| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institute of Biomedical Engineering (A.M.), National Research Council, I-35127 Padova, Italy; Department of Endocrinology, Diabetes, and Rheumatology (W.A.S.), University Hospital Duesseldorf, D-40225 Duesseldorf, Germany; University Hospital (P.M.N.), 205 02 Malmö, Sweden; Medical Group (G.L.), 40000 Mont de Marsan, France; Novartis Pharma AG (A.S.), CH-4056 Basel, Switzerland; PharmaWrite (B.E.D.), Princeton, New Jersey 08540; and Novartis Pharmaceuticals Corporation (S.J., J.E.F.), East Hanover, New Jersey 07936
Address all correspondence and requests for reprints to: Andrea Mari, Ph.D., ISIB-CNR, Corso Stati Uniti 4, 35127 Padova, Italy. E-mail: andrea.mari{at}isib.cnr.it.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Design: A 52-wk double-blind, randomized, parallel-group study comparing vildagliptin (50 mg every day) and placebo was conducted in 306 patients with mild hyperglycemia (glycosylated hemoglobin of 6.2–7.5%). Plasma glucose and C-peptide levels were measured during standard meal tests performed at baseline, wk 24 and 52, and after 4-wk washout. Insulin secretory rate (ISR) was calculated by C-peptide deconvolution, and β-cell function was quantified with a mathematical model that describes ISR as a function of absolute glucose levels (insulin secretory tone and glucose sensitivity), the glucose rate of change (rate sensitivity), and a potentiation factor.
Results: Vildagliptin significantly increased fasting insulin secretory tone [between-group difference in adjusted mean change from baseline to wk 52 (AM
) = +34.1 ± 9.5 pmol·min–1·m–2, P < 0.001] glucose sensitivity (AM = +20.7 ± 5.2 pmol·min–1·m–2·mM–1, P < 0.001), and rate sensitivity (AM = +163.6 ± 67.0 pmol·m–2·mM–1, P = 0.015), but total insulin secretion (ISR area under the curve at 0–2 h) and the potentiation factor excursion during meals were unchanged. These improvements in β-cell function were accompanied by a decrease in the glucose area under the curve at 0–2 h (AM = –1.7 ± 0.5 mM/h, P = 0.002) and in glycosylated hemoglobin (AM = –0.3 ± 0.1%, P < 0.001). None of the effects of vildagliptin remained after 4-wk washout from study medication.
Conclusions: Consistent with previous findings from shorter-term studies in patients with more severe hyperglycemia, in patients with mild hyperglycemia, improved β-cell function is maintained throughout 52-wk treatment with vildagliptin and underlies a sustained improvement in glycemic control. However, no effects remain after washout.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
- and β-cell responsiveness to glucose (1, 2, 3). However, the effects of vildagliptin on β-cell function have only been examined in detail in one small, 4-wk study of drug-naive patients with type 2 diabetes mellitus (T2DM) and relatively severe hyperglycemia [glycosylated hemoglobin (A1C) of 6.5–10%; mean fasting plasma glucose (FPG) of 8.9 mmol/liter)] (4). In that study, vildagliptin (100 mg twice per day) was found to increase insulin secretory tone [insulin secretory rate (ISR) at 7 mmol/liter glucose], but only a nonsignificant trend toward increased glucose sensitivity (slope of the glucose dose/β-cell response curve) was observed (4). Because endogenous glucagon-like peptide-1 (GLP-1) is thought to make a major contribution to the therapeutic effects of DPP-4 inhibitors (5) and because exogenous GLP-1 increases β-cell glucose sensitivity in healthy volunteers (6, 7), the lack of a significant effect of vildagliptin to increase glucose sensitivity in our previous 4-wk study of patients with moderate-to-severe hyperglycemia could be attributable to inadequate statistical power (n = 9 and 11 for the vildagliptin and placebo treatment groups, respectively), to the short duration of treatment, or to the advanced disease state, and relatively severe β-cell impairment. The present multicenter, randomized, placebo-controlled study examined the effects of 1-yr treatment with vildagliptin on model-assessed β-cell function in patients with T2DM and only mild hyperglycemia.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This was a 52-wk, double-blind, randomized, placebo-controlled, parallel-group study with a 4-wk, single-blind washout period conducted in 69 sites in Finland (three), France (four), Germany (42), Romania (five), Spain (seven), and Sweden (eight). Each patient attended one screening visit (wk –2), during which inclusion/exclusion criteria were assessed. Eligible patients were randomized at visit 2 (baseline) to receive vildagliptin 50 mg once per day (qd) or placebo. Patients received individualized lifestyle counseling (recommendations to reduce weight by decreasing fat and total caloric intake, follow a healthy diet, and increase physical activity) at each study visit. Vildagliptin was given once daily, 30 min before the breakfast meal. Model-assessed β-cell function was determined at baseline and at wk 24 and 52 of active treatment and at wk 56, after a 4-wk active-treatment-free period.
