| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Pediatric and Reproductive Endocrinology Branch, National Institute of Child Health and Human Development (K.P., I.I., L.K.N.), and Nuclear Medicine Department, Clinical Center (C.C.C., J.A.C., M.W.), National Institutes of Health, Bethesda, Maryland 20892-1583
Address all correspondence and requests for reprints to: Dr. Karel Pacak, Unit on Clinical Neuroendocrinology, Pediatric and Reproductive Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Building 10, Room 9D42, 10 Center Drive, Bethesda, Maryland 20892-1583. E-mail: karel{at}mail.nih.gov.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Functional imaging with nuclear medicine techniques enables in vivo imaging of physiological and pathophysiological processes. Among these techniques, positron emission tomography (PET) studies are increasingly used in oncology (2). [18F]Fluorodeoxyglucose (FDG) whole body scanning is by far the most widely used and recognized application of PET for tumor identification (3), and although it often fails to visualize the more common differentiated tumors (4), poorly differentiated tumors with high proliferative activity may take up FDG (5, 6).
Scintigraphy with the somatostatin analog [111In]-diethylenetriaminepentaacetate-D-Phe-pentetreotide (octreoscan; OCT) at the usual injected radionuclide dose of 6 mCi (L-OCT) may not always detect neuroendocrine tumors, and previous studies with L-OCT have shown mixed results (7, 8, 9, 10, 11, 12). Detectability of lesions in scintigraphic studies depends on multiple factors, including lesion size, location, type, or degree of somatostatin receptor expression (13, 14, 15, 16, 17), and also on the amount of radioactivity in the lesion. Scans obtained after very high doses of OCT for therapy (180 mCi) have been reported to detect more lesions than pretherapy diagnostic scans in patients with gastroenteropancreatic tumors (18). Before this study a patient evaluated for EAS at our institution underwent OCT scintigraphy with a 12-mCi dose after a 6-mCi dose scan that yielded equivocal results. This higher dose gave definite results, and the patient went on to have a pulmonary carcinoid successfully resected (our unpublished observations). This case led us to consider the hypothesis that a higher dose of OCT might identify ACTH-secreting tumors not detectable on L-OCT investigations because of their small size or lesser degree of somatostatin receptor expression by increasing the absolute amount of radioactivity taken up by the tumor. In the present study, we also performed FDG-PET to assess the sensitivity of FDG for detection of ACTH-secreting tumors. These modalities were compared with the results of conventional imaging (CT, MRI, and L-OCT).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The research protocol was approved by the National Institute of Child Health and Human Development institutional review board, and written informed consent was obtained from all subjects. Subjects were eligible for this study if they were hypercortisolemic and had a normal or increased plasma ACTH concentration. To verify the cause of CS, we performed and interpreted bilateral inferior petrosal sinus sampling (19), an ovine CRH stimulation test (20), and/or an 8-mg overnight dexamethasone suppression test (21), as previously described.
Patients included in this study did not show a central to peripheral ACTH gradient on inferior petrosal sinus sampling and had normal petrosal sinus venous anatomy. They were considered to have EAS and underwent anatomical imaging studies (CT and MRI of the neck, chest, and abdomen) as well as L-OCT for the localization of ACTH-secreting tumor. Additional research imaging studies included higher than standard dose OCT (H-OCT) scintigraphy, if L-OCT was negative, and FDG-PET. Urinary catecholamines and metanephrines and plasma metanephrines were measured in patients with hypertension to exclude pheochromocytoma.
After evaluation during the protocol, patients with presumed EAS tumor underwent surgery or biopsy. Patients with an unknown/occult source of EAS received steroidogenesis inhibitors or adrenalectomy to normalize cortisol levels. Some of these patients returned to the NIH Clinical Center to have repeated imaging studies. The protocol allowed only two FDG and two H-OCT studies at more than a 12-month interval. Other imaging studies were performed on multiple occasions. In three patients with negative evaluation for the localization of the ACTH-secreting tumor at the time of the initial evaluation, subsequent studies guided successful surgery. Four patients continue with occult disease at the time of this report (two of them had multiple follow-up evaluations at the NIH Clinical Center).
