help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grinspoon, S.
Right arrow Articles by Klibanski, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grinspoon, S.
Right arrow Articles by Klibanski, A.
The Journal of Clinical Endocrinology & Metabolism Vol. 88, No. 3 1142-1149
Copyright © 2003 by The Endocrine Society

Effects of Recombinant Human Insulin-Like Growth Factor (IGF)-I and Estrogen Administration on IGF-I, IGF Binding Protein (IGFBP)-2, and IGFBP-3 in Anorexia Nervosa: A Randomized-Controlled Study

Steven Grinspoon, Karen Miller, David Herzog, David Clemmons and Anne Klibanski

Neuroendocrine Unit (S.G., K.M., A.K.) and The Eating Disorders Unit (D.H.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114; and Endocrine Division (D.C.), University of North Carolina, Chapel Hill, North Carolina 25799

Address all correspondence and requests for reprints to: Steven Grinspoon, M.D., Neuroendocrine Unit, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: sgrinspoon{at}partners.org.


   Abstract
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Discussion
 References
 
Administration of recombinant human (rh) IGF-I has been shown to have positive effects on bone density in anorexia nervosa, but the effects of rhIGF-I and estrogen on IGF binding protein (IGFBP)-2 and IGFBP-3 in anorexia nervosa are not known. Sixty-five osteopenic women with anorexia nervosa were randomized to rhIGF-I (30 µg/kg sc twice daily) alone (n = 15), daily ethinyl estradiol (Ovcon 35) with rhIGF-I (n = 15), estradiol and placebo (n = 15), or placebo (n = 14) for 9 months. Subjects were 25.6 ± 0.8 yr of age, low weight (body mass index 16.6 ± 0.2 kg/m2) and osteopenic (T scores -2.06 ± 0.09 for spine and -1.76 ± 0.13 for hip). IGFBP-3 correlated with total hip bone density (r = 0.47, P = 0.0002) and was a significant predictor of hip bone density (P = 0.010) independent of IGF-I and body mass index in a multivariate regression model. During therapy, IGFBP-2 increased by 48 ± 19 ng/ml in response to rhIGF-I and decreased by -38 ± 22 ng/ml in response to placebo (P = 0.011). IGFBP-3 decreased (-895 ± 120 ng/ml) in response to rhIGF-I but showed a minimal change (-53 ± 99 ng/ml) in response to placebo (P < 0.0001). In contrast, no significant effect of estrogen was seen on IGF-I, IGFBP-2 or IGFBP-3. Among patients receiving rhIGF-I, the change in IGFBP-2 was inversely associated with the change in total hip bone density (R = -0.47, P = 0.013). In conclusion, our data suggest that chronic rhIGF-I administration increases IGF-I and IGFBP-2 and decreases IGFBP-3 in women with anorexia nervosa. IGFBP-2 and IGFBP-3 may be important determinants of bone density in this population.


   Introduction
Top
 Abstract
 Introduction
Patients and Methods
 Results
 Discussion
 References
 
ALTHOUGH SEVERE BONE loss is seen in women with anorexia nervosa (1, 2, 3), no effective treatment strategies have been developed for reduced bone density in this large population of women. Among patients with anorexia nervosa, low weight contributes to bone loss independently of degree or duration of estrogen deficiency, but the specific mechanism by which low weight results in bone loss is not understood (4). We have previously shown that serum IGF-I concentrations are severely reduced in anorexia nervosa in association with low weight and reduced markers of bone formation (5). Bone formation indices increase in association with IGF-I levels during refeeding and weight recovery in women with anorexia nervosa (6). Moreover, we have recently shown that recombinant human (rh) IGF-I at physiologic concentrations increases bone density, particularly in combination with estrogen, in women with anorexia nervosa (7). It is therefore possible that reduced serum concentrations of IGF-I resulting from acquired GH resistance in severe undernutrition (8) contribute to reduced bone density.

Changes in related IGF-I binding proteins, such as IGF binding protein (IGFBP)-2 and IGFBP-3 occur secondary to weight loss, and these changes can alter the concentration of free IGF-I that is available to tissues. The relative contribution of changes in IGF-I binding proteins to bone density and the effects of rhIGF-I on IGFBP-2 and IGFBP-3 (the two most abundant forms of IGFBPs in plasma) in anorexia nervosa have not been investigated. In this study, we assessed the relationship of IGF-I and IGFBP-2 and 3 to bone density at multiple skeletal sites in patients with anorexia nervosa. Furthermore, we determined the independent effects of short-term (3 month) administration of rhIGF-I and estrogen on IGF-I, IGFBP-2, and IGFBP-3 among patients participating in a longer-term study of rhIGF-I effects on bone density. We also determined the relationship between the short-term changes in IGFBP-2 and IGFBP-3 to longer-term changes in bone density in response to rhIGF-I. We show that rhIGF-I administration significantly increases IGF-I and IGFBP-2 and reduces IGFBP-3 over 3 months, whereas estrogen administration has little effect on IGF-I, IGFBP-2, or IGFBP-3 in women with anorexia nervosa. IGFP-3 is associated with bone density, but short-term changes in IGFBP-2 predict changes in bone density in response to rhIGF-I administration in this population.