Study population
The study enrolled drug-naive patients aged 
18 yr who were diagnosed with T2DM at least 8 wk previously and who had A1C in the range of 6.2–7.5% at the screening visit (upper limit of 7.0% for centers in Finland and Spain). Patients who had taken no oral antidiabetic drug for at least 12 wk before screening and no oral antidiabetic drug for more than 3 consecutive months at any time in the past were considered to be representative of a drug-naive population. Male and female (nonfertile or of childbearing potential using a medically approved birth control method) patients with a body mass index (BMI) in the range of 22–45 kg/m2, inclusive, were eligible to participate.
Patients were excluded if they had a history of type 1 or secondary forms of diabetes, acute metabolic diabetic complications within the past 6 months, or evidence of significant diabetic complications. A history of significant cardiac arrhythmia, congestive heart failure, New York Heart Association class III or IV, and liver disease such as cirrhosis or chronic active hepatitis also precluded participation, as did any significant laboratory abnormalities.
Study assessments
Standard breakfast meal tests (500 kcal; 60% carbohydrate, 30% fat, and 10% protein) were performed after an overnight fast at baseline and at wk 24, 52, and 56. Samples for determination of glucose, insulin, and C-peptide were obtained at times –20, 0, 15, 30, 60, 90, 120, 180, and 240 min, with the meal beginning immediately after the time 0 sample and consumed within 15 min. During the active treatment period, study medication was taken 15 min before the standard meal challenge. A1C was measured at all study visits (wk –2, baseline, wk 4, 8, 12, 16, 24, 32, 40, and 52 of active treatment, and wk 56). All laboratory assessments were made by Bioanalytical Research Corporation-European Union (Ghent, Belgium). Assays were performed according to standardized and validated procedures according to good laboratory practice.
Modeling analysis of β-cell function
ISRs were calculated from plasma C-peptide levels by deconvolution (8) and expressed per square meter of estimated body surface area. The dependence of ISR on glucose levels was modeled separately for each patient and each study day. The β-cell model used in the present study, describing the relationship between insulin secretion and glucose concentration, has been described previously in detail (9, 10).
Insulin secretion consists of two components. The first component represents the dependence of insulin secretion on absolute glucose concentration (G) at any time point and is characterized by a dose-response function, f(G), relating the two variables. Characteristic parameters of the dose response are insulin secretion at a fixed glucose concentration of 7 mmol/liter (approximately the fasting glucose level in mildly diabetic subjects) and the mean slope in the observed glucose range. The dose response is modulated by a potentiation factor, P(t), which accounts for several potentiating agents (prolonged exposure to hyperglycemia, nonglucose substrates, gastrointestinal hormones, and neurotransmitters). The first secretion component is thus the product, P(t)f(G).
The potentiation factor is set to be a positive function of time and to average one during the experiment. It thus expresses a relative potentiation of the secretory response to glucose. The excursion of the potentiation factor was quantified using ratios between mean values at times 220–240 and 0–20 min.
The second insulin secretion component represents a dynamic dependence of insulin secretion on the rate of change of glucose concentration. This component is termed the derivative component and is determined by a single parameter. Rate sensitivity is related to early insulin release (9, 10).