CT and MRI studies were obtained before the research studies. CT was performed with a Hi Speed Advantage Scanner (General Electric Medical Systems, Milwaukee, WI). Section thickness was 5 mm through the chest and upper abdomen to the adrenal gland and 10 mm through the lower abdomen. All sections were contiguous. A contrast agent was given by mouth. Most studies were performed during a bolus of nonionic water-soluble contrast (130 ml, injected at 2 ml/sec), given at the radiologist’s discretion.
MRI was obtained with a 1.5-T scanner (Picker, Highland Heights, OH). T1- and T2-weighted imaging was performed in all patients. Images were obtained in the coronal and axial planes on all studies. T1-weighted spin-echo (SE) imaging was performed with a repetition time (TR) of 300 msec, an echo time (TE) of 10 msec (TR/TE = 300/10), and eight excitations. T2-weighted spin-echo imaging was performed with a TR/TE ratio of 2000/80 and two excitations. In four patients short-time inversion-recovery sequence imaging was also performed with a TR of 1600 msec, a TE of 30 msec, an inversion time of 100 msec, and four excitations.
FDG was prepared by the method of Hamacher et al. (22). After fasting for at least 6 h, during which patients were encouraged to drink water, 20 mCi FDG was administered iv. After injection, the patients rested for 40 min, followed by whole body imaging on a GE Advance PET scanner (General Electric Medical Systems). Ten-minute two-dimensional mode emission and 8-min transmission scans were acquired. Images were acquired in a 256 x 256 matrix, and reconstruction was performed with an iterative algorithm. Images were reconstructed and reviewed by use of an orthogonal display as well as viewing maximal intensity reprojected images.
For L-OCT, patients were injected with 6 mCi OCT prepared from a single pentetreotide kit (10 µg; Mallinckrodt, St. Louis, MO) and were imaged at 4 and 24 h. For H-OCT, patients received 18 mCi OCT (prepared using three kits or 30 µg pentetreotide) and were imaged at 4, 24, and occasionally 48 h. At 4 h, whole body and single photon emission computed tomography (SPECT) images of skull, chest, abdomen, and pelvis were obtained, and at 24 h and 48 h, SPECT images were repeated as needed. Repeat scans (for both L-OCT and H-OCT examinations) were made when 4 h images were negative or equivocal, or when a positive finding at 4 h might be due to physiological uptake. For example, SPECT was repeated to ascertain that faint foci persisted on follow-up or to differentiate abdominal foci of uptake from activity in normal bowel. Images were obtained using an ADAC (Milipitas, CA) Vertex dual-head and a Trionix (Twinburg, OH) triple-head XLT 
- camera equipped with medium energy, parallel hole collimators centered over both 111In photon peaks (173 and 247 KeV) with 20% windows. For SPECT, 120 sequential, 30-sec (dual-head) or 120 sequential, 40-sec (triple-head) images were obtained. The images were reconstructed with the manufacturers’ software using a standard filtered back-projection algorithm. Hamming (Trionix) or Hanning (ADAC) filters were used.
Based on our experience, tumors producing ACTH are most commonly localized in the lungs, thymus, and pancreas (23). Therefore, any questionable finding (in CT/MRI or scintigraphies) in these regions was reevaluated with high resolution CT or MRI.
Readers of CT and MRI were not blinded to the results of other imaging studies, but these studies were generally interpreted before the research studies were performed, except when repeat studies for questionable findings were obtained. FDG-PET and OCT studies were each read by a nuclear medicine physician blinded to the results of all other imaging studies, except that as the H-OCT was performed only if L-OCT was negative, the L-OCT result was known by the H-OCT reader.
