| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Endocrinology (C.P., M.E.R., P.J.T.), Christie Hospital, Manchester M20 4BX, United Kingdom; Department of Endocrinology, St. Bartholomews Hospital (W.M.D., G.M.B.), West Smithfield, London EC1A 7BE, United Kingdom; and Department of Endocrinology, Hammersmith Hospital (K.M.), London W12 0HS, United Kingdom
Address all correspondence and requests for reprints to: Dr. P. J. Trainer, Department of Endocrinology, Christie Hospital, Manchester M20 4BX, United Kingdom. E-mail: . peter.trainer{at}man.ac.uk
Abstract
Standard medical therapy for patients with acromegaly includes somatostatin analogs. Owing to the widespread expression of somatostatin receptors, these may be associated with unwanted effects, such as altered glucose tolerance and impaired gut hormone release. Pegvisomant is a novel pegylated GH analog that competes with wild-type GH for GH-receptor binding sites but contains a position 120, amino acid substitution that prevents functional GH receptor dimerization, a known prerequisite for GH signal transduction and generation of IGF-I. We have studied the short-term effects of these two therapies (pegvisomant 20 mg/d for 7 d and octreotide 50 µg thrice daily for 7 d) on glucose tolerance and stimulated gut hormone release in six healthy male volunteers in an open-label, random-order, cross-over study. Subjects were assessed at baseline (oral glucose tolerance test and standard mixed meal) and on d 6 and 7 of each therapy with a minimum washout of 2 wk between treatments. Area under the curve and peak responses were analyzed using one-way repeated-measures ANOVA (on ranks where appropriate). Pegvisomant had no effect on glucose tolerance or stimulated gut hormone response during an oral glucose tolerance test and a standard meal. In contrast, octreotide significantly increased fasting plasma glucose, lowered fasting plasma insulin, and led to deterioration in glucose tolerance; three subjects developed impaired glucose tolerance and one diabetes mellitus by World Health Organization criteria. Octreotide significantly impaired stimulated release of cholecystokinin, gastrin, insulin, and pancreatic polypeptide. In conclusion, pegvisomant, unlike octreotide, is not associated with deterioration in glucose tolerance and impairment of stimulated gut hormone release in normal males.
STANDARD TREATMENT ALGORITHMS for acromegaly include pituitary surgery and, if insufficient, pituitary irradiation and/or adjuvant medical therapy with analogs of somatostatin (SMS). Native SMS exists in two forms, SMS-14 and an amino-terminal extended SMS-28, and both peptides are derived from a common preprosomatostatin gene, located on chromosome 3q28 (1, 2). SMS mediates its various actions through cell surface receptors that are expressed in the central nervous system, leptomeninges, anterior pituitary, mucosa of the gastrointestinal tract (GIT), endocrine, and exocrine pancreas (3). Five human subtypes of the SMS receptor have been cloned, all G protein receptors that possess seven trans-membrane-spanning domains (4, 5, 6, 7, 8, 9, 10). Octreotide (Sandostatin, Novartis, Basel, Switzerland) was the first SMS analog introduced into clinical practice and, in patients with acromegaly, maximally suppresses serum GH and IGF-I at doses up to 300600 µg/d (11, 12, 13). Long-term studies report serum IGF-I normalization in 5060% of patients (13, 14, 15, 16, 17). The ability of octreotide to normalize serum IGF-I is dependent on preoctreotide serum GH and the number of functional SMS receptors expressed by somatotroph tumors (15, 18). In the largest series analyzing the efficacy of octreotide over 6 months, GH levels fell below 2.5 µg/liter (5 mU/liter) in 2245% of patients, with normalization of serum IGF-I in 45% (15, 17).
The action of SMS analogs at sites other than the pituitary gland, particularly the GIT, produces a number of unwanted effects in patients with acromegaly (19). SMS has a complex role in neuroendocrine regulation of the GIT in which it inhibits the release of many gut regulatory peptides; blocks exocrine function of the stomach, pancreas, and bile; decreases gastric motility; and lengthens GIT transit time (20). SMS is known to inhibit the release of gastrin, cholecystokinin (CCK), VIP, glucagon, insulin, motilin, pancreatic polypeptide (PPP), secretin, and glucagon-like peptide-1 (GLP-1). The abolition of CCK release in response to food contributes to the high incidence of cholesterol gallstone formation in patients receiving these agents (21, 22).
