| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Royal Melbourne Hospital (A.P.G., P.S., J.S., P.R.E.), Parkville, Victoria 3050, Australia; Peter MacCallum Cancer Centre (J.R.), East Melbourne, Victoria 3002, Australia; Alfred Hospital (A.P.S.), Prahran, Victoria 3181, Australia; Westmead Hospital (K.B.), Westmead, New South Wales 2145, Australia; Royal Adelaide Hospital (C.H.), Adelaide 5000, Australia; and Royal Perth Hospital (R.H.), Perth, Western Australia 6000, on behalf of the Australasian Bone Marrow Transplant Cooperative Study Group
Address all correspondence and requests for reprints to: Professor Peter R. Ebeling, Department of Medicine, University of Melbourne, Western Hospital, Footscray 3011, Australia. E-mail: peterre{at}unimelb.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to evaluate effects of high-dose pamidronate therapy on bone loss (BMD) after alloSCT.
Design: This was a randomized, multicenter, open-label, 12-month prospective study of iv pamidronate (90 mg/month) beginning before conditioning vs. no pamidronate. All 116 patients also received calcitriol (0.25 µg/d) and calcium (1000 mg/d), which were continued for another year.
Main Outcome Measures: Primary objectives were to compare changes in BMD 12 months after alloSCT at the femoral neck, lumbar spine, and total hip between the treatment arms and assess influences of glucocorticoid and cyclosporin therapy on these changes.
Results: Pamidronate reduced bone loss at the spine, femoral neck, and total hip by 5.6, 7.7, and 4.9% (all P 
0.003), respectively, at 12 months. However, BMD of the femoral neck and total hip was still 2.8 and 3.5% lower than baseline, respectively (P < 0.05) with pamidronate. Only differences at the total hip remained significant between the two groups at 24 months. Benefits were restricted to patients receiving an average daily prednisolone dose greater than 10 mg and cyclosporin therapy for more than 5 months within the first 6 months of alloSCT.
Conclusions: Pamidronate markedly reduced but did not completely prevent postallogeneic bone marrow transplantation bone loss. BMD benefits were greatest in patients on higher doses of immunosuppressive therapy, but most were lost 12 months after stopping pamidronate. Studies of more potent bisphosphonates or anabolic therapy with PTH after alloSCT are warranted with the aim of durable maintenance of bone mass.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The pathogenesis of alloSCT-related bone loss is complex (5). Both reduced bone formation and increased bone resorption occur. The main factor is the average daily dose or cumulative dose of glucocorticoids (2, 3, 6) given for treatment of graft vs. host disease (GVHD). Contributing factors include the duration of cyclosporin or tacrolimus therapy (6), immobilization, hypogonadism, vitamin D deficiency (7), and possibly a direct effect of conditioning regimens and GVHD on osteoprogenitor cells. Postchemotherapy ovarian failure is associated with rapid bone loss in women (5, 8). Bone marrow stromal cells, colony-forming unit fibroblasts, are of host origin after transplant and are damaged by high-dose chemotherapy and irradiation (9, 10), leading to reduced osteoblastic differentiation from osteoprogenitors (11). Osteoclasts are also activated, both directly by cytokines (12) and indirectly by increased PTH secretion (5).
Clinically overt fragility fractures occurred in 10 of 280 allogeneic bone marrow transplantation (BMT) survivors followed up for 4–7.4 yr after transplant (3). Another study reported an incidence of nontraumatic fractures in 11% of patients within 3 yr of transplant (6). Of note, 30–50% of nontransplant patients taking prednisolone at more than 7.5 mg/d for longer than 6 months will ultimately develop an osteoporosis-related fracture (13, 14).
Standard approaches for the prevention of posttransplant bone loss have included calcium and vitamin D supplementation and hormone therapy (HT). There are no prospective studies of vitamin D treatment after allograft, although results of trials with 1,25-(OH)2D3 (calcitriol) after solid organ transplants are promising (15). Although calcium and vitamin D are ineffective in preventing bone loss in this context (7, 16), higher doses of vitamin D or calcitriol improve BMD in patients with vitamin D deficiency and intestinal GVHD (17, 18). Whereas osteopenia diagnosed late after transplant in women not receiving HT can improve with subsequent estrogen therapy (19), early glucocorticoid-induced bone loss is not prevented by HT (2).
