| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Clinical Research Centers |
Medical Service, Geriatric Research Education Clinical Center, and Psychiatry Service, Palo Alto Veterans Administration Health Care System, and Departments of Medicine and Psychiatry, Stanford University, Palo Alto, California 94304
Address all correspondence and requests for reprints to: Anne L. Friedlander, Ph.D., Geriatric Research Education Clinical Center Building MB2, 182B, Palo Alto Veterans Administration Health Care System, 3801 Miranda Avenue, Palo Alto, California 94304-1207. E-mail: friedlan{at}leland.stanford.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Short-term treatment with GH or IGF-I has been moderately successful at
reversing some age-associated changes in body composition. Data suggest
that short-term GH treatment can improve nitrogen balance, increase
lean mass, decrease fat mass, and increase markers of bone formation
and resorption in older adults (7, 12, 13, 14, 15). However, GH
treatment has also been shown to be associated with side effects, such
as fluid retention and carpal tunnel syndrome, that may preclude its
use as a long-term therapy (15, 16). In addition, GH
treatment may not prove to be efficacious as a long-term therapy,
because many of the changes initially observed in response to GH
disappear after several weeks, suggesting a developing resistance to
the hormone (14, 15). High-dose IGF-I treatment (
60
µg/kg bi-injection daily) has also been shown to be effective
at promoting body composition changes in the elderly, but it too has
been associated with intolerable side effects (7, 15). In
contrast, short-term low-dose IGF-I therapy (15–30 µg/kg b.i.d.) may
yield similar, albeit smaller, anabolic responses with fewer
side-effects (15, 17). The small changes observed in the
short-term studies could result in significant improvements in body
composition if allowed to accumulate over an extended period of time.
Because low-dose IGF-I has few side effects, does produce changes in
body composition, and may be effective for more extended periods of
time than GH, it has potential as long-term hormone-replacement therapy
for the somatopause. Therefore, the purpose of this study was to test
the hypothesis that 1 yr of low-dose IGF-I treatment that aimed to
increase serum IGF-I into the normal range for young women would
reverse some of the changes in body composition, bone density, blood
chemistry, and psychology often associated with aging and menopause.
Specifically, we expected IGF-I treatment to increase bone density,
decrease fat mass, increase lean mass, and improve mood and memory in
postmenopausal women.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twenty-four healthy, postmenopausal women over 60 yr of age were recruited from the surrounding community by flyer, newspaper advertisement, or mailing. All were nonsmoking, nonobese, and free of chronic or systemic disease, including diabetes, coronary heart disease, and uncontrolled hypertension or hyperlipidemia. Women who were both on and off estrogen-replacement therapy (ERT) were recruited for the study. All the women who were taking ERT were required to have been on a stable dose of ERT for a minimum of 1 yr before commencement of the study. Ten of the 16 subjects who were taking estrogen/progesterone hormone-replacement therapy completed the study protocol, but the ratios of ERT/non-ERT subjects in the placebo and IGF-I groups did not differ.
The protocol was approved by the Administrative Panel for the Protection of Human Subjects in Research at Stanford University; and each volunteer gave written, informed consent. Subjects underwent an initial screening visit, during which time they filled out a medical history questionnaire and had a resting ECG, physical examination, fasting laboratory blood panel, urinalysis, and a 2-h postprandial glucose analysis. Subjects were also required to have a normal mammogram and pap smear before study enrollment.
General experimental design and drug intervention
Participants were tested at baseline, at 6 months, and at 1 yr. During the testing periods, subjects consumed a standardized diet (described below) for approximately 1 week before and during the testing periods. The testing (which is described in detail below) required 2 days, during which time the subjects remained in the Aging Studies Unit (ASU). Subjects were instructed by the nursing staff on how to deliver the appropriate dose of the IGF-I/placebo by sc injection, and they began the injections immediately after the completion of the baseline testing. Subjects remained in the ASU through the first three injections (approximately 24 h), to be observed in case of adverse events. During that time, blood glucose was monitored by a One Touch II glucose system (Lifespan, Milpitas, CA), and orange juice was available to counteract possible hypoglycemia from the IGF-I.
The recombinant human (rh)IGF-I and placebo were generously supplied by Genentech, Inc. (South San Francisco, CA) and administered under investigational drug 36944. The dose of rhIGF-I was 15 µg/kg twice daily, and the study was double-blinded. The dose of IGF-I was selected because it has been shown previously to be effective in, and well-tolerated by, elderly women (12, 15). Subjects returned to the ASU, 1 week after starting the protocol and every 6 weeks between measurement periods, to pick up supplies and have a brief physical examination. Subjects were given forms to check off the twice-daily injections and to record any adverse events that occurred throughout the study intervention. All the women had follow-up mammograms upon completion of the study.