For sensitivity calculations, positive results of CT/MRI imaging were considered as either true positives (when the ACTH-secreting tumor was identified) or false positives (when the lesion seen was not the ACTH-secreting tumor), whereas negative results were considered to be false negatives (because the ACTH-secreting lesion was not seen). The PET images were evaluated qualitatively; focal areas of uptake believed to be nonphysiological were scored as positive (true or false; in the same way as CT/MRI) in a manner considered to be adequate for oncological PET studies (24, 25). In the same manner as PET studies, OCT scintigraphy was considered abnormal if definite nonphysiological uptake was observed. Mild bilateral adrenal uptake with either PET or OCT was not considered abnormal. Sensitivity was calculated as the number of true positive results per number of all patients studied with the modality (26).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Patients 1, 2, and 8 were initially admitted to the NIH Clinical Center 5–10 yr before the present study, at which time they were diagnosed with EAS and underwent surgical treatment. They were reevaluated in this study because of EAS recurrence. Ten of the 17 patients had successful tumor identification based on the initial imaging. Patients 5, 7, and 10, with negative initial imaging, later had successful tumor resection based on subsequent imaging studies. Four patients remain with occult disease.
|
|
Only the results of the patients’ initial imaging studies are presented (Table 1
). Of these, nine of 17 CT, six of 16 MRI, six of 17 FDG-PET, eight of 17 L-OCT, and one of nine H-OCT were true positives, as confirmed at subsequent surgery or biopsy in 13 patients (Fig. 1).
|
). On CT/MRI, false positive findings were seen in three patients (no. 2, 11, and 17), although in patient 11 the lesions were attributed to pulmonary infection. FDG-PET studies showed false positive uptake foci in four patients (no. 4, 6, 11, and 13). These included uptake in the chest wall (no. 4), in rib fractures (no. 6), bilateral hilar uptake in another patient (no. 13), and multiple lung foci in the patient with lung infection (no. 11). No false positive findings were noted with OCT.
CT and/or MRI in six patients (no. 4–7, 12, and 15) showed bilaterally hyperplastic adrenal glands. FDG-PET scanning in six patients (no. 3, 4, 5, 7, 11, and 15) and L-OCT and H-OCT in five patients (no. 3, 5, 7, 14 and 17) showed radionuclide uptake by the adrenals.
For sensitivity calculations, studies were scored as being true positive if the ACTH-secreting tumor was detected, even if false positive findings were also seen. Thus, the FDG-PET study of patient 6 was considered as being overall true positive because the focus of uptake in the lung coincided with the surgically verified lesion, and the false positive foci of uptake in the ribs coincided with healed rib fractures. The FDG-PET study of patient 11 was also considered as being true positive because the ACTH-secreting tumor was imaged whereas the multiple false positive foci of uptake in the chest were attributed to concomitant lung infection (nocardiosis). In patient 13, FDG-PET had a false positive finding of bilateral hilar uptake (Fig. 1); nevertheless, it also showed the adrenal lesion that was surgically verified, so overall, this study was also considered to be true positive. In localizing ACTH-secreting tumors, the sensitivity of CT or OCT scintigraphy (combining the results of both L-OCT and H-OCT) was higher (both 53%; 95% confidence interval, 29–76%) than that of MRI (37%; 95% confidence interval, 16–64%) or FDG-PET (35%; 95% confidence interval, 15–61%; Table 1).
Management and outcome
Based on initial or subsequent imaging results, 12 patients underwent surgery (no. 1–10, 13, and 17), including one with negative anatomical imaging and positive H-OCT (no. 4). In all patients who underwent thoracotomy, lymph node dissection was performed for oncological staging.
One patient (no. 17) was operated upon based on positive CT and MRI findings, but no tumor was found; in retrospect, these imaging results were considered to be false positives. Two others with positive imaging studies and biopsy-proven ACTH-secreting tumors did not undergo surgical resection (one with an ACTH-secreting olfactory esthesioneuroblastoma and one with small cell lung cancer; no. 11 and 12, respectively).