This multicenter Australian study evaluated the efficacy of iv pamidronate in preventing bone loss after alloSCT. Bisphosphonates are potent inhibitors of bone resorption, which act by inhibiting the action of osteoclasts and inducing their apoptosis (20). Oral bisphosphonates reduce glucocorticoid-induced bone loss (21) and prevent bone loss after cardiac or renal transplantation (15).
Based on the experience in myeloma (22, 23), the study randomized eligible patients undergoing alloSCT to receive monthly pamidronate infusions for 12 months or no pamidronate.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Eligible patients had undergone an alloSCT irrespective of donor source, stem cell type, or intensity of conditioning regimen and had greater than 20% expected long-term survival. Patients with myeloma were not eligible. The Human Research Ethics Committee at each of the participating institutions approved the protocol, and all patients gave informed consent.
Protocol
Patients were randomized to receive iv pamidronate 90-mg infusions or no pamidronate for 12 months, beginning a week before pretransplant chemotherapy. The study was not blinded. All female patients received HT with an oral estrogen-containing preparation and progestogen in the majority. All patients received calcitriol 0.25 µg daily and oral calcium carbonate supplements 1000 mg/d until study completion at 24 months. Testosterone supplementation was not routine, even in men with low serum testosterone concentrations.
Investigations
Before transplant and at 3, 6, 12, and 24 months after transplant. The following were investigated: 1) lumbar spine, total hip, and femoral neck BMD and total body bone mineral content; 2) serum markers of bone formation and urine markers of bone resorption; and 3) serum total testosterone levels in men and estradiol levels in women.
Before transplant and at 6, 12, and 24 months. Serum concentrations of 25-hydroxyvitamin D [25(OH)D], 1,25-dihydroxyvitamin D [1,25(OH)2D], PTH, FSH, and LH were investigated.
Before transplant and at 12 months. In a subset of 20 patients, plain x-rays of the thoracolumbar spine were investigated.
Other measures. In addition to these investigations, the following data were collected. First, the following pretransplant variables were investigated: age, sex, conditioning regimen, donor source, cell type (peripheral blood vs. marrow), and disease status at transplant. Myeloablative regimens were busulfan-cyclophosphamide and total body irradiation with either etoposide or cyclophosphamide. Reduced intensity regimens generally contained fludarabine with either cyclophosphamide or melphalan.
Second, after transplant the mean daily prednisolone dose between 0–3, 3–6, and 6–12 months after transplant. Glucocorticoid use was not protocolized. In general, significant GVHD was treated with 2 mg/kg oral or iv prednisolone, with subsequent doses and duration of therapy according to response. In addition, the number of days of cyclosporin or tacrolimus, maximum grade of acute and chronic GVHD, and episodes of clinical fractures and avascular necrosis were noted. Cyclosporin was given in full dose to 60 d in all patients. In the absence of GVHD, it was then tapered according to type of donor (sibling matched vs. unrelated) and risk of relapse. Cyclosporin was not tapered in the presence of GVHD and was resumed in patients who redeveloped GVHD.
Primary study end points were the changes in BMD at 12 months after alloSCT. The changes in BMD between treatment arms were compared with account for any influences of glucocorticoid and cyclosporin therapy. Secondary end points included the changes in BMD at 3, 6, and 24 months; changes in biochemical markers of bone turnover; the incidence of vertebral fractures assessed quantitatively by loss of vertebral height greater than 20%; and the incidence of clinical fractures and avascular necrosis.
