L; nor-mal range 0–50 U/L) at the post period I evaluation. Thiselevation resolved and was rated as possibly drug related bythe investigator.

FIG. 6. Left panel, Individual serum IGFBP-3 responses (mg/L) atbaseline and on day 4 of 10 mg/day (n 5 9) and 50 mg/day (n 5 5) oralMK-677. Shaded area represents normal range. Right panel, Per-centage change from baseline (geometric mean 6 SE) in serumIGFBP-3 concentrations after placebo (n 5 4), 10 mg/day (n 5 9), and50 mg/day (n 5 5) oral MK-677. P # 0.05 vs. baseline. ¥, Two subjectshad same values for baseline and on 10 mg.

TABLE 3. Analysis of glucose, insulin, and other hormone concentrations at baseline and after four doses of oral treatment with MK-677

Parameter Treatment(per day) N

Geometric mean 6 intrasubject SE

Baseline Day 4 % Change % Change range

Fasting glucose (mg/dL) Placebo 4 79 6 7 79 6 7 0 6 13 210 to 1510 mg 9 77 6 4 85 6 5 11 6 9 212 to 10250 mg 5 75 6 6 89 6 7 17 6 13 212 to 140

Peak glucose (mg/dL) post-Sustacal Placebo 4 103 6 8 105 6 8 1 6 11 220 to 1210 mg 9 101 6 4 118 6 5a 16 6 7 215 to 4650 mg 5 100 6 7 120 6 8 20 6 12 29 to 98

Glucose AUC (mgzh/dL) post-Sustacal Placebo 4 342 6 20 362 6 21 6 6 9 28 to 2110 mg 9 338 6 11 389 6 13a 15 6 5 27 to 3750 mg 5 334 6 17 363 6 19 9 6 8 29 to 98

Fasting insulin (mU/mL) Placebo 4 4.7 6 0.9 6.3 6 1.2 35 6 36 12 to 7010 mg 9 6.4 6 0.7 7.7 6 0.9 19 6 19 246 to 14050 mg 5 8.3 6 1.4 19.5 6 3.2a 136 6 56 36 to 350

Peak insulin (mU/mL) post-Sustacal Placebo 4 24 6 6 50 6 13 107 6 78 7 to 49210 mg 9 36 6 6 60 6 9a 67 6 37 259 to 55450 mg 5 50 6 11 87 6 20 74 6 58 257 to 533

Insulin AUC (mUzh/mL) post-Sustacal Placebo 4 55 6 6 108 6 23a 96 6 60 6 to 41410 mg 9 76 6 11 130 6 19a 71 6 36 250 to 36850 mg 5 98 6 11 157 6 29 60 6 43 221 to 179

Mean PRL (mg/L)b Placebo 4 13.5 6 1.1 11.1 6 1.1 218 6 9 249 to 1710 mg 9 13.5 6 0.7 13.9 6 0.7 3 6 7 238 to 7650 mg 5 13.4 6 1.0 12.3 6 1.0 28 6 10 235 to 44

Mean cortisol (mg/dL)b Placebo 4 6.0 6 0.6 4.6 6 0.6 224 6 12 240 to 1110 mg 9 4.6 6 0.8 5.3 6 0.8 17 6 26 274 to 62050 mg 5 3.4 6 0.5 4.9 6 0.5 44 6 26 225 to 462

Urinary free cortisol (mg/day) Placebo 4 21 6 3 23 6 6 8 6 29 230 to 12510 mgc 8 22 6 2 19 6 2 213 6 10 244 to 4050 mgc 4 24 6 2 19 6 5 224 6 13 245 to 24

a P # 0.05 vs. baseline.b Values are arithmetic means 6 SE. PRL and cortisol values are mean of samples collected at times 0, 0.5, 1, 2, 4, and 8 h at baseline (day-1)

and on day 4 of oral MK-677 treatment.c One subject’s data were excluded because of high baseline cortisol levels thought to be caused by hydrocortisone treatment.


Discussion

In this study, once daily oral administration of MK-677 for4 days significantly increased circulating concentrations ofGH, IGF-I, and IGFBP-3 in men with childhood-onset GHdeficiency. None of these adults had previously undergonepituitary or hypothalamic surgery or irradiation, and all hadmeasurable, albeit reduced, levels of serum GH, excludingthe absence of the GH gene as the cause of their deficiency.Therefore, although a greater number of subjects than mighthave been expected had deficiencies of other pituitary hor-mones, all were thought to have idiopathic GH deficiency.This is the most common cause of GH deficiency in child-hood and is believed to result from inadequate stimulationof the pituitary by hypothalamic GHRH rather than from aprimary lesion of the pituitary (18). It is therefore theoreti-cally possible to treat not only with GH, but with agents thatdirectly stimulate GH secretion by somatotrophes. Admin-istration of GHRH, the major endogenous GH secretagogue,stimulates normal or near-normal GH release in the majorityof children with GH deficiency and short stature (21), andchronic treatment with GHRH has been shown to signifi-cantly increase growth rates in these children (18, 22, 23).Unfortunately, like GH, GHRH must be given parenterally.