In primary or metastatic tumors from eight patients, ACTH staining was positive on immunohistochemistry (no. 1, 2, 4, 5, and 7–10). Patient 13 had negative tumor staining for ACTH, but hypercortisolism resolved after tumor resection. No ACTH stain was available in tumors from two patients (no. 3 and 6). ACTH staining was positive on biopsy tissue in two patients who were not operated upon (no. 11 and 12).
The resected tumors ranged in size from 0.6–7.0 cm. The smallest tumor detected in this study (excluding subsequent evaluations) with CT, MRI, FDG-PET, and L-OCT was 1.3 cm (in maximum dimension), and the smallest detected with H-OCT was 0.6 cm.
Four patients remained with occult disease, including two after reevaluation (no. 15 and 16).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A variety of factors may influence the ability of FDG-PET to localize a tumor. Increased tumoral metabolic rate and glucose transport through the cell membrane are necessary for increased uptake of FDG (27). FDG-PET is most likely to identify tumors with high proliferative activity. Relatively few endocrine tumors have been studied with FDG-PET, but, in general, the results support this concept. As shown in Table 2, a review of the literature shows that regardless of the pathological type of tumor, increased uptake of FDG is most likely in less well differentiated neuroendocrine tumors with high proliferative activity (4, 28, 29, 30, 31, 32, 33, 34, 35, 36). By contrast, FDG-PET seems to work less well to identify tumors with low proliferative activity (30, 31).
|
In contrast to the importance of proliferation for FDG-PET uptake, OCT images depend on the presence of somatostatin receptors. Somatostatin receptors (subtypes 1–5) have been identified on many cells of neuroendocrine origin (42). OCT, the currently available radiolabeled somatostatin analog for scintigraphic imaging, binds with high affinity to somatostatin receptors 2 and 5. These receptor subtypes are predominantly expressed in primary gastroenteropancreatic neuroendocrine tumors (gastrinomas, insulinomas, and carcinoid tumors) (43), lung tumors (typical and atypical carcinoids, neuroendocrine carcinomas, and small cell lung cancer) (15), and their metastases (44). Thus, visualization of these tumor types would be likely (30, 31, 32, 33, 36). Nevertheless, pulmonary carcinoids (particularly atypical ones) and pulmonary neuroendocrine tumors may show heterogeneity in the degree of expression of somatostatin receptors (45, 46).
One of the factors that determines whether a tumor can be imaged with OCT is the density of somatostatin receptors expressed in the tumor (47). Other factors implicated are tumor size, the type of receptors expressed, the affinity of radioligand for the subtype of receptors expressed by the tumor, the efficiency of receptor subtype-mediated binding and internalization of radioligand, the trapping of radioisotopes within the tumor cells, and the mass of the injected radiopharmaceutical (47).
In previous studies, various levels of success with L-OCT scintigraphy have been reported, perhaps reflecting differences in patient referral patterns or study methodology. In a retrospective, unblinded series of 10 patients with EAS, an ACTH-secreting tumor was visualized with L-OCT in eight patients and was missed in two patients (12). In a more recent retrospective, unblinded report of patients with EAS, L-OCT examinations were positive in two of two patients with medullary thyroid carcinoma and six of six patients with histologically proven ACTH-secreting pulmonary carcinoids (11). Of the latter group, three of six L-OCT were initially negative, becoming positive on follow-up 8–27 months later. In contrast, in an earlier retrospective series of 18 patients assessed at the NIH, no ACTH-secreting tumor was detected by L-OCT when CT/MRI were negative, and seven L-OCT results were falsely positive (9). As in the present study, interpretation of octreoscans, but not CT or MRI, was blinded. In another study, L-OCT (as well as CT and/or MRI) was positive in four of 12 patients with CS due to EAS; one patient had a false positive L-OCT, and three patients had true positive CT and/or MRI, but negative L-OCT studies (10).