Patient numbers
Previous studies of oral bisphosphonates in patients receiving glucocorticoids demonstrated a 6% difference in femoral neck BMD between treatment and control groups at 12 months (20). To detect a 6% difference in femoral neck BMD at 12 months with 80% power and one-tailed P = 0.05, 40 patients were required in each arm of this study. Assuming a 5% transplant-related mortality at 3 months and 15% mortality between 3 and 12 months, it was calculated that 20% of enrolled patients would not be evaluable at 12 months. Thus, 120 patients were enrolled in the study, aiming to accrue 80 evaluable patients.
Definition of candidate covariate factors
The investigation of potential factors associated with changes in BMD at 12 months was confined to the following: sex; age at BMT categorized into four levels (<30 yr, 30–39, 40–49, 50+ yr); maximum severity of acute GVHD at 3 months (grade 0, 1, 2, 3+); chronic GVHD in the first 6-month period (nil, limited, extensive); average daily steroid dose in the first 6 and the second 6 months after transplant, categorized into three levels: low (<10 mg), medium (10–39 mg), or high (40+ mg); duration of cyclosporin use in the first 6 months categorized as short (less than 90 d in first 3 months or less than 60 d in the second 3 months) or long (continuous use in the first 3 months and more than 60 d in the second 3 months); and duration of cyclosporin use in the second 6 months after transplant categorized as short (<3 months) or long (>3 months).
Bone densitometry
BMD was measured at the spine and proximal femur by dual-energy x-ray absorptiometry using Hologic (Bedford, MA) or Lunar (Madison, WI) bone densitometers. Coefficients of variation at the lumbar spine femoral neck were 1 and 1.7%, respectively. Lumbar spine and proximal femur BMD values obtained using Lunar instruments were adjusted using standardization formulae at baseline to make them comparable with those obtained using Hologic instruments (24).
Biochemical and hormone measurements
Total testosterone was measured by RIA (DPC Coat-A-Count; Diagnostic Products Corp., Los Angeles, CA). SHBG was measured by immunoradiometric assay (Orion SHBG kit; Espoo, Finland). LH and FSH were measured by AxSym autoanalyzer (Abbott, Abbott Park, IL). PTH was measured by immunochemiluminometric assay (Immulite 2000 Intact PTH; Diagnostic Products Corp.). 25(OH)D was measured by RIA (Diasorin, Stillwater, MN). 1,25(OH)2D was measured by RIA (Immunodiagnostic Systems, Bolden, UK). Total urine deoxypyridinoline corrected for creatinine was measured by HPLC. Serum total alkaline phosphatase (ALP) was measured by the AFCC method using an Olympus 2700 autoanalyzer (Tokyo, Japan) and osteocalcin by a two-site immunoradiometric assay (Diagnostic Systems Laboratories, Inc., Webster, TX).
Statistical methods
For each assessment time, 3, 6, 12, and 24 months, ANOVA was used to compare percentage changes in BMD, from the pretransplant value, in the two treatment groups. Analyses for changes in BMD at 1 yr were also repeated with the inclusion, one factor at a time, of the candidate covariate factors for age group; sex; incidence and severity of GVHD; steroid dose; and cyclosporin use. F tests based on the type II sums of squares were used to assess treatment differences adjusted for a covariate and the interaction of the treatment with the covariate factor. For incidence of avascular necrosis and clinical fractures, Fisher’s exact test was used to test for associations between incidence and treatment group. Fisher’s exact test was also used to test for associations between treatment group and severity of acute and chronic GVHD.
For serum osteocalcin, hormone levels, total ALP, albumin, and calcium, the ANOVA was used to compare treatment groups, sex, and age groups at each sample period. Overall survival in the two groups was estimated by the Kaplan-Meier method and compared using the log-rank test. For all comparisons, a significance level of 0.05 (not adjusted for multiple comparisons) was used. All analyses were conducted using procedures in SAS (version 9.1; SAS Institute, Cary, NC) and S-Plus 2000 (MathSoft Inc., Cambridge, MA).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
details the baseline variables of the study population. Only transplant type differed significantly (P = 0.043), with proportionally more myeloablative transplants in the control group than in the pamidronate group. The majority of patients had normal spinal and proximal femur BMDs. Although corrected serum calcium levels were normal, serum PTH levels were increased in both groups at baseline (see Table 4). Serum 25(OH)D and 1,25(OH)2D concentrations were normal in both groups. ALP and osteocalcin levels were normal, whereas urine deoxypyridinoline levels were increased in both groups, indicating increased bone resorption and uncoupling of bone remodeling (see Table 4). In women FSH levels were increased, although estradiol levels were in the low premenopausal normal range. Men had normal testosterone levels.