The increased IGF-I and GH concentrations induced byshort-term oral MK-677 treatment most likely resulted fromactions on both the pituitary and hypothalamus. MK-677 andGHRP-6, whose GH stimulatory actions it is thought tomimic (24), bind to a recently identified unique G protein-coupled receptor on the pituitary (9–11, 25, 26) to directlystimulate somatotrophe GH secretion via phospholipase Cand calcium-dependent mechanisms (27–31). In addition,GHRP and its analogs have a synergistic stimulatory effecton GH secretion when coadministered with GHRH (17, 32).They also increase c-fos activity and electrical activity inhypothalamic arcuate nucleus neurons that secrete GHRH(12, 33, 34). Stimulation of GH release by GHRP-6 and GHRP-mimetics such as MK-677 is therefore likely to be dependent,at least in part, on extrapituitary factors, particularly thebackground GHRH and somatostatin tone. The smaller GHsecretory response to MK-677 treatment in subjects withlower baseline GH and IGF-I concentrations in this studymight reflect a greater degree of GHRH deficiency in thesesubjects, and suggests the possibility of combined therapywith MK-677 and GHRH in such subjects.

Inspection of the GH profiles suggested that MK-677 in-creased serum GH concentrations by enhancing the preex-isting pulsatile pattern of GH release. This was supported bythe results of cluster analysis, which revealed a significantincrease in GH peak height without change in peak number.Interpeak nadir GH concentrations were also significantlyincreased by MK-677 treatment. We have previously admin-istered a compound related to MK-677 (L-692,429) to healthyolder subjects by continuous 12- and 24-h intravenous infu-sions and assessed pulsatility by deconvolution analysis (17).We have also administered daily oral MK-677 to healthyolder subjects for up to 4 weeks and assessed GH pulsatilityby the cluster and ultra algorithms and deconvolution (35).There is agreement among all methods and studies that thesecompounds increase circulating GH concentrations by in-

creasing the size, but not the number, of existing pulses, andthat despite an increase in interpulse GH concentrations, GHsecretion remains pulsatile. This enhancement of GH pulsa-tility occurs whether these compounds are administered con-tinuously as intravenous infusions or as daily administra-tions of the long-acting compound MK-677. This suggeststhat these compounds amplify the normal signals responsi-ble for episodic GH release. This could occur via relief of aninhibitory effect, such as that of somatostatin, enhancementof a stimulatory effect, such as that of GHRH, or a combi-nation of both.

The stimulatory effect of MK-677 on mean GH concentra-tions in the 8 h after drug administration declined betweenthe first and fourth day of drug administration, although thefourth day value was still significantly greater than baseline.This decline may indicate desensitization to the GH stimu-latory effects of the drug and foreshadow an eventual loss ofstimulatory effect. Alternately, and probably more likely, itmay result from negative feedback effects of IGF-I on GHsecretion. There is evidence that IGF-I acts at pituitaryand/or hypothalamic sites to suppress GH secretion (36–39).The negative feedback effects of IGF-I would be expected toincrease as circulating IGF-I concentrations increase in thedays to weeks after starting MK-677 treatment. This wouldeventually result in a new set point at which IGF-I and GHconcentrations are higher than at baseline, but GH concen-trations are lower than immediately after initiation of treat-ment. Such changes have been reported in beagle dogstreated with oral MK-677 for 2 weeks (40).

IGF-I and GH concentrations were significantly increasedafter 2 weeks treatment with oral MK-677 in healthy oldersubjects (35). Moreover, IGF-I concentrations increased fur-ther between 2–4 weeks of treatment, indicating that the fulleffect of MK-677 on IGF-I concentrations took longer than 2weeks to develop, and that significant stimulation of theGH/IGF-I axis was sustained for at least a month. In addi-tion, IGF-I concentrations increased slightly more after dailymorning than evening administration of MK-677 in thatstudy. The drug was administered at night to subjects in thepresent study. We do not yet know whether the response toMK-677 is sustained for as long in persons with idiopathicGH deficiency as in healthy older persons. Nevertheless, theresponse suggests that mean IGF-I concentrations, whichincreased significantly (but not into the normal range) within4 days of starting MK-677 treatment in the present study,may have increased even more if the drug had been admin-istered in the morning rather than at night, and if the treat-ment period had been longer. Gonadal steroids stimulate GHsecretion (41). Although there was no association betweenbaseline serum testosterone and 24-h mean serum GH con-centration either at baseline or in response to MK-677, anumber of subjects were testosterone deficient during thisstudy, and the effect of adequate testosterone replacement ontheir GH response to MK-677 is not known.