The present prospective study, using blinded readers for interpretation of scintigraphic studies, but not CT or MRI, found that conventional L-OCT worked as well as CT and better than MRI to visualize ACTH-secreting tumors, and H-OCT identified a tumor that was not seen with any other modality in one patient. Importantly, L-OCT indicated involvement of lymph nodes that was missed by CT, MRI, and FDG-PET in two patients (no. 2 and 3). In another patient (no. 9), L-OCT identified two abdominal lesions missed by CT, MRI, and FDG-PET, one of which was the primary ileal neuroendocrine tumor. Thus, L-OCT appears to complement CT and MRI, providing additional diagnostic accuracy. Its lack of false positive results in the current study implies that this problem may not be as great as previously suggested and also shows how its specificity can aid in the interpretation of anatomical imaging findings.
As OCT uptake is dependent on the presence of somatostatin receptors, we hypothesized that a larger dose of radionuclide might improve its absolute uptake by small or somatostatin receptor-poor tumors. This hypothesis was supported by a report in which 48% of patients receiving H-OCT for therapy had lesions detected on their posttherapy scans that were not seen on their diagnostic scans (18). In the same report, posttherapy scans provided better count statistics and enabled delayed imaging with improved background clearance (18). More recently, Verhaegen et al. (48) reported a number of patients in whom posttherapy scans (200–270 mCi) also showed more lesions than diagnostic scans, although in some patients, fewer lesions were seen on the posttherapy planar imaging scans. Although the number of lesions detected after very high doses was not significantly different from that using diagnostic doses, posttherapy scans revealed some previously undetected lesions (48).
Lending further credence to the rationale of H-OCT, studies in the literature have not shown reduced imaging quality at octreotide mass doses up to 120 µg (49). Consequently, we did not anticipate that the increased mass dose of octreotide used for H-OCT compared with L-OCT (30 vs. 10 µg) would lead to saturation of tumor receptor sites.
In our patients, eight of nine patients with negative initial L-OCT had subsequent H-OCT that was also negative. However, one patient with a negative L-OCT study (and negative CT/MRI) had a positive H-OCT study that led to resection of an intrathoracic, ACTH-secreting tumor. Thus, H-OCT may be justified in the evaluation of patients with ectopic ACTH secretion when CT, MRI, and L-OCT are negative.
This study is limited by its small size, a result of the rarity of EAS. Thus, we studied only pulmonary carcinoids, neuroendocrine tumors, small cell lung cancer, pheochromocytoma, and olfactory esthesioneuroblastoma. Because of the small sample size, differences in sensitivity of the various modalities may not represent the results for all patients with EAS. Because other tumor types or larger groups may have different rates of detection, there is a need for studies of more patients.
Recently, other groups have published promising results for the detection of neuroendocrine tumors using other PET ligands, specifically [11C]5-hydroxytryptophan ([11C]5-HTP), 5-[11C]dihydroxyphenyl-L-alanine ([11C]DOPA), [11C]harmine, and 6-[18F]fluorodopamine (50, 51, 52, 53, 54, 55). Neuroendocrine tumors, such as foregut carcinoids, have been classified as APUDomas based on the demonstration of amine precursor uptake and decarboxylation. In particular, tryptophan is taken up and 5-hydroxylated to 5-HTP. Eriksson and colleagues (50) demonstrated that [11C]5-HTP PET identified tumors in three of five patients with ACTH-producing foregut carcinoids. In two other studies, including 22 patients with fore-, mid-, and hindgut carcinoids and three with endocrine pancreatic tumors, all tumors had radionuclide uptake after [11C]5-HTP-PET (51, 53). In another study of patients with pancreatic endocrine tumors, 11 of 16 patients had tumor uptake after [11C]DOPA-PET, and the images after [11C]5-HTP-PET correlated well with those after [11C]DOPA-PET (52). PET with the ligand [11C]harmine visualized all neuroendocrine gastroenteropancreatic tumors in 11 patients in another study (54). Norepinephrine transporter systems are present in medullary thyroid cancer, and a metastatic medullary thyroid cancer lesion was detected with 6-[18F]fluorodopamine (55). Thus, all of these ligands deserve further investigation in patients with ectopic ACTH secretion.