|
|
depicts the study profile and the available data. Missing BMD data at 3 and 6 months occurred predominantly because of investigator oversight, but scans were performed on 79 of 81 eligible patients at 12 months. BMD results were not available in a quarter of patients at 2 yr, predominantly due to patient withdrawal from the study because of progressive disease or death.
|
Mean percentage changes in lumbar spine, femoral neck, and total hip BMD at 3, 6, 12, and 24 months after transplant are depicted in Figs. 2-4, respectively. At the lumbar spine, the differences were generally smaller but significant at 12 months, being 5.6% greater in the pamidronate than the no-pamidronate group (Fig. 2). However, at 24 months the 3.0% difference between treatment groups was not significant.
|
|
|
). Pamidronate therapy decreased femoral neck bone loss by 7.7% at 12 months; however, significant bone loss (–2.8%) from baseline still occurred. Pamidronate treatment also decreased femoral neck bone loss at 3 and 6 months, but at 24 months the 3.4% difference between treatment groups was not significant. From 12 to 24 months, further bone loss of 3.0% [95% confidence interval (CI) 0.9–5.1%] occurred in the group previously treated with pamidronate (P = 0.005).
Total hip BMD decreased by 8.4% at 12 months in patients not receiving pamidronate (Fig. 4). Pamidronate therapy decreased bone loss at the total hip by 4.9% at 12 months; however, significant bone loss (–3.5%) from baseline also occurred in this group. Pamidronate treatment also decreased bone loss from the total hip at 3 and 6 and 24 months, with a persisting 3.9% (95% CI, 0.1–7.6%) greater BMD in the group previously treated with pamidronate at 24 months. Inclusion of a covariate to account for the imbalance in type of conditioning between the arms (Table 1) resulted in an estimate of 3.5% (95% CI, –1.0 to 8.0) greater total hip BMD in the pamidronate group at 24 months with neither the effect of the treatment (P = 0.128) nor the covariate (P = 0.116) being significant. This covariate was not significant, and the significance of treatment group differences was unaltered in analyses of covariance for all other BMD changes at all other times.
Effect of treatment and use of glucocorticoids or cyclosporin
Tables 2 and 3 detail the percentage changes in BMD at 12 months at each site according to mean daily glucocorticoid dose and the duration of cyclosporin use in the first 6 months, respectively, in each treatment group. Statistical interactions of glucocorticoid dose in the first 6 months and duration of cyclosporin use in the first 6 months with the treatment were significant for BMD changes at 12 months at all sites except for the interaction of cyclosporin duration with treatment at the total hip (P = 0.074). Patients receiving low mean glucocorticoid doses lost bone from the total hip and femoral neck but not from the spine. Site-specific changes in BMD were not altered by pamidronate therapy in this group.
|
|
There were no significant interactions of sex, age, severity of acute GVHD in the first 3 months, chronic GVHD at 6 months, mean steroid dose in the second 6 months, and duration of cyclosporin use in the second 6 months with the treatment on bone loss at 12 months at any site.
Bony complications
There was no difference in rates of avascular necrosis in each group. Three cases occurred in the first year (one in the pamidronate group and two in the no-pamidronate group). One case occurred in the second year in the pamidronate group. Magnetic resonance imaging was used to diagnose three of four cases, and the fourth was a clinical diagnosis. Two cases occurred in multiple lower limb joints.
No differences in clinical fractures occurred in each group. Two occurred in the first year, and a hip fracture occurred during the second year in a patient who had received pamidronate. In a subset of 20 patients, only one patient developed a new mild vertebral deformity at 12 months.