GH infusion in normal subjects stimulates gluconeogenesis and lipolysis, resulting in increased blood glucose and FFA levels, which may serve as primary stimuli for enhanced insulin secretion and sustained hyperglycemia (23, 24). Similarly, acromegaly is characterized by increased glucose turnover, insulin resistance, and hyperinsulinemia (25). In acromegaly, the degree of glucose intolerance may independently influence hypertension and has also been correlated with the severity of cardiomyopathy (26, 27). Improved insulin sensitivity is reported when circulating GH is reduced by octreotide administration, but SMS analogs also inhibit insulin release (28, 29). As a result, the effects of these agents on glucose tolerance in patients with acromegaly may be multifactorial (29, 30).
Despite employment of pituitary surgery, radiotherapy, dopamine agonists, and SMS analogs, many patients with acromegaly (approximately 3035%) do not achieve the recommended biochemical criteria for remission (31). Knowledge of the interaction between GH and its receptor has led to the design of a pegylated GH analog (pegvisomant) that acts as a GH receptor antagonist by preventing GH-mediated GH receptor dimerization and signal transduction. By inhibiting GH action at the cellular level, rather than lowering circulating GH, pegvisomant normalizes serum IGF-I in 97% of patients with active acromegaly and is effective in patients who are resistant to SMS analogs (31A, 32, 33, 34). Pegvisomant contains a G120R mutation and is highly specific for the GH receptor, such that unwanted effects on gut peptide release and glucose tolerance are unlikely to be observed during pegvisomant therapy for acromegaly, but before this study in normal volunteers, this has not been established.
Here we describe the effects of sc octreotide and pegvisomant on gut peptide secretion and glucose tolerance in male volunteers, during an oral glucose tolerance test (OGTT) and standard mixed meal.
Materials and Methods
Subjects
Six healthy male volunteers aged 2163 yr (mean age 43 yr) were studied on three separate occasions in an open-label, random-order, cross-over study. All subjects provided informed consent to the study, which was approved by the South Manchester Local Research Ethics Committee.
Protocol
Each subject performed a baseline assessment of gut hormone response to a standard meal and 75-g OGTT. Subjects were then randomized to receive octreotide (50 µg sc three times daily) or pegvisomant (80 mg loading dose followed by 20 mg/d, sc) and instructed on self-administration. Assessment of the gut hormone response to a standard meal and OGTT was undertaken on d 6 and 7 of therapy, respectively. A minimum washout period of 2 wk elapsed before subjects entered the second treatment phase, receiving pegvisomant if octreotide was used initially and vice versa. A final assessment of the gut hormone response to a standard meal and OGTT was undertaken on d 6 and 7 of the second treatment.
All assessments were performed, in random order, following an overnight fast. For blood sampling a 19G iv cannula was inserted into a forearm vein. Samples were collected -5, 0, +15, +30, +45, +60, +90, and +120 min after consumption of 75 g oral anhydrous glucose dissolved in 300 ml water or 400 ml Ensure Plus (Abbott Laboratories, Queenborough, Kent, UK) (protein 25 g, carbohydrate 80.8 g, fat 20.6 g). During octreotide therapy, subjects were instructed to self-administer octreotide less than 1 h before each assessment.
Assays
Venous samples were taken into fluoride oxalate for plasma glucose measurement and (10 ml) lithium-heparin tubes containing 4000 kallikrein inhibitor units aprotinin (Trasylol VLE, Serologicalas Corporation, Kankakee, IL) for gut peptide measurement. Plasma was separated by immediate centrifugation for 10 min at 3000 rpm and frozen at once. Samples were stored at -80 C for subsequent measurement of plasma glucose, insulin, PPP, glucagon, gastrin, and CCK as single-batch analyses. Glucose was determined by a glucose oxidase/linked peroxidase assay on an ADVIA 1650 analyzer (Bayer Corp. Diagnostics, Newbury, Berkshire, UK) with a typical within-run coefficient of variation (CV) of less than 3.0%. Plasma CCK was determined by RIA after concentrating samples by Sep-Pak C18 cartridges as described previously (35). The assay used Dino (1:1,500,000) (35), which is specific for the sulfated C terminus of CCK-8 and does not react significantly with sulfated gastrin. Bolton-Hunter-labeled CCK-8 (Amersham Pharmacia Biotech, Buckinghamshire, UK) was used as label (36). Plasma concentrations of the remaining gastrointestinal regulatory peptides (insulin, gastrin, PPP, and pancreatic glucagon) were measured by RIA, as previously described (intraassay CV <10% for all assays, specifically, 5.2%, 3.8%, and 7.2% for glucagon, gastrin, and PPP, respectively) (37). Serum IGF-I was measured using the Advantage method (Nichols Institute Diagnostics, Heston, Middlesex, UK) with an intraassay CV of 7.4%, 5.7%, and 4.5% at 49, 229, and 493 ng/ml, respectively.