Diet
Subjects were fed a standardized diet aimed at maintaining body weight and controlling nutrient intake during the testing periods. Basal energy expenditure was estimated using the Harris-Benedict equation, and caloric need was calculated by multiplying the basal energy expenditure by an activity factor of 1.5–1.6 (18). Subjects were weighed every other day during the stabilization period, and the diet was adjusted as needed. Protein intake was set at 1.0–1.2 g/kg BW, fat at 30% of the total calories, with carbohydrate comprising the rest of the diet. Subjects ate the same diet every day for 7–9 days during each of the three testing periods, and foods were generally consistent between subjects. The diet provided 100% of the recommended daily allowances for nutrients. A daily calcium-and-mineral supplement (1200 mg Ca2+, 40 mg Mg, 1.5 mg Zn, 0.2 mg Cu, 1.5 mg Mn) was provided by Shaklee Corporation (San Francisco, CA) for consumption by the subjects throughout the study intervention.
Biochemical analysis
At baseline, 6 months, and 12 months, fasting blood samples were collected and centrifuged, and the serum was frozen at -80 C for future analyses. Serum IGF-I concentrations were determined by RIA after acid chromatography was used to separate the IGF peptides from the IGFBPs (19). IGFBP3 and IGFBP1 were measured using our previously described RIA (20). Insulin and leptin concentration was determined by a single- and two-site RIA, respectively (21). Routine serum chemistries were performed by the Veterans Administration clinical laboratory, at the time of collection, to determine blood lipids [total cholesterol, low-density lipoprotein (LDL), high-density lipoprotein (HDL), triglycerides] and blood glucose.
Bone resorption and formation were measured using urinary excretion of the type I collagen N-teleopeptide (NTx, Ostex International, Inc., Seattle, WA) and serum concentrations of intact osteocalcin (Diagnostic Systems Laboratories, Inc., Webster, TX), respectively, as we have previously reported (12). The urine samples were taken from three consecutive 24-h urine collections obtained while on the standardized diet at baseline, 6 months, and 12 months.
Body composition and strength measurements
Body weight was measured using a digital scale, with subjects wearing only a light hospital gown. BMD of the lumbar spine, total hip, forearm (midradius), and whole body, along with total lean mass and percent body fat, were assessed using DXA (QDR2000, software version 6.3; Hologic, Inc., Waltham, MA). One repetition maximum (isotonic muscle strength of the dominant hand, arms, and legs) was measured on Universal, Inc. (Columbus, OH) exercise equipment by doing a grip squeeze, arm press, and leg press, respectively (22).
Meal tolerance test
Blood samples were collected at regular intervals for 3 h after a standardized breakfast containing 100 g carbohydrate. The samples were analyzed for glucose and insulin concentration (as described above) to determine the physiologic response to the meal.
Psychological measures
Subjects performed a series of psychological tests and completed several self-rating measures designed to assess memory, sleep attributes, and level of depression and anxiety. All testing was performed by the Mental Illness Research Education and Clinical Center at the Palo Alto Veterans Administration Health Care System. Memory was tested using a variety of tests, including name-face and word-list recall (23). Three different versions of these 2 memory tests were used, and the order of these tests was counterbalanced across subjects. Depression was measured using the Geriatric Depression Scale (24), which is widely used to assess depression in the elderly. The state version of the State-Trait Anxiety Inventory (25) was used to assess the subject’s perception of anxiety at the time of testing. Subjects’ ability to sleep was assessed by 20 questions designed to illuminate the length and soundness of their sleep.
Statistical analyses
Data were analyzed using the Statview II (Abacus Concepts,
Berkeley, CA) statistics software package for MacIntosh computers.
ANOVAs with repeat measures were used to analyze significance between
groups over time (treatment x time). When appropriate, post
hoc tests of significance were made using Fisher
least-squares-difference test. Between-group comparisons at baseline
were made using t tests. The level of significance for all
tests was set at 
< 0.05. Area under the curve for insulin
data were determined using the trapezoidal method.
Premature termination
To ensure the safety of the subjects, the data were unblinded
and were checked throughout the course of the study by a scientist not
directly associated with the project. When it was determined by the
outside investigator that the IGF-I was having no measurable effect, we
felt an ethical responsibility to discontinue the protocol. Asking the
participants to submit to extensive testing, while continuing to inject
themselves twice daily with IGF-I/placebo for a year to augment the
study sample size, did not seem appropriate. Thus, because the
randomization scheme was developed for a full cohort of 40 subjects but
was never completed, the data are presented with an uneven number of
subjects in each of the groups (Table 1
).