The drug was generally well tolerated with few reportedand no significant clinical adverse events. The increase infasting and postprandial (Sustacal) insulin concentrations,and less marked but still statistically significant increases inpostprandial glucose concentrations, are consistent with theknown effect of GH to enhance insulin resistance. However,


an insulin resistance-enhancing action of MK-677, indepen-dent of its effect on GH secretion, cannot be excluded. It isworth noting that fasting insulin concentrations increasedsignificantly during placebo as well as MK-677 treatments.This may reflect alterations to their normal exercise anddietary patterns that the subjects experienced in the ClinicalResearch Center. Therefore, although the findings suggestthat a degree of glucose intolerance and hyperinsulinemiamay accompany chronic use of this drug, it remains to bedetermined whether this is the case, and if so, whether it isof clinical significance. This will need to be carefully evalu-ated in subsequent studies. Favorable changes in body com-position because of MK-677-induced increases in GH secre-tion could conceivably counteract any unfavorable effects oninsulin resistance. GH therapy in GH-deficient subjects hasbeen reported to increase insulin resistance at 6 weeks, buthave a diminished effect at 26 weeks when significant de-creases of body fat have occurred (42).

In conclusion, daily oral administration of the GH secre-tagogue MK-677 for 4 days to selected GH-deficient men wasgenerally well tolerated and was associated with signifi-cantly increased circulating concentrations of GH, IGF-I, andIGFBP-3. Although responses to the drug were modest (rel-ative to responses seen with exogenous GH), some degree ofstimulation of the GH/IGF-I axis was observed in all sub-jects. These subjects were severely GH deficient as deter-mined by prestudy insulin tolerance test. Therefore, the smallstatistically significant increases are encouraging. It is likelythat these severely deficient adults represent less than 25%of childhood patients currently defined as GH deficient.These preliminary results suggest that oral administration ofthis compound may have a therapeutic role in the treatmentof some patients with GH deficiency. Further studies areneeded to determine its effects in other populations, such aswomen and children with idiopathic GH deficiency, as wellas its long-term safety and efficacy.

Acknowledgments

We thank Ms. Sandra Ware Jackson and the nursing staffs of theUniversity of Virginia GCRC and the Indiana University GCRC for theirexpert assistance. We also thank the General Clinical Research CenterCore Laboratory for performing the GH assays, the University of Vir-ginia Medicine Clinical Laboratory for performing the testosterone as-says, Dr. Robert Blizzard for assistance in recruitment of subjects, andRob Abbott for his statistical consultation.

References

1. Salomon F, Cuneo RC, Hesp R, Sonksen PH. 1989 The effects of treatmentwith recombinant human growth hormone on body composition and metab-olism in adults with growth hormone deficiency. N Engl J Med. 321:1797–1803.

2. Rudman D, Feller AG, Nagraj HS, et al. 1990 Effects of human growthhormone in men over 60 years old. N Engl J Med. 323:1–6.

3. de Boer H, Blok GJ, Voerman HJ, Phillips M, Schouten JA. 1994 Serum lipidlevels in growth hormone-deficient men. Metabolism. 43:199–203.

4. Rosen T, Eden S, Larson G, Wilhelmsen L, Bengtsson BA. 1993 Cardiovas-cular risk factors in adult patients with growth hormone deficiency. ActaEndocrinol (Copenh). 129:195–200.

5. Markussis V, Beshyah SA, Fisher C, Sharp P, Nicolaides AN, Johnston DG.1992 Detection of premature atherosclerosis by high-resolution ultrasonogra-phy in symptom-free hypopituitary adults [see comments]. Lancet.340:1188–1192.

6. Rosen T, Bengtsson BA. 1990 Premature mortality due to cardiovasculardisease in hypopituitarism. Lancet. 336:285–288.

7. Bowers CY, Alster DK, Frentz JM. 1992 The growth hormone-releasing ac-

tivity of a synthetic hexapeptide in normal men and short statured childrenafter oral administration. J Clin Endocrinol Metab. 74:292–298.