The model parameters [the parameters of the dose response, f(G), and the potentiation factor, P(t)] were estimated from glucose and C-peptide concentration by regularized least squares, as described previously (9, 10). Regularization involves the choice of smoothing factors that were selected to obtain glucose and C-peptide model residuals with SDs close to the expected measurement error (
1% for glucose and 5% for C-peptide). Estimation of the individual model parameters was performed blinded to the randomization of patients to treatment.
Data analysis
The changes from baseline to wk 52 or endpoint in the β-cell model parameters were analyzed using an analysis of covariance (ANCOVA) model with treatment and pooled center as classification variables and baseline value as the covariate. The changes from baseline to wk 56 were analyzed with the same ANCOVA model to determine whether any treatment-associated effects remained after the 4-wk washout. Analyses were conducted using two-sided tests and a significant level of 0.05.
Ethics and good clinical practice
All participants provided written informed consent. The protocol was approved by the independent ethics committee/institutional review board at each study site, and the study was conducted in accordance with the Declaration of Helsinki, using Good Clinical Practice.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Table 1
reports the baseline demographic and metabolic characteristics and disposition of all randomized patients. The groups were well balanced at baseline, with A1C averaging 6.7% and FPG averaging 7.1 mmol/liter. Approximately 60% of patients were male and nearly all were Caucasian, with a mean age of approximately 63 yr. More than 45% of patients in either treatment group were obese, with a mean BMI of 30.2 kg/m2 and mean disease duration of 2.6 yr. Anti-glutamic acid decarboxylase antibodies were undetectable in more than 95% of participants. A similarly high percentage of patients in each treatment group completed the study. A higher percentage of patients assigned to vildagliptin discontinued because of an adverse event (9 vs. 4% in the placebo group). Otherwise, a similarly low percentage of patients in each treatment group discontinued for any specific reason.
|
Figure 1 depicts plasma glucose, insulin, and C-peptide levels during standard meal tests performed at baseline and after 52-wk treatment. As seen in Fig. 1, A and D, the glucose profiles in the two groups of patients were similar at baseline, but postmeal hyperglycemia worsened somewhat during 52 wk of placebo administration and improved during vildagliptin treatment. At baseline, the glucose area under the curve at 0–2 h (AUC0–2 h) averaged 19.8 ± 0.3 mmol/l·h in patients randomized to vildagliptin and 20.2 ± 0.3 mmol/l·h in those randomized to placebo. The adjusted mean change (AM) was –1.0 ± 0.4 mmol/l·h in patients receiving vildagliptin and +0.7 ± 0.4 mmol/l·h in those receiving placebo. The between-group difference in the AM glucose AUC0–2 h was –1.7 ± 0.5 mmol/l·h (P = 0.002).
|
). In contrast, at wk 52, the postmeal insulin levels were slightly higher in the vildagliptin group than in the placebo group at the early time points and essentially identical at the later time points (Fig. 1E), despite a reduced glycemic stimulus. The insulin AUC0–2 h averaged 702 ± 42 pmol/l·h in patients randomized to vildagliptin and 800 ± 45 pmol/l·h in those randomized to placebo. The AM was 1 ± 28 pmol/l·h in patients receiving vildagliptin and –107 ± 28 pmol/l·h in those receiving placebo. The between-group difference in the AM insulin AUC0–2 h was 108 ± 39 pmol/l·h (P = 0.006).
As would be expected, the C-peptide profiles were qualitatively very similar to the insulin profiles, with postmeal C-peptide levels being somewhat higher in the placebo group at baseline (Fig. 1C) and somewhat higher at the early time points in the vildagliptin group at wk 52 (Fig. 1F). The C-peptide AUC0–2 h averaged 4.28 ± 0.14 nmol/l·h in patients randomized to vildagliptin and 4.40 ± 0.14 nmol/l·h in those randomized to placebo. The AM was 0.04 ± 0.08 nmol/l·h in patients receiving vildagliptin and –0.20 ± 0.09 nmol/l·h in those receiving placebo. The between-group difference in the AM C-peptide AUC0–2 h was 0.24 ± 0.12 nmol/l·h (P = 0.043).