   Patients and Methods
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Discussion
 References
 
One hundred eighty women were screened for a longitudinal study of bone loss in women with anorexia nervosa. One hundred seventeen subjects were eligible based on World Health Organization-defined osteopenia of the lumbar spine (anteroposterior T score -1.0 SD or less). Of this group, 31 subjects receiving estrogen were excluded, and 7 patients with weight not fulfilling the Diagnostic and Statistical Manual of Mental Disorders (DSM) IV criteria of anorexia nervosa (>85% ideal body weight) were also excluded. Fourteen subjects declined enrollment. Sixty-five eligible subjects with DSM-IV confirmed anorexia nervosa and amenorrhea for at least 3 months were thus recruited into the study. No patient had received estrogen or related hormones known to affect bone density or turnover within 6 months of the study. Screening thyroid function, FSH, and prolactin (PRL) levels were normal in all patients. Screening bone density in a subset of the subjects was previously reported (4). In addition, serial changes in regional body composition with spontaneous weight recovery were also reported in a subset of the patients (9), and the effects of rhIGF-I and estrogen on bone density were recently reported (7).

Baseline assessment

At the baseline visit, weight and height were measured. Bone density was measured in all patients at multiple sites, including total body, radius, total hip, femoral neck, AP, and lateral lumbar spine by dual x-ray absorptiometry (DXA), and IGF-I, IGFBP-2 and IGFBP-3 were assessed. Whole body lean and fat mass were also assessed by DXA. Caloric intake, fat, carbohydrate, and protein intake were determined from analysis of food records and frame size was determined as previously reported (4). Serum procollagen carboxyl-terminal propeptide (PICP) and urinary N-telopeptide (NTX) excretion were determined. After the baseline assessment, subjects were simultaneously randomized to rhIGF-I [30 µg/kg·d sc twice daily (BID)] (Genentech, Inc., San Francisco, CA) or identical placebo and to estrogen (Ovcon 35, 35 µg ethinyl estradiol, and 0.4 mg of norethindrone) (Bristol Myers Squibb, Inc., Princeton, NJ) or no estrogen in a two-by-two factorial model for a period of 9 months, as previously reported (7). Placebo was prepared by the Massachusetts General Hospital pharmacy and was identical in color and consistency to active drug. The administration of rhIGF-I was blinded to the patient and the nursing staff on the General Clinical Research Center, but not to investigator or study personnel, to permit safety monitoring. Estrogen administration was not blinded to either patient or investigator. The randomization schema was prepared by the study statistician. Blood glucose was determined immediately before and for 30, 60, 90, and 120 min after the first injection of rhIGF-I or placebo at baseline and at 3 months. Subjects kept an injection log and returned unused or empty vials. In addition, patients recorded menses and estrogen pills taken in a separate log. Returned estrogen pills were also recorded.

Subsequent study visits

Subjects were assessed for safety at 1 wk by telephone and returned for safety monitoring and assessment of IGF-I levels at 2 wk and then monthly after study initiation. Weight was measured at each visit. Serum IGF-I was determined at each visit, and a dose reduction of 25% was performed for patients with an IGF-I level above the normal age-adjusted range for the assay. Dose adjustment was further made as required based on measured weight at each visit to maintain a dose of 30 µg/kg sc BID. No dose adjustment was made for placebo-treated patients. At the 12-wk visit, repeat measurements of IGF-I, IGFBP-2, IGFBP-3 and blood glucose were made. Bone density was not performed at the 3-month visit. Six patients were unable to return at 3 months, and therefore 59 patients had paired 0- and 12-wk data available. One patient in the rhIGF-I group was discontinued for local skin irritation associated with injections. Two patients in the rhIGF-I group and one patient in the placebo group dropped out after randomization but before the baseline visit. One patient in the rhIGF-I group dropped out for personal reasons unrelated to the study after the baseline visit. An additional patient in the placebo group was unable to make the 3-month visit of the study.