8. Penalva A, Carballo A, Pombo M, Casaneuva FF, Dieguez C. 1993 Effect ofgrowth hormone (GH)-releasing hormone (GHRH), atropine, pyridostigmine,or hypoglycemia on GHRP-6-induced GH secretion in man. J Clin EndocrinolMetab. 76:168–171.

9. Bowers CY, Momany FA, Reynolds GA, Hong A. 1984 On the in vitro and invivo activity of a new synthetic hexapeptide that acts on the pituitary tospecifically release growth hormone. Endocrinology. 114:1537–1545.

10. Ilson BE, Jorkasky DK, Curnow RT, Stote RM. 1989 Effect of a new synthetichexapeptide to selectively stimulate growth hormone release in healthy humansubjects. J Clin Endocrinol Metab. 69:212–214.

11. Bowers CY, Sartor AO, Reynolds GA, Badger TM. 1991 On the actionsof the growth hormone-releasing hexapeptide, GHRP. Endocrinology.128:2027–2035.

12. Dickson SL, Leng G, Robinson ICAF. 1993 Systemic administration of growthhormone-releasing peptide activates hypothalamic arcuate neurons. NeurosciLett. 53:303–306.

13. Popovic V, Damjanovic S, Micic D, Djurovic M, Dieguez C, Casanueva FF.1995 Blocked growth hormone-releasing peptide (GHRP-6)-induced GH se-cretion and absence of the synergic action of GHRP-6 plus GH-releasinghormone in patients with hypothalamopituitary disconnection: evidence thatGHRP-6 main action is exerted at the hypothalamic level. J Clin EndocrinolMetab. 80:942–947.

14. Hartman ML, Farello G, Pezzoli SS, Thorner MO. 1992 Oral administrationof growth hormone (GH)-releasing peptide (GHRP) stimulates GH secretionin normal men. J Clin Endocrinol Metab. 74:1378–1384.

15. Gertz BJ, Barrett JS, Eisenhandler R, et al. 1993 Growth hormone response inman to L-692,429, a novel nonpeptide mimic of growth hormone releasingpeptide (GHRP-6). J Clin Endocrinol Metab. 77:1393–1397.

16. Aloi JA, Gertz BJ, Hartman ML, et al. 1994 Neuroendocrine responses to anovel growth hormone secretagogue, L-692,429, in healthy older subjects [seecomments]. J Clin Endocrinol Metab. 79:943–949.

17. Chapman IM, Hartman ML, Pezzoli SS, Thorner MO. 1996 Enhancement ofpulsatile growth hormone secretion by continuous infusion of a growth hor-mone-releasing peptide mimetic, L-692,429, in older adults—a clinical researchcenter study. J Clin Endocrinol Metab. 81:2874–2880.

18. Thorner MO, Rochiccioli P, Colle M, et al. 1996 Once daily subcutaneousgrowth hormone-releasing hormone therapy accelerates growth in growthhormone-deficient children during first year of therapy. J Clin EndocrinolMetab. 81:1189–1196.

19. Chapman IM, Hartman ML, Straume M, Johnson ML, Veldhuis JD, ThornerMO. 1994 Enhanced sensitivity growth hormone (GH) chemiluminescenceassay reveals lower postglucose nadir GH concentrations in men than women.J Clin Endocrinol Metab. 78:1312–1319.

20. Veldhuis JD, Johnson ML. 1986 Cluster analysis: a simple, versatile, androbust algorithm for endocrine pulse detection. Am J Physiol. 250:E486–E493.

21. Arrigo T, Martino F, Lombardo F, et al. 1992 Diagnostic value of growthhormone-releasing hormone test in children and adolescents with idiopathicgrowth hormone deficiency. Eur J Pediatr. 151:263–265.

22. Lanes R, Carrillo E. 1994 Long-term therapy with a single daily subcutaneousdose of growth hormone releasing hormone (1–29) in prepubertal growthhormone deficient children. Venezuelan Collaborative Study Group. J PediatrEndocrinol. 7:303–308.

23. Duck SC, Schwarz HP, Costin G, et al. 1992 Subcutaneous growth hormone-releasing hormone therapy in growth hormone-deficient children: first year oftherapy. J Clin Endocrinol Metab. 75:1115–1120.

24. Smith RG, Cheng K, Schoen WR, et al. 1993 A nonpeptidyl growth hormonesecretagogue. Science. 260:1640–1643.

25. Chaung LYP, Pong SS, Dean D, Schaeffer JM, Smith RG. 1996 Character-ization of a G-protein-linked receptor for peptidyl and nonpeptidyl growthhormone secretagogue in rat and porcine hypothalamic and pituitary mem-branes. Program of the 10th International Congress on Endocrinology, SanFrancisco, CA; P1–613:288 (Abstract).