Statistical analysis
Plasma glucose and hormone concentrations are expressed as mean ± SEM or median (range) if not normally distributed. The area under the curve (AUC) of postoral glucose and postprandial increase in glucose and gut hormones was calculated as total and incremental integrated responses using the trapezoidal method, the incremental integrated response representing the AUC above the baseline level. Peak response was also analyzed. Statistical significance of the differences was evaluated using one-way repeated-measures ANOVA and, where appropriate, one-way repeated-measures ANOVA on ranks. All pair-wise comparison used the t-Newman-Keels test. Differences with P less than 0.05 were considered significant.
Results
Compared with baseline, serum IGF-I was significantly lower with pegvisomant (mean ± SEM, 172 ± 28 to 127 ± 15 ng/ml, P = 0.03). Although octreotide administration lowered serum IGF-I, this did not reach statistical significance (172 ± 28 to 147 ± 22, P = 0.09). Serum IGF-I levels after 7 d of either pegvisomant or octreotide were not statistically different.
During octreotide administration, all six subjects reported varying degrees of diarrhea, steatorrhea, and abdominal bloating, but the severity of these symptoms did not warrant octreotide withdrawal. No adverse symptoms were reported during pegvisomant administration.
Glucose and insulin
Fasting.
The effect of pegvisomant and octreotide on fasting/preprandial insulin and glucose concentrations are shown in Table 1
. Octreotide produced a statistically significant fall in mean fasting plasma insulin and a corresponding increase in mean fasting plasma glucose (FPG). All subjects had normal FPG levels at baseline and during pegvisomant administration. Two subjects developed impaired FPG (>6.1 mmol/liter) during octreotide administration.
|
At all time points measured, mean glucose levels following 75 g of oral glucose during pegvisomant administration were not statistically different from those observed at baseline (Fig. 1
). During octreotide administration, a more marked rise in mean plasma glucose was observed, such that between 60 and 90 min following glucose ingestion, mean plasma levels were statistically greater than at baseline or with pegvisomant (Fig. 1
). Peak plasma glucose and mean AUC glucose was significantly higher when subjects received octreotide but not pegvisomant (Table 2
). According to World Health Organization criteria (38), all subjects had normal glucose responses to the OGTT at baseline and during pegvisomant administration but when receiving octreotide, three developed impaired glucose tolerance (IGT) and one diabetes mellitus (DM).
|
|
|
During a standard mixed meal, there was no statistically significant difference among peak plasma glucose, peak plasma insulin, AUC glucose, and AUC insulin at baseline or during octreotide and pegvisomant administration (Table 3
and Figs. 12![]()
).
|
Fasting.
The effects of pegvisomant and octreotide on PPP, gastrin, CCK, and pancreatic glucagon release are shown in Figs. 36,![]()
![]()
![]()
respectively. Unlike pegvisomant, octreotide produced a statistically significant fall in median preprandial/fasting levels of CCK and PPP (Table 1
).
|
|
|
|
Except for pancreatic glucagon, the baseline OGTT produced a significant increase in all postglucose hormone concentrations measured. The increase in plasma PPP, gastrin, and CCK following glucose ingestion was significantly blunted during octreotide administration but was unaffected by pegvisomant. Specifically, mean PPP and CCK concentrations remained at basal levels throughout the OGTT during octreotide therapy, whereas mean gastrin concentration fell below basal levels and was significantly below levels observed at baseline for all time points beyond 30 min (Figs. 3
and 5
). Compared with baseline, mean peak and AUC PPP, CCK, and gastrin were significantly reduced by octreotide but not pegvisomant.
Standard mixed meal
The standard meal produced a statistically significant increase in all gut regulatory peptides measured except glucagon. Unlike pegvisomant, octreotide blunted the expected rise in gastrin, PPP, and CCK during the standard meal. Median plasma CCK levels were not statistically different from those observed during the baseline standard meal test when subjects received pegvisomant. In contrast, when receiving octreotide, median plasma CCK was significantly lower than at baseline and most other time points. Median peak and AUC CCK and gastrin were significantly reduced during octreotide administration, compared with baseline (Figs. 4
and 5
). Plasma PPP levels failed to rise above basal values in response to the standard meal when subjects were receiving octreotide but were unaffected by pegvisomant (Fig. 3
). Mean peak and AUC PPP were significantly lower when subjects were receiving octreotide.