Model-derived parameters of β-cell function
Based on the data presented in Fig. 1, several parameters describing β-cell function were derived. These are illustrated in Fig. 2. Baseline values for ISR at 7 mmol/liter glucose averaged 204 pmol·min–1·m–2 in both treatment groups. Relative to placebo, vildagliptin significantly increased ISR at 7 mmol/liter glucose (P < 0.001), with the between-group difference in the AM (34 pmol·min–1·m–2) representing an approximately 17% increase (Fig. 2A). Baseline values for glucose sensitivity averaged 51 and 53 pmol·min–1·m–2·mM–1 in the vildagliptin and placebo groups, respectively. Relative to placebo, vildagliptin significantly increased glucose sensitivity (P < 0.001), with the between-group difference in the AM (20.7 pmol·min–1·m–2·mM–1) representing an approximately 40% increase (Fig. 2B). Baseline values for rate sensitivity averaged 526 and 430 pmol·m–2·mM–1 in patients randomized to vildagliptin and placebo, respectively. Vildagliptin significantly increased rate sensitivity (P = 0.015), with the between-group difference in the AM (164 pmol·m–2·mM–1) representing an approximately 34% increase (Fig. 2C). Baseline values for the excursion of the potentiation factor averaged 1.5 and 1.6 in patients randomized to vildagliptin and placebo, respectively. This excursion did not change during vildagliptin treatment and changed minimally (AM of 0.04) during placebo administration. The between-group difference in the AM was –0.03 ± 0.06 (P = 0.543) (Fig. 2D). The potentiation factor time course was similar in the two groups and was not affected by vildagliptin (data not shown).
|
s in ISR at 7 mmol/liter glucose, glucose sensitivity, and rate sensitivity were 38.5 ± 9.3 pmol·min–1·m–2, 18.5 ± 4.9 pmol·min–1·m–2·mM–1, and 176.5 ± 63.3 pmol·m–2·mM–1, respectively. In obese patients receiving vildagliptin (n = 64), the AMs in ISR at 7 mmol/liter glucose, glucose sensitivity, and rate sensitivity were 29.2 ± 10.3 pmol·min–1·m–2, 18.4 ± 6.0 pmol·min–1·m–2·mM–1, and 113.5 ± 76.0 pmol·m–2·mM–1, respectively. Because of the small sample size and reduced statistical power, not all of these changes achieved full statistical significance relative to placebo.
At baseline, the ISR at time 0 (ISR0min) averaged 143 and 147 pmol·min–1·m–2 in patients randomized to vildagliptin and placebo, respectively. ISR0min changed minimally in both groups, and there was no significant between-group difference (data not shown). Similarly, total postmeal insulin secretion (ISR AUC0–240min) was similar at baseline in the two treatment groups (65.5 and 68.9 nmol/m2 in the vildagliptin and placebo groups, respectively), and vildagliptin had no significant effect on this parameter (between-group difference in AM of 2.9 ± 2.0 nmol/m2, P = 0.143) (data not shown).
Time course of changes in β-cell function
In addition to the standard meal tests at baseline and wk 52, meal tests were performed and β-cell function was assessed at wk 24 and at wk 56, after the 4-wk placebo washout period. Table 2 summarizes β-cell function parameters throughout the 56-wk study. Although no statistical analyses were performed with data from the wk 24 meal test, it appears that the increases in ISR at 7 mmol/liter glucose and glucose sensitivity with vildagliptin treatment were fully manifest by wk 24. In contrast, mean values for the rate sensitivity parameter, which were substantially higher at baseline in patients randomized to vildagliptin than in those randomized to placebo, increased in both groups of patients at wk 24, so no vildagliptin treatment effect was apparent. As presented in Fig. 2, ISR at 7 mmol/liter glucose, glucose sensitivity, and rate sensitivity were each significantly increased in vildagliptin-treated patients relative to placebo. However, none of these effects on β-cell function remained at wk 56 after 4-wk washout from study medication.