26. Pong SS, Chaung LYP, Dean DC, Nargund RP, Patchett AA, Smith RG. 1996Identification of a new G-protein-linked receptor for growth hormone secre-tagogues. Mol Endocrinol. 10:57–61.

27. Wu D, Chen C, Zhang J, Katoh K, Clarke I. 1994 Effects in vitro of new growthhormone releasing peptide (GHRP-1) on growth hormone secretion from ovinepituitary cells in primary culture. J Neuroendocrinol. 6:185–190.

28. Cheng K, Chan WW, Butler B, Barreto, Jr, A, Smith RG. 1991 Evidence fora role of protein kinase-C in His-D-Trp-Ala-Trp-D-Phe-Lys-NH2-inducedgrowth hormone release from rat primary pituitary cells. Endocrinology.129:3337–3342.

29. Adams EF, Petersen B, Lei T, Buchfelder M, Fahlbusch R. 1995 The growthhormone secretagogue, L-692,429, induces phosphatidylinositol hydrolysisand hormone secretion by human pituitary tumors. Biochem Biophys ResCommun. 208:555–561.

30. Smith RG. 1996 Modulation of pulsatile growth hormone release by an orallyactive growth hormone secretagogue. Program of the 10th International Con-gress on Endocrinology, San Francisco, CA; S10–1:26 (Abstract).

31. Howard AD, Feighner SD, Cully DF, et al. 1996 A receptor in pituitary and


hypothalamus that functions in growth hormone release [see comments].Science. 273:974–977.

32. Bowers CY, Reynolds GA, Durham D, Barrera CM, Pezzoli SS, Thorner MO.1990 Growth hormone (GH)-releasing peptide stimulates GH release in normalmen and acts synergistically with GH-releasing hormone. J Clin EndocrinolMetab. 70:975–982.

33. Dickson SL, Leng G, Dyball RE, Smith RG. 1995 Central actions of peptideand non-peptide growth hormone secretagogues in the rat. Neuroendocrinol-ogy. 61:36–43.

34. Guillaume V, Magnan E, Cataldi M, et al. 1994 Growth hormone (GH)-releasing hormone secretion is stimulated by a new GH-releasing hexapeptidein sheep. Endocrinology. 135:1073–1076.

35. Chapman IM, Bach MA, Van Cauter E, et al. 1996 Stimulation of the growthhormone (GH)-insulin-like growth factor I axis by daily oral administration ofa GH secretagogue (MK-677) in healthy elderly subjects. J Clin EndocrinolMetab. 81:4249–4257.

36. Hartman ML, Clayton PE, Johnson ML, et al. 1993 A low dose euglycemicinfusion of recombinant human insulin-like growth factor I rapidly suppressesfasting-enhanced pulsatile growth hormone secretion in humans. J Clin Invest.91:2453–2462.

37. Bermann M, Jaffe CA, Tsai W, DeMott-Friberg R, Barkan AL. 1994 Negativefeedback regulation of pulsatile growth hormone secretion by insulin-likegrowth factor I. Involvement of hypothalamic somatostatin. J Clin Invest.94:138–145.

38. Berelowitz M, Szabo M, Frohman LA, Firestone S, Chu L, Hintz RL. 1981Somatomedin-C mediates growth hormone negative feedback by effects onboth the hypothalamus and the pituitary. Science. 212:1279–1281.

39. Yamashita S, Melmed S. 1986 Insulin-like growth factor 1 action on rat anteriorpituitary cells: suppression of growth hormone secretion and messenger ri-bonucleic acid levels. Endocrinology. 118:176–182.

40. Hickey GJ, Jacks T, Schleim K, et al. 1996 Repeat administration of the GHsecretagogue MK-0677 increases and maintains elevated IGF-I levels in bea-gles. J Endocrinol. 152:183–192.

41. Giustina A, Scalvini T, Tassi C, et al. 1997 Maturation of the regulation ofgrowth hormone secretion in young males with hypogonadotropic hypogo-nadism pharmacologically exposed to progressive increments in serum tes-tosterone. J Clin Endocrinol Metab. 82:1210–1219.

42. Fowelin J, Attvall S, Lager I, Bengtsson BA. 1993 Effects of treatment withrecombinant human growth hormone on insulin sensitivity and glucose metab-olism in adults with growth hormone deficiency. Metabolism. 42:1443–1447.


Documents

Oral administration of growth hormone (GH)-releasing peptide stimulates GH secretion in normal men