Data analysis

Univariate and multivariate regression analyses were performed relating IGF-I, IGFBP-2, IGFBP-3, and bone density at baseline. The effects of rhIGF-I and estrogen were determined simultaneously in a two-by-two factorial model, with an interaction term. There was no significant interaction between rhIGF-I or estrogen in any model. Therefore, all rhIGF-I treated (n = 30) were compared with all placebo-treated patients (n = 29), independent of estrogen status, and conversely all estrogen treated (n = 30) were compared with all nonestrogen-treated (n = 29), independent of rhIGF-I status, by analysis of covariance, controlling for baseline values, in a pooled factorial analysis. The effects of the four individual treatment groups [rhIGF-I and estrogen (n = 15), rhIGF-I (n = 15), estrogen (n = 15), and no active treatment (n = 14)] on IGF-I, IGFBP-2 and IGFBP-3 were also compared by t test. The 3-month changes in IGFBP-2 and IGFBP-3 were also compared with the 9-month changes in bone density among patients receiving rhIGF-I in univariate regression analyses. With 59 evaluable patients, and a 10% dropout rate, the study was powered at 99% to detect a difference of 850 ng/ml in IGFBP-3 (two-sided 5% significance level) between treatment groups in the factorial model, assuming an SD of 584 ng/ml. The power of the study to determine a significant interaction between treatments was more than 95% (two-sided 5% significance level) based on a root square mean of 584 ng/ml for IGBP-3 in the factorial model.

Safety monitoring

An independent Data Safety Monitoring Board met every 3 months to review all adverse events associated with the study, as well as safety data related to glucose, weight, and IGF-I levels. The protocol was approved by the Subcommittee on Human Studies of the Massachusetts General Hospital, and informed consent was obtained from all patients, in accordance with the Helsinki II guidelines on the conduct of human research.

Methods

Height was determined by stadiometer. Weight was determined on a calibrated scale to the nearest 0.1 kg. Bone density and body composition was performed by DXA (Hologic, Inc. 45000, Waltham, MA). The precision of this technique is less than 1.5% for the lumbar spine, 1% for total body fat mass, and 3% for total body lean mass. PRL, FSH, and TSH were determined at screening by previously reported methods (10). Glucose was determined by fingerstick (Accucheck III meter, Roche Molecular Biochemicals, Indianapolis, IN). For safety monitoring, IGF-I assays were performed by RIA after alcohol extraction with an intraassay coefficient of variation of 2.4–3.0% (Nichols Institute, San Juan Capistrano, CA) at each visit rather than frozen for subsequent analysis. Baseline and 12 wk IGF-I, IGFBP-2, and IGFBP-3 were also batched for subsequent analysis by RIA, in which paired samples from baseline and 12 wk were run in the same assay. These serum samples underwent acid ethanol extraction to remove IGFBPs before RIA of IGF-I (11). IGFBP-2 and IGFBP-3 were also measured by RIA as described previously (12). Caloric intake was determined by food records and analyzed for total caloric, protein and fat intake (Minnesota Data Nutrition Systems, Minneapolis, MN). PICP was determined by RIA with an intraassay coefficient of variation of 2.1–3.2% (DiaSorin, Inc., Stillwater, MN). Urinary excretion of NTX was determined by EIA with an intraassay coefficient of variation of 5.0–8.6% (Ostex International, Inc., Seattle, WA) and adjusted for urinary creatinine excretion.


   Results
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Discussion
 References
 
Baseline clinical characteristics are shown in Table 1Go. IGF-I and IGFBP-3 were positively associated with body mass index (BMI), whereas IGFBP-2 was inversely associated with BMI at baseline (Table 2Go). Lean body mass was only associated with IGF-I, whereas fat mass was associated with IGFBP-3. In univariate regression analyses, IGF-I and IGFBP-3 were positively associated with femoral neck and total hip bone density (Fig. 1Go and Table 3Go), whereas IGFBP-2 was not associated with bone density at any site. In contrast, neither IGF-I nor IGFBP-3 was associated with lumbar spine bone density. Frame size correlated significantly only with distal one third-radius bone density (R = 0.28, P = 0.032). In multivariate regression analyses, controlling for BMI, height, and IGF-I, IGFBP-3 was a significant independent predictor of bone density at the total hip (P = 0.01, R2 for whole model = 0.51) and femoral neck (P = 0.02, R2 for whole model = 0.51) (Table 3Go).


View this table:
[in this window]
[in a new window]
  		Table 1. Baseline clinical characteristics and bone density

 

View this table:
[in this window]
[in a new window]
  		Table 2. Univariate regression analysis for IGF-I, IGFBP-3, and IGFBP-2

 


View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Univariate regression analysis of IGFBP-3 and total hip bone density, R = 0.47, P < 0.0001 for baseline measurements, before administration of rhIGF-I and estrogen.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Univariate and multivariate regression analyses for bone density

 
After 12 wk, the increase in IGF-I for the pooled groups was 164 ± 18 compared with 7 ± 6 ng/ml with placebo (P < 0.0001) and IGFBP-2 increased by 48 ± 19 compared with -38 ± 22 ng/ml with placebo (P = 0.011). In contrast, IGFBP-3 decreased significantly more in response to rhIGF-I than placebo, (-895 ± 120 compared with -53 ± 99, P < 0.0001) (Table 4Go). No significant changes for IGF-I, IGFBP-2, or IGFBP-3 were seen in response to estrogen (Table 4Go). Neither change in IGF-I (r = 0.02, P = 0.86), IGFBP-2 (r = -0.10, P = 0.50) or IGFBP-3 (r = 0.06, P = 0.63) were related to change in weight over the 3-month period of observation among all patients or in the subset receiving rhIGF-I (data not shown).