|
Baseline A1C averaged 6.7% in patients randomized to vildagliptin and 6.8% in those randomized to placebo. A1C decreased modestly in vildagliptin-treated patients (AM of –0.2 ± 0.1%, n = 153) and increased in patients receiving placebo (AM of 0.1 ± 0.1%, n = 149). The between-group difference in AM was –0.3 ± 0.1% (P < 0.001). In those patients completing the 4-wk washout period (n = 135 and 130 for vildagliptin and placebo, respectively), the between-group difference in AM A1C (–0.1 ± 0.1%) was no longer statistically significant (P = 0.102).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The effects of vildagliptin to increase glucose sensitivity as well as insulin secretory tone seen in the present study are in keeping with the concept that GLP-1 makes a major contribution to the hormonal and metabolic effects of DPP-4 inhibitors (5). Thus, it has been shown that acute infusion of GLP-1 increases ISR at 7 mmol/liter glucose and glucose sensitivity in a small number of subjects with a range of glucose tolerance (7). Similar effects were also seen in patients with T2DM treated for 7 d with the GLP-1 receptor agonist liraglutide (11). Another study assessed β-cell function parameters after 30-wk treatment with exenatide in patients with T2DM continuing maximally effective doses of metformin or metformin plus a sulfonylurea (12). In that study, exenatide was also found to increase insulin secretory tone and to produce an upward shift in the dose response but with no change in slope (glucose sensitivity). The participants’ advanced disease state (FPG 9.3 mmol/liter while receiving maximal metformin with or without sulfonylurea) seems a plausible explanation for the lack of effect of exenatide to increase glucose sensitivity.
Two aspects of the influence of the DPP-4 inhibitor vildagliptin on β-cell function parameters differ from those of GLP-1 agonists. First, vildagliptin did not increase the excursion of the potentiation factor [potentiation factor ratio (PFR)], whereas GLP-1, exenatide, and liraglutide were each found to increase this component of insulin secretion (7, 11, 12). This may suggest that very high levels of GLP-1 receptor activation are required to increase potentiation, and the more physiologic levels achieved with DPP-4 inhibition do not affect this aspect of β-cell function. Consistent with this interpretation, Ahrén et al. (7) noted that the PFR was positively correlated with GLP-1 increments and that a more than 10-fold increase in GLP-1 was required to substantially increase the PFR. In addition, it has been shown that endogenous incretin hormones increase glucose and rate sensitivity but not the PFR (13), as with vildagliptin.
Also unlike findings obtained with GLP-1 receptor agonists, 52-wk treatment with vildagliptin significantly increased rate sensitivity. This component is necessary to describe the early phase of insulin secretion during a meal (10), which, in turn, may be related to the acute insulin response to iv glucose (14). Thus, the improvement in rate sensitivity seen with vildagliptin treatment in the present study is consistent with previous findings that vildagliptin partially restores the acute insulin response to iv glucose in patients with T2DM (15). This effect of vildagliptin is also in keeping with the observation that endogenous incretin hormones increase rate sensitivity in subjects with impaired glucose tolerance (13). The discrepancy between the actions of endogenous incretins [both GLP-1 and glucose-dependent insulinotropic peptide (GIP), presumably mediating the effects of vildagliptin] and exogenously administered GLP-1 agonists may again relate to the more physiologic mechanism of action of vildagliptin and the lower degree of GLP-1 receptor signaling. Because both GLP-1 (16) and exenatide (17) retard gastric emptying, an alteration of the glucose absorption pattern may blunt the efficacy of these agonists to enhance early insulin release during a meal.
In summary, the findings from the present study are primarily consistent with the concept that the endogenous incretin hormones mediate the effects of vildagliptin on β-cell function. However, it should be recognized that the data do not rule out contributions from DPP-4 substrates other than GLP-1 and GIP.