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Twenty-four women were recruited for participation
in the study and had baseline measurements, but only 16 completed the
1-yr intervention program. Four subjects withdrew for personal or
health reasons unrelated to the protocol. Two participants withdrew
because of adverse events possibly related to the protocol (Table 2), and 3 participants were excluded from
the protocol for noncompliance or because the protocol was discontinued
early (see Statistical analyses above). Adverse events
associated with the protocol were minor and, except for those mentioned
above, required no adjustment in the treatment protocol. No significant
baseline differences between groups were observed.
|
). Circulating levels of
the IGFBPs (IGFBP1 and IGFBP3) were unaffected by the IGF-I treatment
and did not change in either group throughout the intervention (Fig. 1, B–C).
|
). Therefore, lean mass, as calculated
as the difference between total body weight and fat mass, also did not
change (Table 3).
|
, A–C). Both osteocalcin and NTx were significantly lower in the ERT
subjects than in the non-ERT subjects at each of the testing time
points (Fig. 3), but the differences in
bone turnover markers were not associated with differences in BMD
between the ERT and non-ERT groups (data not shown).
|
|
Fasting blood glucose, LDL, HDL, total cholesterol, and triglyceride
concentrations did not differ at baseline, nor were they altered by
IGF-I treatment (Table 4). Fasting blood
glucose was lower in ERT vs. non-ERT subjects at baseline
(86.7 ± 2.4, 96.0 ± 2.6, respectively, P <
0.05) and remained lower at each of the testing periods. The response
of blood glucose to a meal containing 100 g simple carbohydrates
is illustrated in Fig. 4. IGF-I treatment
did not impact glucose disposal, because the groups did not differ
significantly at any of the measured time points. However, in response
to the same meal, the area under the insulin curve, which did not
differ significantly between groups at baseline or 6 months, was lower
in the IGF-I vs. placebo subjects after 12 months of
treatment (Fig. 5). Leptin values also
were not affected by the study intervention (Table 4).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
300
vs. 250 ng/mL). On the other hand, short-term IGF-I
treatment has been shown previously to cause significant metabolic
changes in elderly subjects (7, 12, 15, 17), suggesting
that chronic IGF-I may lead to tachyphylaxis. Metabolic parameters and body composition
We did observe significant changes in glucose metabolism in the IGF-I-treated subjects. IGF-I may cause hypoglycemia by binding to the insulin receptor and increasing glucose uptake (26). IGF-I may also stimulate glucose uptake by contributing to both the translocation and synthesis of GLUT4 in muscle cells (27). In addition, IGF-I administration leads to increased insulin sensitivity under conditions of rest, exercise, and euglycemic, hyperinsulemic clamps (3, 6, 28). After 12 months of treatment, similar glucose concentrations, in response to a meal tolerance test, were observed despite significantly lower insulin values in the IGF-I treated women, compared with the placebo-treated women. These data suggest a more prolonged and sustained impact of IGF-I on glucose metabolism than is evident with the other metabolic variables. This data are consistent with other studies showing that IGF-I might counter insulin resistance.
Maintenance of lean mass improves functional capacity and quality of life in the elderly. Short-term IGF-I treatment increases protein synthesis in humans (7, 29, 30) and in animal models (31, 32). In our previous study (15), low-dose IGF-I treatment (15 µg/kg bi-injection daily) resulted in small positive changes in nitrogen balance for 4 weeks, without the reversal over time observed with high-dose treatment. In the current investigation, we hypothesized that this low-dose IGF-I would cause small changes in nitrogen balance that would continue to accumulate over the year to produce significant gains in lean mass. However, the data did not support the hypothesis and suggest that tachyphylaxis develops with low-dose IGF-I as well. Similarly, the decreases in fat mass that were observed in response to IGF-I treatment in the earlier study were not reproduced in the current long-term investigation.