View this table:
[in this window]
[in a new window]
  		Table 4. Effects of A) rhIGF-I vs. placebo and B) estrogen vs. no estrogen on clinical endpoints

 
The changes in four individual treatment groups are shown in Table 5Go and Fig. 2Go, A and B. IGF-I increased significantly more in the two groups receiving rhIGF-I compared with the two groups not receiving rhIGF-I, but there was no significant difference in the IGF-I response between the group receiving rhIGF-I alone vs. rhIGF-I and estrogen. The IGF-I response did not differ in response to estrogen (estrogen alone vs. no estrogen, group 3 vs. 4 in Fig. 2AGo). IGFBP-3 decreased significantly in both groups receiving rhIGF-I compared with the other groups, but there was no significant difference in the IGFBP-3 response between subjects receiving rhIGF-I alone and rhIGF-I and estrogen. The IGFBP-3 response did not differ in response to estrogen (estrogen alone vs. no estrogen, group 3 vs. 4 in Fig. 2AGo). IGFBP-2 increased significantly more among subjects receiving rhIGF-I alone, compared with subjects receiving either estrogen alone or no therapy, but there was no difference in the absolute increase in IGFBP-2 among patients receiving rhIGF-I alone compared with the response in subjects receiving rhIGF-I and estrogen. However, comparison of the percent change in IGFBP-2 between the groups demonstrated a greater percent change in those subjects receiving rhIGF-I alone compared with the response in subjects receiving rhIGF-I plus estrogen (37.3 ± 11.0 vs. 8.2 ± 9.5%, P < 0.5, respectively).


View this table:
[in this window]
[in a new window]
  		Table 5. Effects of rhIGF-I and estrogen on clinical endpoints

 


View larger version (27K):
[in this window]
[in a new window]
  		Figure 2. A, Effects of rhIGF-I and estrogen on IGF-I and IGFBP-2. *, P < 0.05 rhIGF-I and estrogen compared with rhIGF-I placebo and estrogen (group 1 vs. 3). **, P < 0.05 rhIGF-I and estrogen compared with rhIGF-I placebo and no estrogen (group 1 vs. 4). #, P < 0.05 rhIGF-I alone compared with rhIGF-I placebo and estrogen (group 2 vs. 3). ##, P < 0.05 rhIGF-I alone compared with rhIGF-I placebo and no estrogen (group 2 vs. 4). B, Effects of rhIGF-I and estrogen on IGFBP-3. *, P < 0.05 rhIGF-I and estrogen compared with rhIGF-I placebo and estrogen (group 1 vs. 3). **, P < 0.05 rhIGF-I and estrogen compared with rhIGF-I placebo and no estrogen (group 1 vs. 4). #, P < 0.05 rhIGF-I alone compared with rhIGF-I placebo and estrogen (group 2 vs. 3). ##, P < 0.05 rhIGF-I alone compared with rhIGF-I placebo and no estrogen (group 2 vs. 4).

 
Mean IGF-I levels were maintained within the normal range during the 12 wk of observation. In 3 of 30 patients randomized to rhIGF-I, IGF-I levels rose above the age-adjusted normative ranges on one or more visits in the study. Furthermore, IGF-I levels were within the normal range in all but 6% of the 113 individual monthly assessments performed on the rhIGF-I treated patients. As few dose reductions were required, the average daily dose for the rhIGF-I treated subjects remained 30 µg/kg sc BID at the 3-month visit. The average blood sugar over 2 h after the baseline injection was 80 ± 2 vs. 84 ± 3 mg/dl (rhIGF-I vs. placebo, respectively, P = 0.36) at baseline and (87 ± 3 vs. 89 ± 3 rhIGF-I vs. placebo, P = 0.51) at 3 months. Blood sugar was not measured to be less than 50 mg/dl in any rhIGF-I treated patient at either the baseline or 3-month visit. No rhIGF-I treated patient demonstrated symptoms of hypoglycemia or tachycardia.