Serum hormone and calcium concentrations
Serum hormone and calcium concentrations for each treatment group are summarized in Table 4. There were no significant differences between the treatment groups and sexes at the assessment times for corrected serum calcium, serum 25(OH)D, and PTH concentrations. PTH levels were increased at baseline but decreased to the normal range thereafter (Fig. 5). Serum 25(OH)D concentrations were within the normal range at all times but were somewhat lower at 3 and 6 months, compared with baseline. Mean estradiol levels in women remained above the values seen in postmenopausal women, consistent with hormone replacement therapy. In men, mean testosterone levels fell to the lower limit of the normal range at 3 and 6 months but increased toward baseline at 12 months.
|
Serum biochemical bone turnover markers for each treatment group are summarized in Table 4. At 1 yr, serum osteocalcin levels were significantly lower in males treated with pamidronate than in the no-pamidronate group (mean change –10.6 vs. +3.2 ng/ml; P < 0.001), an effect seen as early as 3 months. Pamidronate therapy did not affect osteocalcin values in females. Total ALP concentrations were normal in all patients at baseline but increased in both study arms after alloSCT. Increases in the pamidronate group were smaller at 3 and 6 months. Because of a change in assay and incomplete transport of urine collections, urine deoxypyridinoline data were not available after baseline.
Survival
There was no overall difference in survival between the two groups at 24 months after BMT; however, there were fewer deaths in the first year after BMT in the patients treated with pamidronate: 83.7% (95% CI 72.3–91.0%) survival vs. 64.5% (95% CI 50.6–76.4%).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
There are caveats, however, in the interpretation of our results. Bone loss (
3%) from the femoral neck and total hip was not completely prevented by pamidronate treatment, and the differences between the two groups tended to diminish in the second year after transplant. We cannot exclude the possibility that the latter observation may represent imbalances in glucocorticoid dose and/or cyclosporin duration during this period. Fewer 2-yr BMD data may have compromised the power of the analysis at this time point to demonstrate a difference. Nevertheless, it is conceivable that the protective effect of pamidronate on bone mass is not sustained after treatment cessation.
Subsequent to the initiation of this study, Kananen et al. (25) demonstrated that six doses of 60 mg iv pamidronate over 9 months after allograft, combined with sex hormone therapy, calcium, and vitamin D prevented spinal bone loss but did not prevent bone loss from the proximal femur, compared with sex hormone therapy, calcium, and vitamin D alone. These findings of a reduction in, but lack of, prevention of bone loss at the proximal femur are similar to ours despite the cumulative dose of pamidronate being about three times higher in our study. This may reflect a failure of pamidronate to inhibit matrix metalloproteinase-mediated bone resorption or reverse the osteoblastic defect present after BMT (26).
Recent studies after solid organ transplantation have also demonstrated the efficacy of both iv and oral bisphosphonates in preventing bone loss (15). Pamidronate 0.5 mg/kg at the time of renal transplant and 1 month later prevented bone loss (27). Oral alendronate is also effective in preventing bone loss after renal or cardiac transplantation (28).
Pamidronate at an average dose of 30 mg thrice monthly also significantly increased spinal BMD in patients on glucocorticoid therapy (29). At a dose of 60 mg thrice monthly, pamidronate also prevented bone loss in the lumbar spine and hip in men receiving androgen deprivation for prostate cancer (30).
The optimal duration and frequency of bisphosphonate therapy for transplantation osteoporosis is unclear. About half of the administered bisphosphonate dose is rapidly bound to active sites of bone resorption on osteoclastic surfaces, and the remainder is renally excreted. Short-term exposure inhibits activity of a single generation of basic multicellular units (BMUs). In the long-term, however, bisphosphonates are also deposited on osteoblastic and resting bone surfaces, extending bisphosphonate activity beyond the life of a BMU as could direct effects on existing BMUs and osteocytes (31). Whereas disturbances in bone metabolism persist more than 5 yr after allograft (32), BMD is relatively constant after 2 yr in most patients (3).