BMD
We previously demonstrated that 28 days of low-dose IGF-I treatment increased blood markers for bone formation, with no concomitant increase in resorption markers, in elderly women (12). These data confirm findings by Ebeling et al. (33), who obtained similar results after 6 days of low-dose IGF-I treatment. There were no differences in bone turnover markers between the IGF-I treated and placebo-treated subjects throughout the current study. Because we selected only one marker for bone formation, it is possible that we missed an osteogenic response that would have been identified by an alternative marker. However, based on the responsiveness of osteocalcin in our earlier study (12), we hypothesized that that marker would be the best indicator of changes associated with low-dose, long-term treatment. We also did not observe any improvements in BMD at any of the measured sites. The subjects who were taking ERT throughout the study had significantly lower levels of both turnover markers, and the difference was maintained throughout the study. However, despite the lower turnover, the ERT subjects did not have measurably higher BMD. There is abundant literature documenting the protective effect of estrogen on postmenopausal bone loss; however, recent data from the PEPI study suggests that bone turnover markers are not a robust predictor of changes in BMD in either ERT or non-ERT subjects (34). In addition, those individuals who had lower BMD at the time of menopause were more likely to be placed on estrogen therapy than those with higher BMD, making cross-sectional comparisons of BMD between groups in this study difficult.
Efficacy of IGF-I
The reason for the resistance to IGF-I action during long-term therapy is unclear, but it may be attributed to feedback inhibition on the GH/IGF-I axis. IGF-I administration leads to a decline in circulating levels of GH in young men (28), adolescents with insulin-dependent diabetes (35), and patients with Laron syndrome (36). The pattern of decreased efficacy over time is not unique to IGF-I and has been noted in patients receiving GH as well. GH treatment was discounted by some for long-term treatment because of the high incidence of side-effects but also because changes observed in the short-term (such as nitrogen balance or lipid parameters) begin to decline after several weeks of treatment (14, 15, 37). Snyder et al. (14) showed the decrease in GH efficacy over time could be attributed to a developing resistance to the chronically elevated IGF-I levels in response to GH treatment. The authors suggested that either down-regulation of the IGF-I receptor or a postreceptor defect was the cause of the developing resistance. In the current investigation, we chose not to measure GH. A single-point measurement would not have yielded sufficient information, and drawing blood samples every 5 min for 24 h was too involved for the subjects, given the already extensive nature of the testing protocol. In this population, we expected the ambient GH levels to be low, and we anticipated that the values would decline with IGF-I treatment (38). It is possible that the predominantly negative findings in this investigation could be attributed to a fall in GH secretion associated with the intervention.
The importance of circulating IGF-I has been questioned by recent IGF-I knock-out experiments. Yakar et al. (39) attempted to determine the relative importance of endocrine vs. paracrine IGF-I by using tissue-specific gene deletion to create mice deficient in the gene for hepatic IGF-I production. The mice grew normally, despite an 80% reduction in circulating total IGF-I, suggesting that endocrine IGF-I is not essential for growth. However, compensatory increases in GH in the deletion mice may have contributed to the normal growth rate. These experiments led Le Roith and Butler (40) to reevaluate the somatomedin hypothesis that suggested that circulating IGF-I was the mediator of growth related GH actions and to suggest instead that paracrine IGF-I mediates GH action. However, that theory does not eliminate circulating IGF-I as having a possible physiologic role in older adults where the short-term benefits on metabolism and body composition parameters have been documented. In chronic IGF-I hormone-replacement therapy, the presumed decrease in GH secretion would lead to decreased tissue production of IGF-I. This lack of paracrine IGF-I may underlie the lack of long-term anabolic activity.
Finally, it should be emphasized that the small number of subjects in this study is problematic. The study was discontinued prematurely when an unblinded monitor reviewed the data and found that IGF-I was having no measurable effect on any of the outcome variables. Rather than continuing the investigation for the sole purpose of increasing the cohort size, we felt an ethical responsibility to discontinue the intervention. As a result, the group sizes are small, making it difficult to draw definitive conclusions.
This paper presents data from the first study investigating the efficacy of long-term, low-dose IGF-I treatment to counteract some of the physical and psychological changes associated with aging. Despite successfully elevating IGF-I levels throughout the intervention, no significant changes were observed in any of the major outcome variables. In contrast, previous studies, using identical treatment doses, demonstrated significant changes after short-term IGF-I treatment. Therefore, we conclude that the effects of IGF-I are relatively transient and that IGF-I monotherapy would not be useful as a continuous, long-term treatment to improve body composition in postmenopausal women.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 12, 2000.
Revised December 7, 2000.
Accepted December 7, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Rubin, C. L. Ackert-Bicknell, L. Zhu, X. Fan, T. C. Murphy, M. S. Nanes, R. Marcus, L. Holloway, W. G. Beamer, and C. J. Rosen IGF-I Regulates Osteoprotegerin (OPG) and Receptor Activator of Nuclear Factor-{kappa}B Ligand in Vitro and OPG in Vivo J. Clin. Endocrinol. Metab., September 1, 2002; 87(9): 4273 - 4279. [Abstract] [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 |
, Significantly different from ERT.