The decrease in IGFBP-3 was inversely associated with the increase in IGF-I over 3 months (R = -0.33, P = 0.001, among all subjects), whereas the change in IGFBP-2 was inversely associated with the change in total hip bone density among those subjects receiving rhIGF-I (R = -0.47, P = 0.013) over 9 months. The changes in IGFBP-3 and IGFBP-2 did not significantly correlate with changes in PICP (r = 0.00, P = 0.97 and r = 0.19, P = 0.16 for IGFBP-3 and IGFBP-2, respectively) or urinary NTX (r = -0.10, P = 0.509 and r = 0.21, P = 0.15 for IGFBP-3 and IGFBP-3, respectively) over 3 months.

Compliance was equivalent among the study groups in terms of rhIGF-I. Over 9 months, subjects returned an average of 11.2 ± 0.6, 9.4 ± 1.0, 10.9 ± 0.7, and 9.1 ± 1.1 vials, P = 0.240, estrogen and rhIGF-I, rhIGF-I, estrogen, placebo, respectively). Subjects receiving estrogen missed on average 5.3 ± 2.0 d of treatment over the study, based on diary and returned pill packs.


   Discussion
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
References
 
IGF-I is a nutritionally regulated protein, with potent trophic effects to stimulate bone collagen formation in vivo and in vitro (13, 14, 15, 16, 17). IGF-I circulates bound to a ternary complex comprised of IGF-I, IGFBP-3, and acid labile subunit (ALS; Refs. 18, 19, 20). IGF-I concentrations are reduced in undernutrition as a result of acquired resistance to the effects of GH due to decreased GH receptors and by post-receptor mechanisms (21, 22, 23, 24). Serum concentrations of IGF-I are reduced in patients with anorexia nervosa (8), but the relationship of IGFBP-2 and IGFBP-3 to bone and the effects of rhIGF-I and estrogen on IGF-I and related binding proteins in anorexia nervosa is unknown.

We have previously demonstrated low IGF-I in association with reduced indices of bone formation in anorexia nervosa (5). Additionally, we have shown in healthy female subjects that the reduction in bone turnover markers that occurs with acute fasting can be reversed over 6 d with administration of rhIGF-I (25). Hotta et al. (6) recently showed that serum concentrations of IGF-I increase in association with increased bone formation indices in response to refeeding in women with anorexia nervosa. Moreover, we have recently shown that rhIGF-I at physiologic concentrations increases bone density, particularly in combination with estrogen, in women with anorexia nervosa (7). These data suggest that reduced IGF-I in association with undernutrition may contribute to decreased bone formation and therefore low bone density in patients with anorexia nervosa. However, the effects of sustained rhIGF-I on the predominate binding proteins in plasma, e.g. IGFBP-2 and IGFBP-3, in severe undernutrition are not known.

In this study, we assess the effects of rhIGF-I on IGF-I related binding proteins IGFBP-2 and IGFBP-3 in women with anorexia nervosa, participating in a long-term treatment study on bone density (7). We demonstrate that IGFBP-3, independent of IGF-I, is a strong predictor of reduced bone density in women with anorexia nervosa, independent of BMI and IGF-I. The strongest relationship between reduced IGFBP-3 and bone density was observed at the hip. It is unclear whether IGFBP-3 has an independent effect on bone or whether alterations in the circulating concentrations of IGFBP-3 affect bone through modulation of tissue-available IGF-I. Sugimoto et al. (26) demonstrated a positive association between IGFBP-3, but not IGFBP-2, and mid-radius bone density in healthy postmenopausal women and Boonen et al. (27) demonstrated reduced IGFBP-3 levels in postmenopausal women with femoral neck osteoporosis compared with healthy aged-matched women. IGFBP-3 and hip bone density may be reduced in severe chronic undernutrition as a result of resistance to the action of GH at multiple tissue sites.

When rhIGF-I was administered at a dose of a dose of 30 µg/kg sc BID, the serum IGF-I concentration increased, whereas the IGFBP-3 concentration decreased. In this regard, the study was adequately powered to detect a relevant change in IGBFP-3, even after accounting for the 10% dropout rate. A similar effect was reported previously with high dose administration of rhIGF-I in calorically restricted healthy subjects (28, 29). The most likely explanation for the effect of rhIGF-I in this regard is to reduce GH and ultimately IGFBP-3 through feedback suppression. However, we did not directly assess GH levels in this study, and therefore the presumed mechanism by which rhIGF-I decreases GH and IGFBP-3 cannot be confirmed. In contrast, short-term changes in IGFBP-2 were significantly associated with long-term (9-month) changes in bone density among patients receiving rhIGF-I in this study. These data are consistent with prior data suggesting that IGFBP-2 is potentially inhibitory to bone and are consistent with recent data of Amin et al. (30) demonstrating an inverse association between IGFBP-2 and bone mineral density in population studies. Short-term changes in IGF binding proteins may therefore be a useful marker for the long-term effects of rhIGF-I on bone density. The mechanism whereby changes in IGFBP-2 might affect bone density remain unknown.