In conclusion, FDG-PET is inferior to CT/MRI and does not detect additional ACTH-secreting tumors causing CS. Because hyperplastic adrenal glands may show FDG-PET and/or OCT uptake, an adrenal ACTH-secreting lesion may be obscured. Nevertheless, in this study L-OCT detected ectopic ACTH-secreting tumors as well as CT and better than MRI and seems to be a useful complementary modality to them. Moreover, when CT/MRI results are questionable or equivocal, a positive L-OCT scan might guide the clinician to review those images at a specific location or to repeat them with thinner sections to correctly localize the ACTH-secreting lesion. Finally, H-OCT may be useful in patients with EAS in whom CT, MRI, and L-OCT are negative, because it may occasionally localize the tumor. These modalities are complementary: a single positive study may represent a falsely positive result, whereas the presence of more than one positive study may confirm a true lesion. Thus, we recommend that CT, MRI, and L-OCT be used to screen for tumors. To reduce costs, initial studies might be performed with only two modalities, followed by the third if a tumor is not identified clearly. H-OCT may be performed only when other imaging modalities fail to identify the ACTH-secreting tumor.
|
Footnotes |
|---|
Received October 17, 2003.
Accepted February 9, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Boscaro and G. Arnaldi Approach to the Patient with Possible Cushing's Syndrome J. Clin. Endocrinol. Metab., September 1, 2009; 94(9): 3121 - 3131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kauhanen, M. Seppanen, J. Ovaska, H. Minn, J. Bergman, P. Korsoff, P. Salmela, J. Saltevo, T. Sane, M. Valimaki, et al. The clinical value of [18F]fluoro-dihydroxyphenylalanine positron emission tomography in primary diagnosis, staging, and restaging of neuroendocrine tumors Endocr. Relat. Cancer, March 1, 2009; 16(1): 255 - 265. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J R Meinardi, B H R Wolffenbuttel, and R P F Dullaart Cyclic Cushing's syndrome: a clinical challenge Eur. J. Endocrinol., September 1, 2007; 157(3): 245 - 254. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Salgado, M. C. B V. Fragoso, M. Knoepfelmacher, M. C. Machado, S. Domenice, M. A. A. Pereira, and B. B. de Mendonca Ectopic ACTH syndrome: our experience with 25 cases. Eur. J. Endocrinol., November 1, 2006; 155(5): 725 - 733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Findling and H. Raff Cushing's Syndrome: Important Issues in Diagnosis and Management J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3746 - 3753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Waintrop, F. Tenenbaum, J. Coulot, O. Vignaux, C. Blanchet, P. Legmann, B. Richard, and X. Bertagna IMAGE IN ENDOCRINOLOGY: Localization of an Occult Adrenocorticotropin-Secreting Carcinoid Tumor with Somatostatin Receptor Scintigraphy Using Single-Photon Emission Computed Tomography/Computed Tomographic Image Scanning J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3271 - 3272. [Full Text] [PDF]
|
|
|
|
|
|
I. Ilias, D. J. Torpy, K. Pacak, N. Mullen, R. A. Wesley, and L. K. Nieman Cushing's Syndrome Due to Ectopic Corticotropin Secretion: Twenty Years' Experience at the National Institutes of Health J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4955 - 4962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Orlefors, A. Sundin, U. Garske, C. Juhlin, K. Oberg, B. Skogseid, B. Langstrom, M. Bergstrom, and B. Eriksson Whole-Body 11C-5-Hydroxytryptophan Positron Emission Tomography as a Universal Imaging Technique for Neuroendocrine Tumors: Comparison with Somatostatin Receptor Scintigraphy and Computed Tomography J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3392 - 3400. [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 |