The potency of the bisphosphonate may impact on the frequency of therapy required to prevent bone loss. However, the use of the potent bisphosphonates, oral risedronate (33) or iv zoledronate (34), has been limited to studies in which therapy was given 12 months after BMT. Both prevented bone loss, and iv zoledronate also increased ex vivo growth of the bone marrow colony-forming unit fibroblasts (34), suggesting that this potent bisphosphonate may be effective at improving osteoblast recovery and increasing osteoblast numbers after BMT. Our own data show that a single 4-mg dose of zoledronate prevents further bone loss from the proximal femur over the subsequent 3 months in patients with bone loss in the first year after alloSCT (35). Zoledronate has recently been shown to prevent bone loss after liver transplantation (36).
In general, pamidronate was well tolerated. After one early patient developed symptomatic hypocalcemia after the first dose (37), we ensured all patients received calcium and calcitriol before this time. Serum calcium levels increased during the study, whereas 25(OH)D levels decreased. The reason for this decrease is unclear, but it may reflect reduced sunlight exposure after alloSCT or be secondary to calcitriol administration, which also may have decreased serum PTH concentrations. The cause of the baseline elevation in serum PTH levels is unclear. Although subtypes of ALP were not measured, an increase in the liver subtype could contribute to the increase in total ALP after alloSCT. This could reflect mild hepatic dysfunction, secondary to GVHD.
We did not observe any cases of jaw osteonecrosis (38) or osteopetrosis (39), associated with prolonged intensive pamidronate therapy. Although first-generation bisphosphonates inhibit bone mineralization and have been associated with stress fractures (40), prolonged iv bisphosphonate therapy has not been associated with stress fractures of long bones (41).
Avascular necrosis is the most serious early skeletal complication after allograft, developing in 5–20% of patients within 2 yr (42). The femoral head is most often affected. The cumulative glucocorticoid dose used in treatment of chronic GVHD is the most important risk factor. Whereas it has been suggested that the pathogenesis of avascular necrosis is immune-related vasculitis in the context of chronic GVHD and unrelated to bone loss (43), these interactions have not been clarified. The incidence of avascular necrosis was low in both arms of our study; a much larger study would be required to evaluate the impact of anti-osteoporosis therapy on this complication.
In conclusion, high-dose monthly iv pamidronate therapy significantly reduced bone loss in the first year after alloSCT. Whereas the greatest impact was at the femoral neck and total hip, clinically relevant loss still occurred at both sites. The benefit was restricted to patients receiving an average daily prednisolone dose greater than 10 mg and prolonged cyclosporin therapy within the first 6 months of alloSCT and was not maintained after cessation of pamidronate. Given these results and the preliminary primary data emerging for zoledronate, studies of this more potent bisphosphonate or anabolic therapy with PTH after alloSCT are warranted with the aim of durable maintenance of bone mass.
|
Footnotes |
|---|
First Published Online July 11, 2006
Abbreviations: alloSCT, Allogeneic stem cell transplantation; ALP, alkaline phosphatase; BMD, bone mineral density; BMT, bone marrow transplantation; BMU, basic multicellular unit; CI, confidence interval; GVHD, graft vs. host disease; HT, hormone therapy; 1,25(OH)2D, 1,25-dihydroxyvitamin D; 25(OH)D, 25-hydroxyvitamin D.
Received March 29, 2006.
Accepted July 5, 2006.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3 serum levels during allogeneic bone marrow transplantation. Bone Marrow Transplant 33:871–873[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  P. R. Ebeling Approach to the Patient with Transplantation-Related Bone Loss J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1483 - 1490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Drake, B. L. Clarke, and S. Khosla Bisphosphonates: Mechanism of Action and Role in Clinical Practice Mayo Clin. Proc., September 1, 2008; 83(9): 1032 - 1045. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Tauchmanova, A. Colao, G. Lombardi, B. Rotoli, and C. Selleri REVIEW: Bone Loss and Its Management in Long-Term Survivors from Allogeneic Stem Cell Transplantation J. Clin. Endocrinol. Metab., December 1, 2007; 92(12): 4536 - 4545. [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 |