Hypoglycemia and symptoms of hypoglycemia were not observed in rhIGF-I-treated patients, and rhIGF-I was well tolerated. Swelling and other potential side effects of IGF-I were not noted by any patient, although one patient was withdrawn because of local erythema around the injection site. IGF-I levels remained within the normal range in 90% of the patients for 94% of all study visits. Dose reduction was necessary in only 10% of patients, and the average administered dose of rhIGF-I remained constant throughout the protocol at 30 µg/kg sc BID. The mean IGF-I level among rhIGF-I-treated patients was within the normal range at each visit.

In contrast to the observed effects of rhIGF-I on IGF-I and IGFBP-3, estrogen alone had little effect on IGF-I or IGFBP-2 and IGFBP-3. In this regard, data from our study differs from other studies in which oral estrogen administration decreased IGF-I in healthy premenopausal and postmenopausal women (31, 32). One possible explanation in this regard is that IGF-I levels are already low related to undernutrition and therefore change less in response to estrogen in severe chronic undernutrition. Altered estrogen metabolism secondary to anorexia nervosa may also potentially explain this finding, but this was not assessed in the current study. The dose of estrogen we used was a standard one and compliance was confirmed to be adequate by menstrual diary and pill count. The mechanism by which oral estrogen administration decreases IGF-I is thought to be via direct effects of estrogen on hepatic synthesis of IGF-I, with a resultant rise in GH from feedback disinhibition, rather than via an inhibitory effect on GH. A potent effect of undernutrition to inhibit IGF-I gene expression may mitigate the known effects of estrogen on IGF-I. We did observe a reduced response of IGFBP-2 to rhIGF-I in the presence of estrogen. This suggests that estrogen may be acting to inhibit the stimulatory effect of rhIGF-I on IGFBP-2 and therefore could lead to a greater increase in free, bioavailable IGF-I when combined rhIGF-I and estrogen is administered compared with rhIGF-I alone.

Our data suggest that IGFBPs may have independent effects on bone density in a model of chronic undernutrition, as demonstrated by the strong relationship between IGFBP-3 and hip bone density in multivariate modeling in this population of women with anorexia nervosa. Administration of rhIGF-I effectively increases serum IGF-I concentrations but simultaneously lowers IGFBP-3 and increases IGFBP-2 in women with anorexia nervosa and severe undernutrition. The change in IGFBP-3 in response rhIGF-I has been shown previously to be due to suppression of pituitary GH and subsequent suppression of serum ALS concentrations (28). Because ALS is an important component of the ternary complex that carries IGF-I in serum, this results in lowering IGFBP-3. A decrease in serum IGFBP-3 could reduce the stimulatory effect of rhIGF-I on bone density. However, the subsequent increase in free IGF-I may partially alter this response. The role of combined rhIGF-I and IGFBP-3 administration on bone density and the role played by IGFBP-2 in modulating changes in bone density in response to rhIGF-I in patients with anorexia nervosa and other chronic conditions of undernutrition merit further investigation.


   Acknowledgments
 
We thank the nursing and bionutrition staffs of the General Clinical Research Center for their dedicated patient care.


   Footnotes
 
This work was funded in part by NIH Grants DK-52625 (A.K.), AG-02331 (D.C.), M01-RR-01066, The Harvard Eating Disorders Center, and The Rubinstein Foundation. RhIGF-I was supplied by Genentech, Inc. under FDA IND 38,809. None of the authors received grant support or financial assistance from Genentech, Inc.

Abbreviations: ALS, Acid labile subunit; BID, twice daily; BMI, body mass index; DSM, Diagnostic and Statistical Manual of Mental Disorders; DXA, dual x-ray absorptiometry; IGFBP, IGF binding protein; NTX, N-telopeptide; PICP, procollagen carboxyl-terminal propeptide; PRL, prolactin; rh, recombinant human.

Received September 6, 2002.

Accepted December 16, 2002.