Discussion
Active acromegaly is associated with increased morbidity and mortality, with cardiovascular disease accounting for the majority of excess deaths (39, 40, 41, 42). Numerous factors such as arterial hypertension, glucose intolerance, insulin resistance, and altered lipid parameters may contribute to this increase in cardiovascular mortality. In addition to age and estimated disease duration, the degree of abnormality of glucose tolerance has been reported to influence systolic and diastolic blood pressure and correlate with the severity of acromegalic cardiomyopathy (26, 27, 43, 44).
We have demonstrated, in six male volunteers, that the peak insulin response and mean AUC insulin during a 75-g OGTT is significantly lower when subjects receive octreotide, compared with baseline or when receiving pegvisomant. There is a statistically significant deterioration in glucose tolerance with octreotide but not pegvisomant administration. All subjects had a normal glucose response to OGTT at baseline and during pegvisomant, but after 1 wk of octreotide, three had developed IGT and one overt DM by World Health Organization criteria.
SMS analogs are an effective and well-tolerated medical treatment for acromegaly (14, 15, 17). However, because of the widespread expression of SMS receptors, these agents, in addition to lowering circulating GH, have complex multiple actions on gut hormone release and glucose tolerance. In normal subjects octreotide inhibits the secretion of insulin and glucagon and delays gastrointestinal glucose absorption (45, 46). The combined effect of SMS analogs on carbohydrate metabolism in patients with acromegaly is determined by effects on GH release, which improves insulin sensitivity, and the inhibition of insulin secretion. After 23 wk of octreotide therapy, no change in the glucose response during a 75-g OGTT, glycated hemoglobin, and FPG is reported, although the mean insulin response during OGTT may be significantly impaired (47, 48). Long-term studies of SMS analog therapy in acromegaly reveal heterogeneous effects on glucose tolerance. Ho et al. (28) studied seven patients receiving sc octreotide (500 µg three times per day) for 914 months in whom the impact of octreotide on glucose tolerance was dependent on pretreatment glycemic status. In the study reported by Ho et al. (28), the morning dose of octreotide was omitted before assessment and the change in mean blood glucose during OGTT was negatively related to pretreatment mean blood glucose response. Mean glucose concentration during OGTT was reduced in those with the highest pretreatment value, and whole-body insulin sensitivity, measured using the euglycemic hyperinsulinemic clamp method, was increased by octreotide therapy.
In contrast, a large prospective study of long-term octreotide therapy involving 90 patients, in which octreotide was not omitted on the day of assessment, observed a significant deterioration in glucose tolerance (30). Of the 55 patients with normal glucose tolerance at baseline, 11 (20%) and 16 (29%), respectively, developed IGT or DM. In this study, pretreatment glucose levels were not predictive of changes in glucose tolerance during therapy (30). Thus, despite improvement in insulin sensitivity in patients with acromegaly receiving octreotide, impaired insulin release may contribute to postprandial hyperglycemia, which has been linked with an increased incidence of cardiovascular disease. Prospective epidemiological studies have demonstrated a 2-fold increase in cardiovascular disease risk in nonacromegalic subjects with IGT, compared with those with normal glucose tolerance (49, 50).
Pegvisomant is not associated with deterioration in glucose tolerance in normal volunteers, and to date this has not been observed in patients with acromegaly receiving pegvisomant. Moreover, the normalization of serum IGF-I in patients with acromegaly who receive pegvisomant has been associated with significant reductions in insulin resistance (calculated using the Homeostatic Model Assessment equation) (50A ). Insulin resistance is a fundamental abnormality in the metabolic syndrome that is associated with increased cardiovascular mortality. An ideal therapeutic agent for acromegaly will reduce insulin resistance and improve glucose tolerance because both of these factors may influence outcomes. Long-term studies will therefore need to address potential differences in mortality reduction between pegvisomant and SMS analogs.
In the present study, pegvisomant, unlike octreotide, was not associated with a reduction in stimulated release of gut regulatory peptides such as gastrin, PPP, and CCK. This action of SMS analogs accounts for GIT side effects such as steatorrhea, diarrhea, abdominal pain, and bloating in patients with acromegaly and may prompt drug withdrawal in some (19). Inhibition of meal-stimulated CCK release by SMS analogs leads to bile stasis within the gall bladder and after 12 yr of treatment 1060% of patients develop cholesterol-rich gallstones (51, 52).