Accompanying the improvements in β-cell function during 52-wk treatment with vildagliptin (50 mg qd) in patients with T2DM and mild hyperglycemia described above, there was a modest but statistically significant decrease in A1C (–0.3%) relative to placebo. This has been discussed in greater detail previously (18). It is suggested by the present work, however, that the improvement in β-cell function made a major contribution to the glucose-lowering efficacy of vildagliptin. Although plasma glucagon levels were not measured in this study, it has been consistently seen in smaller studies (in which assessments of glucagon, GLP-1, and GIP were more feasible) that vildagliptin also suppresses inappropriate glucagon secretion (1, 4, 19, 20). Hence, it is likely that improved -cell function also contributes to the effects of vildagliptin on glycemic control.
The primary aim of the present work was to fully characterize the effects of vildagliptin on β-cell function, and this goal was achieved. In addition, we sought to determine whether 1-yr treatment was sufficient to detect a disease-modifying effect of this DPP-4 inhibitor, which could be hypothesized based on animal data demonstrating that GLP-1, GIP, and DPP-4 inhibitors (21) can reduce β-cell apoptosis, stimulate neogenesis and/or replication, and increase β-cell mass. Toward this end, the study included a 4-wk washout period, and β-cell function was assessed at wk 56. None of the improvements in β-cell function that were seen after 52-wk treatment with vildagliptin remained after 4-wk washout, and the decrease in A1C was similarly evanescent once treatment was discontinued. Thus, it must be concluded that DPP-4 inhibitors do not increase functional β-cell mass in a clinically meaningful manner over the course of 1-yr treatment.
In summary, in patients with T2DM and mild hyperglycemia, 52-wk treatment with vildagliptin (50 mg qd) improved three aspects of β-cell function, namely, insulin secretory tone, glucose sensitivity, and rate sensitivity. This was associated with a significant decrease in A1C. Because none of the effects of vildagliptin were maintained after a 4-wk washout period, we conclude that longer-term studies would be necessary to determine whether DPP-4 inhibition modifies disease progression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: A1C, Glycosylated hemoglobin; AM, adjusted mean change; ANCOVA, analysis of covariance; AUC, area under the curve; BMI, body mass index; DPP-4, dipeptidyl peptidase-IV; FPG, fasting plasma glucose; GIP, glucose-dependent insulinotropic peptide; GLP-1, glucagon-like peptide-1; ISR, insulin secretory rate; PFR, potentiation factor ratio; qd, once per day; T2DM, type 2 diabetes mellitus.
This study was funded by Novartis Pharmaceuticals Corporation.
This trial (NCT 00101712) is registered with ClinicalTrials.gov
Disclosure Statement: A.M. has received research grants and travel expenses from Novartis Pharmaceuticals Corporation. W.A.S. consults for Novartis Pharmaceuticals Corporation, Pfizer, Eli Lilly, Sanofi-Aventis, and Roche Diagnostics. P.M.N. has received fees for speaking for Merck, GlaxoSmithKline, AstraZeneca, Sanofi-Aventis, and Pfizer. G.L. has nothing to disclose. A.S. is an employee of and shareholder in Novartis Pharma AG. B.E.D. is a consultant for Novartis Pharmaceuticals Corporation. S.J. and J.E.F. are employees and shareholders of Novartis Pharmaceuticals Corporation.
Received July 23, 2007.
Accepted October 1, 2007.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Ahren, A. Schweizer, S. Dejager, B. E. Dunning, P. M. Nilsson, M. Persson, and J. E. Foley Vildagliptin Enhances Islet Responsiveness to Both Hyper- and Hypoglycemia in Patients with Type 2 Diabetes J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1236 - 1243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. D'Alessio, A. M. Denney, L. M. Hermiller, R. L. Prigeon, J. M. Martin, W. G. Tharp, M. L. Saylan, Y. He, B. E. Dunning, J. E. Foley, et al. Treatment with the Dipeptidyl Peptidase-4 Inhibitor Vildagliptin Improves Fasting Islet-Cell Function in Subjects with Type 2 Diabetes J. Clin. Endocrinol. Metab., January 1, 2009; 94(1): 81 - 88. [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 |