   References
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 References
 

  1. Rigotti NA, Nussbaum SR, Herzog DB, Neer RM 1984 Osteoporosis in women with anorexia nervosa. N Engl J Med 311:1601–1606[Abstract]
  2. Rigotti NA, Neer RM, Skates SJ, Herzog DB, Nussbaum SR 1991 The clinical course of osteoporosis in anorexia nervosa. J Am Med Assoc 265:1133–1138[Abstract/Free Full Text]
  3. Biller BMK, Saxe V, Herzog DB, Rosenthal DI, Holzman S, Klibanski A 1989 Mechanisms of osteoporosis in adult and adolescent women with anorexia nervosa. J Clin Endocrinol Metab 68:548–554[Abstract/Free Full Text]
  4. Grinspoon S, Thomas E, Pitts S, Gross E, Mickley D, Miller K, Herzog D, Klibanski A 2000 Prevalence and predictive factors for regional osteopenia in women with anorexa nervosa. Ann Int Med 133:790–794[Abstract/Free Full Text]
  5. Grinspoon S, Baum H, Lee K, Anderson E, Herzog D, Klibanski A 1996 Effects of short-term recombinant human Insulin-Like Growth Factor-I administration on bone turnover in osteopenic women with anorexia nervosa. J Clin Endocrinol Metab 81:3864–3870[Abstract/Free Full Text]
  6. Hotta M, Fukuda I, Sato K, Hizuka N, Shibasaki T, Takano K 2000 The relationship between bone turnover and body weight in IGF-I and IGFBP levels in anorexia nervosa patients. J Clin Endocrinol Metab 85:200–206[Abstract/Free Full Text]
  7. Grinspoon S, Thomas L, Miller K, Herzog D, Klibanski A 2002 Effects of recombinant human IGF-I and oral contraceptive administration on bone density in anorexia nervosa. J Clin Endocrinol Metab 87:2883–2891[Abstract/Free Full Text]
  8. Counts DR, Gwirtsman H, Carlsson LM, Lesem M, Cutler Jr GB 1992 The effect of anorexia nervosa and refeeding on growth hormone-binding protein, the insulin-like growth factors (IGFs), and the IGF-binding proteins. J Clin Endocrinol Metab 75:762–767[Abstract]
  9. Grinspoon S, Thomas L, Miller K, Pitts S, Herzog D, Klibanski A 2001 Changes in regional fat redistribution and the effects of estrogen during spontaneous weight gain in women with anorexia nervosa. Am J Clin Nutr 73:865–869[Abstract/Free Full Text]
  10. 1992 Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Normal reference laboratory values. N Engl J Med 327:718–724
  11. Davenport ML, Svoboda ME, Koerber KL, Van Wyk JJ, Clemmons DR, Underwood LE 1988 Serum concentrations of insulin-like growth factor II are not changed by short-term fasting and refeeding. J Clin Endocrinol Metab 67:1231–1236[Abstract/Free Full Text]
  12. Smith WJ, Nam TJ, Underwood LE, Busby WH, Celnicker A, Clemmons DR 1993 Use of insulin-like growth factor-binding protein-2 (IGFBP-2), IGFBP-3, and IGF-I for assessing growth hormone status in short children. J Clin Endocrinol Metab 77:1294–1299[Abstract]
  13. Canalis E, Centrella M, Burch W, McCarthy TL 1989 Insulin-like growth factor I mediates selective anabolic effects of parathyroid hormone in bone cultures. J Clin Invest 83:60–65
  14. McCarthy TL, Centrella M, Canalis E 1989 Regulatory effects of insulin-like growth factors I and II on bone collagen synthesis in rat calvarial cultures. Endocrinology 124:301–309[Abstract/Free Full Text]
  15. Schoenle E, Zapf J, Humbel RE, Froesch ER 1982 Insulin-like growth factor I stimulates growth in hypophysectomized rats. Nature (Lond) 296:252–253[CrossRef][Medline]
  16. Hock JM, Centrella M, Canalis E 1988 Insulin-like growth factor I has independent effects on bone matrix formation and cell replication. Endocrinology 122:254–260[Abstract/Free Full Text]
  17. Isaksson OGP, Lindahl A, Nilsson A, Isgaard J 1987 Mechanisms of the stimulatory effect of growth hormone on longitudinal bone growth. Endocr Rev 8:426–438[Abstract/Free Full Text]
  18. Baxter RC, Martin JL 1986 Radioimmunoassay of growth hormone-dependent insulinlike growth factor binding protein in human plasma. J Clin Invest 78:1504–1512
  19. Baxter RC, Martin JL 1989 The structure of the Mr 140, 000 growth hormone dependent insulin-like growth factor binding protein complex: determination by reconstitution and affinity labeling. Proc Nat Acad Sci USA 86:6898–6902[Abstract/Free Full Text]
  20. Baxter RC 1991 Insulin-like growth factor (IGF) binding proteins: role of serum IGFBPs in regulating IGF availability. Acta Paediatr Scand (Suppl) 372:107–114
  21. Clemmons DR, Klibanski A, Underwood LE, McArthur JW, Ridgway EC, Beitins IZ, Van Wyk JJ 1981 Reduction of plasma immunoreactive Somatomedin C during fasting in humans. J Clin Endocrinol Metab 53:1247–1250[Abstract/Free Full Text]
  22. Jackson Smith W, Underwood LE, Clemmons DR 1995 Effects of caloric or protein restriction on insulin-like growth factor-I (IGF-I) and IGF-binding proteins in children and adults. J Clin Endocrinol Metab 80:443–449[Abstract]
  23. Thissen J, Ketelslegers J, Underwood L 1994 Nutritional regulation of the insulin-like growth factors. Endocr Rev 15:80–101[Abstract/Free Full Text]
  24. Zhang J, Chrysis D, Underwood L 1998 Reduction of hepatic insulin-like growth factor I (IGF-I) messenger ribonucleic acid (mRNA) during fasting is associated with diminished splicing of IGF-I pre-MRNA and decreased stability of cytoplasmic IGF-I mRNA. J Clin Endocrinol Metab 139:4523–4530
  25. Grinspoon SK, Baum HB, Peterson S, Klibanski A 1995 Effects of rhIGF-I administration on bone turnover during short-term fasting. J Clin Invest 96:900–906
  26. Sugimoto T, Nishiyama K, Kuribayashi F, Chihara K 1997 Serum levels of insulin-like growth factor (IGF-I), IGF-binding protein (IGFBP-2), and IGFBP-3 in osteoporotic patients with and without spinal fractuures. J Bone Miner Res 12:1272–1279[CrossRef][Medline]
  27. Boonen S, Mohan S, Dequeker J, Aerssens J, Vanderschueren D, Verbeke G, Broos P, Bouillon R, Baylink DJ 1999 Down-regulation of the serum stimulatory components of the insulin-like growth factor (IGF) system (IGF-I, IGF-II, IGF binding protein 3, and IGFBP-5) in age-related (type II) femoral osteoporosis. J Bone Miner Res 14:2150–2158[CrossRef][Medline]
  28. Kupfer SR, Underwood LE, Baxter RC, Clemmons DR 1993 Enhancement of the anabolic effects of growth hormone and insulin-like growth factor I by use of both agents simultaneously. J Clin Invest 91:391–396
  29. Hartman ML, Clayton PE, Johnson ML, Celniker A, Perlman AJ, Alberti KG, Thorner MO 1993 A low dose euglycemic infusion of recombinant human insulin-like growth factor I rapidly suppresses fasting-enhanced pulsatile growth hormone secretion in humans. J Clin Invest 91:2453–2462
  30. Amin S, Atkinson EJ, Melton LJ, Khosla S, Oberg AL, Riggs BL, A potentially deleterious role of IGFBP-2 on mineral bone density in aging men and women. Program of the American Society of Bone and Mineral Research 24th Annual Meeting, San Antonio, TX, 2002, p S204 (Abstract F307)
  31. Bellantoni MF, Vittone J, Campfield AT, Bass KM, Harman M, Blackman MR 1996 Effects of oral versus transdermal estrogen on the growth hormone/insulin-like growth factor I axis in younger and older postmenopausal women: a clinical research center study. J Clin Endocrinol Metab 81:2848–2853[Abstract/Free Full Text]
  32. Balogh A, Kauf E, Vollanth R, Graser G, Klinger G, Oettel M 2000 Effects of two oral contraceptives on plasma levels of insulin-like growth factor I (IGF-I) and growth hormone (hGH). Contraception 62:259–269[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Endocr. Rev.Home page
A. Giustina, G. Mazziotti, and E. Canalis
Growth Hormone, Insulin-Like Growth Factors, and the Skeleton
Endocr. Rev., August 1, 2008; 29(5): 535 - 559.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
V. E. DeMambro, D. R. Clemmons, L. G. Horton, M. L. Bouxsein, T. L. Wood, W. G. Beamer, E. Canalis, and C. J. Rosen
Gender-Specific Changes in Bone Turnover and Skeletal Architecture in Igfbp-2-Null Mice
Endocrinology, May 1, 2008; 149(5): 2051 - 2061.
[Abstract] [Full Text] [PDF]

Home page
 FocusHome page
J. Yager, M. J. Devlin, K. A. Halmi, D. B. Herzog, J. E. Mitchell, P. S. Powers, and K. J. Zerbe
Guideline Watch: Practice Guideline for the Treatment of Patients With Eating Disorders, 2nd Edition
Focus, October 1, 2005; 3(4): 546 - 551.
[Full Text] [PDF]

Home page
 J ANIM SCIHome page
K. E. Govoni, T. A. Hoagland, and S. A. Zinn
The ontogeny of the somatotropic axis in Hereford calves from birth to one year of age and its response to administration of exogenous bovine somatotropin
J Anim Sci, June 1, 2004; 82(6): 1646 - 1655.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grinspoon, S.
Right arrow Articles by Klibanski, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grinspoon, S.
Right arrow Articles by Klibanski, A.

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