Although still a matter of debate, acromegaly may be associated with an increased prevalence of colonic polyps, which may undergo malignant transformation. A number of studies have reported an increased prevalence of colorectal adenomatous polyps in acromegaly, but this is not a universal finding (40, 53, 54, 55, 56, 57, 58). An increased incidence of colonic carcinoma has also been reported, with estimates ranging from a 14-fold increase in one single institute colonoscopic screening program to a standardized incidence ratio of 1.68 in a large epidemiological study (n = 1362) (40, 56). Bile acids have long been implicated in the pathogenesis of colorectal malignancy. In humans, deoxycholic acid (DCA) is formed by anaerobic bacterial conversion (i.e. deconjugation and 7
- hydroxylation) of conjugated or amidated cholic acid in the caecum or colon (59, 60). DCA is then absorbed by passive nonionic diffusion and concentrations can be measured in serum (61). Serum DCA levels are elevated in nonacromegalic patients with colorectal neoplasia and are also increased in patients with acromegaly, compared with controls (62, 63). SMS analog therapy increases large bowel transit time allowing greater DCA formation and absorption, elevating high serum DCA levels further (63). Although circumstantial, this suggests that further study of the effects of SMS analogs on the incidence of colonic neoplasia are needed. In this study of normal male volunteers, pegvisomant, in contrast to octreotide, was not associated with altered gut peptide release. Prolongation of large bowel transit and increased serum DCA levels are not anticipated during pegvisomant therapy for acromegaly, but further study is required.
In conclusion, we have demonstrated in normal males that pegvisomant, in contrast to octreotide, is not associated with a significant deterioration in glucose tolerance consequent on impaired insulin release. Pegvisomant fails to alter the gastrin, PPP, and CCK responses to glucose or a standard mixed meal, whereas octreotide inhibits the secretion of them all. It is anticipated that, in patients with acromegaly receiving pegvisomant, homogenous improvement in insulin sensitivity with no change in insulin secretion or deterioration of glucose intolerance will be observed. Moreover, pegvisomant is unlikely to be associated with gallstone formation.
Acknowledgments
Footnotes
This study was supported by Sensus Drug Development Corp. (Austin, TX).
Abbreviations: AUC, Area under the curve; CCK, cholecystokinin; CV, coefficient of variation; DCA, deoxycholic acid; DM, diabetes mellitus; FPG, fasting plasma glucose; GIT, gastrointestinal tract; IGT, impaired glucose tolerance; OGTT, oral glucose tolerance test; PPP, pancreatic polypeptide; SMS, somatostatin.
Received September 23, 2001.
Accepted January 14, 2002.
References
This article has been cited by other articles:
![]() |
H Biering, B Saller, J Bauditz, M Pirlich, B Rudolph, A Johne, M Buchfelder, K Mann, M Droste, I Schreiber, et al. Elevated transaminases during medical treatment of acromegaly: a review of the German pegvisomant surveillance experience and a report of a patient with histologically proven chronic mild active hepatitis Eur. J. Endocrinol., February 1, 2006; 154(2): 213 - 220. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Stanley, K. Wynne, B. McGowan, and S. Bloom Hormonal Regulation of Food Intake Physiol Rev, October 1, 2005; 85(4): 1131 - 1158. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. O. L. Jorgensen, U. Feldt-Rasmussen, J. Frystyk, J.-W. Chen, L. O. Kristensen, C. Hagen, and H. Orskov Cotreatment of Acromegaly with a Somatostatin Analog and a Growth Hormone Receptor Antagonist J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5627 - 5631. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Barkan, P. Burman, D. R. Clemmons, W. M. Drake, R. F. Gagel, P. E. Harris, P. J. Trainer, A. J. van der Lely, and M. L. Vance Glucose Homeostasis and Safety in Patients with Acromegaly Converted from Long-Acting Octreotide to Pegvisomant J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5684 - 5691. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Wynne, S. Stanley, B. McGowan, and S. Bloom Appetite control J. Endocrinol., February 1, 2005; 184(2): 291 - 318. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. F. Muller, J. J. Kopchick, A. Flyvbjerg, and A. J. van der Lely Growth Hormone Receptor Antagonists J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1503 - 1511. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |