| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Department of Endocrinology and Medical Sciences and Center of Excellence for Biomedical Research (E.R., F.B., A.R., F.M., D.F.), and Departments of Pathology (P.D., J.-L.R.) and Neurosurgery (G.Z., R.S.), San Martino Hospital University of Genova, 16132 Genova, Italy; Interactions Cellulaires Neuroendocriniennes (A.S., P.J.), Unité Mixte de Recherche, Centre National de la Recherche Scientifique, Faculté de Médecine Nord, Institut Fédératif Jean Roche, 13916 Marseille Cedex, France; and Biomeasure Inc./IPSEN (M.D.C.), Milford, Massachusetts 01757
Address all correspondence and requests for reprints to: Diego Ferone, M.D., Ph.D., Department of Endocrinology and Metabolism, University of Genova, Viale Benedetto XV, 6, 16132 Genova, Italy. E-mail: ferone{at}unige.it.
| Abstract |
|---|
|
|
|---|
Objective: The objective of the study was to investigate the dissociation between antiproliferative and antisecretive effects of SSA in an octreotide-resistant patient displaying dramatic tumor shrinkage during primary therapy with octreotide LAR.
Design and Setting: We characterized somatostatin and dopamine D2 receptor expression by immunohistochemistry and real-time RT-PCR. The effects of different receptor-selective, bispecific analogs, and chimeric somatostatin/dopamine compounds on GH secretion and cell proliferation in primary cell cultures of the tumor were assessed.
Results: The expression of SS receptor subtypes (sst)5 and D2 receptor was higher, compared with the other receptor subtypes. GH inhibition by SS-14 and the two chimeric somatostatin/dopamine compounds was scant but greater than subtype-selective and sst2/sst5 bispecific agonists. Conversely, cell growth was potently inhibited by all test substances. However, SS-14, sst2/sst5 bispecific agonist, and chimeric molecules were more potent than the other compounds.
Conclusions: The significant antiproliferative effect of octreotide seems to be related to the higher expression of sst5 and the negligible antihormonal effect to the lower expression of sst2. However, activation of multiple receptors by new analogs may produce better control of tumor cell activities. The dissociation between antisecretive and antiproliferative effects observed in vivo and in vitro confirms that SSAs may induce tumor shrinkage despite the lack of effect on GH secretion.
| Introduction |
|---|
|
|
|---|
SS exerts its inhibitory effects via five SS receptor (SSR) subtypes (sst15) (5). The currently available SSAs (octreotide, lanreotide) bind with high affinity to sst2 and relatively lower affinity to sst3 and sst5. Most of the in vivo effects of SSAs are mediated by the activation of these SSR subtypes. In vitro, it has been shown that SSAs activate distinct signaling pathways to exert their antisecretory and antitumoral effects (6, 7, 8). Moreover, the in vitro inhibition of cell proliferation may occur independently of the effects on GH secretion, although the mechanisms and significance of this dissociation are still unknown (8).
Recently a functional association between sst2 and sst5 in inhibiting GH secretion from human fetal pituitary and adenoma cells has been suggested (9), and new bispecific analogs, with high affinity for both sst2 and sst5, or compounds with a broader binding profile have been developed. Because the inhibitory activity of these experimental compounds is particularly evident in SSA-resistant GH-secreting tumors, which may display a heterogeneous expression of sst2 and sst5, these new drugs may achieve better control of GH hypersecretion in selected acromegalic patients (10, 11).
Here we describe the case of a young acromegalic patient who experienced dramatic pituitary tumor shrinkage after several months of therapy with octreotide LAR, although GH levels were unchanged during the treatment. To explore the potential mechanisms causing the dissociation observed in vivo, tumor-derived cells were tested in vitro with new receptor subtype-selective, bispecific, and chimeric SS/dopamine compounds.
| Patient and Methods |
|---|
|
|
|---|
A 33-yr-old man had a 4-yr history of hand and feet enlargement and facial disfigurement as well as osteoarticular pain, fatigue, and headache. Hormonal evaluation showed elevated GH (4.1 ± 0.5 ng/ml, mean ± SEM of three samples) not suppressible after oral glucose load, elevated IGF-I (791 ng/ml), and normal prolactin and
-subunit levels. The remaining pituitary function as well as visual perimetry was normal, whereas the initial magnetic resonance imaging (MRI) showed a large intra- and extrasellar pituitary adenoma with optic chiasm and right carotid siphon involvement. A left deviation of the pituitary stalk was also evident (Fig. 1A
).
|
|
Hormone assays
Serum GH levels were determined by means of a chemiluminescent immunoradiometric assay (Immulite; Diagnostic Products Corp., Los Angeles, CA), whereas IGF-I was measured by RIA using immunochemicals and tracer provided by Biosource (Nivelles, Belgium), as previously reported in detail (12).
MRI and visual perimetry
Tumor mass was evaluated by MRI performed as previously reported (13).
Visual fields were evaluated by means of a Humphrey perimeter (750 II VFA model, program central 302, full threshold strategy). Particularly, indexes of automatic static perimetry, namely mean defect and pattern SD, were evaluated. The ophthalmological examination was performed at baseline and monthly for the first 6 months of treatment and then quarterly.
Samples and products
An informed consent was obtained from the patient for the use of the surgical specimens. A large portion of the pituitary tumor sample obtained during transsphenoidal surgery was washed in buffer and immediately divided into two pieces. The first was processed for RNA extraction and the other was mechanically dispersed in for in vitro functional experiments, yielding more than 9 x 106 viable cells. Moreover, 12 paraffin-embedded sections from the same adenoma sample were obtained from the Department of Pathology for immunohistochemistry.
BIM compounds were provided by Biomeasure, Inc. (Milford, MA), whereas SS-14 was purchased from Bachem Inc. (Hanover, Germany). BIM-23023 is an sst2-selective compound, BIM-23206 is an sst5-selective compound, BIM-53097 is a D2R-selective compound, BIM-23244 is an sst2/sst5 bispecific analog, whereas BIM-23A387 and BIM-23A761 are chimeric molecules that combine structural elements of both SS and dopamine and retain affinity for both the sst2 and D2R and sst2, sst5 and D2R, respectively. BIM-23023, BIM-23206, and BIM-53097, which are subtype-selective analogs, also served as control peptides for chimeras. The IC50 values for BIM compounds are reported in Table 1
.
|
Immunohistochemistry for SSR and D2R was performed on 5-µm paraffin-embedded sections as previously reported (14, 15). Negative controls for immunohistochemistry included omission of the primary antibody and preabsorption of the antibodies with the respective immunizing receptor peptides (at a concentration of 100 nM). A tissue was considered positive when the immunostaining was abolished by preabsorption of the antibody with the respective peptide antigen. Immunostaining for GH and Ki67 was performed on sequential sections using specific antibodies at the standard dilution.
One cubic millimeter of tumor tissue was fixed overnight in buffered 2.5% glutaraldehyde solution, without Ca and Mg (pH 7.2) and subsequently postfixed in buffered 2% osmium for 1 h. The sample was dehydrated in acetone and infiltrated in epoxides resin. Semithin (1-µm thick) slides were stained for toluidin blue. Subsequently ultrathin slides were mounted to copper one-hole grids (200 mesh). For electron microscopy an EM 900 electron microscope (Zeiss, Jena, Germany) was used.
Cell cultures
The preparation and culture methods for isolated pituitary tumor cells have been published previously (11). To avoid contamination with fibroblasts, the cells, after enzymatic digestion, were treated with antihuman fibroblasts antibodies coated on magnetic beads (Dianova GmbH, Hamburg, Germany). In this study, cells were plated at a density of 104 and 105 cells/well, respectively, for hormone secretion and 3H-thymidine incorporation. The experiments were performed at d 2 of culture. All test substances were added at 108 M concentrations and refreshed after 24 h. Hormone release was measured in the media after a 48-h incubation period. Proliferation was determined by measuring thymidine incorporation after adding 1 mCi [methyl-3H-]thymidine (Amersham, Milan, Italy) during the last 24-h period of incubation, as previously described (16).
Real-time PCR
(Taq Man) assay for quantitative measurement of the SSR and D2R mRNAs as been presented in detail elsewhere (17). The results are normalized to the ß-glucuronidase (ß-Gus) mRNA levels measured in the same reaction. In GH-secreting adenomas sensitive to octreotide, the mean mRNA levels for sst2, sst5, and D2R were 1.7 ± 0.3, 4.8 ± 0.7, and 4.0 ± 0.5 copy/copy ß-Gus, respectively.
Statistical analysis
Statistical analysis was performed by parametric ANOVA for unpaired samples and followed by t test. P < 0.05 was considered significant. Results are expressed as the mean ± SEM. To minimize variation among different experiments, the results are expressed as relative variation from untreated control value.
| Results |
|---|
|
|
|---|
By real-time RT-PCR, the expression of sst5 (4.6 sst5 per ß-Gus copy mRNA) was higher than that of all SSR subtypes (sst4 was undetectable). In fact, among the other subtypes, the expression of sst2 (2.4 sst2 per ß-Gus) was relatively higher than that of sst1 (0.1 sst1 per ß-Gus), whereas the expression of sst3 was very low (0.01 sst3 per ß-Gus). Conversely, D2R was the second highest receptor expressed (3.5 D2R per ß-Gus copy mRNA) in this tumor.
Histology and immunohistochemistry
Histology revealed an acidophilic adenoma without signs of karyokinesis, necrosis, or fibrosis (Fig. 3A
), whereas electron microscopy showed a large number of dense granules in the tumor cells (Fig. 3G
). GH immunostaining was strongly positive, whereas Ki67 was less than 1% (Fig. 3
, B and C). The remaining pituitary hormones were not detected (data not shown). The tumor has been classified as a pure GH-secreting densely granulated adenoma. Diffuse sst2A, sst5, and D2R immunoreactivity was detected; however, sst2A was less intense, compared with sst5 and D2R immunoreactivity (Fig. 3
, DF). Conversely, sst1 and sst3 staining was weak and limited to scattered cells. Because sst4 mRNA was undetectable, sst4 immunohistochemistry was not performed.
|
Prolactin was undetectable in the cell cultures, whereas GH secretion was 18.4 ± 2.7 µg/liter. All test substances, except BIM-53097, significantly inhibited GH secretion, compared with control (P < 0.05). However, the inhibitory effects were scant (ranging from 4.6 ± 0.6 to 22.1 ± 1.4,
% vs. control). SS-14 and the two chimeric compounds, BIM-23A387 (sst2/D2R) and BIM-23A761 (sst2/sst5/D2R), were more potent in inhibiting GH release, compared with the other test substances, including the bispecific analog BIM-23244 (sst2/sst5) and the subtype-selective analogs tested in combination (controls for the chimeras). No significant differences have been found between SS-14 and the two chimeras, BIM-23A387 and BIM-23A761, in inhibiting GH secretion. Similarly, there were no differences in the inhibition of GH secretion between the BIM-23023 (sst2) and BIM-23206 (sst5), whereas BIM-23023 (sst2) was more potent than BIM-53097 (D2R). The inhibition of GH secretion by the bispecific analog BIM-23244 (sst2/sst5) was more potent, compared with BIM-23206 (sst5), but not with BIM-23023 (sst2). The effects of the test substances on GH secretion are illustrated in Fig. 4A
with the respective P values.
|
The basal 3H-thymidine incorporation activity was approximately 800 cpm/105 cells. All test substances significantly inhibited 3H-thymidine incorporation, compared with control (P < 0.05), and were more potent in inhibiting cell proliferation than GH secretion (ranging from 14.6 ± 0.4 to 33.5 ± 2.0,
% vs. control). However, SS-14 was more potent in inhibiting 3H-thymidine incorporation, compared with BIM-23023 (sst2) and BIM-53097 (D2R), tested alone and in combination. Conversely, there were no differences between SS-14 and all the other test substances, including both chimeras BIM-23A387 (sst2/D2R) and BIM-23A761 (sst2/sst5/D2R) as well as the bispecific analog BIM-23244 (sst2/sst5). However, BIM-23A761 (sst2/sst5/D2R) was more potent than BIM-23A387 (sst2/D2R). Both sst-selective compounds BIM-23023 and BIM-23206, tested alone or in combination, as well as the bispecific analog BIM-23244 (sst2/sst5) were more potent than BIM-53097 (D2R). The bispecific analog BIM-23244 (sst2/sst5) was more potent than both BIM-23023 (sst2) and BIM-23206 (sst5), tested alone. The combination of BIM-23023 (sst2) and BIM-23206 (sst5) was more potent than only BIM-23023 (sst2). The effects of the test substances on 3H-thymidine incorporation are illustrated in Fig. 4B
with the respective P values.
| Discussion |
|---|
|
|
|---|
Initial evidence in GH secreting adenomas suggested that resistance to octreotide might be associated with low expression of sst2 mRNA (11, 22, 23). However, other authors have demonstrated in vitro that, independent of the density of a given receptor, a combined activity of sst2 and sst5 confers higher efficacy in suppressing GH from tumor cells (9). Moreover, most of the tumors that are partially responsive to octreotide may better respond in vitro to a bispecific sst2/sst5 (10, 24) or chimeric sst2/D2R and sst2/sst5/D2R ligands (25). The enhanced activity of these compounds may be related to the induction of receptor dimerization or interaction of these ligands with preexistent receptor dimers (26, 27). Interestingly, it has been shown that in adenomas expressing different profiles of sst2, sst5, and D2R, treatment with the sst2-selective agonist or the sst2/sst5 dual agonist significantly inhibited GH secretion, whereas treatment with the sst5-selective agonist did not modify the secretory activity of the adenoma cells when D2R was coexpressed (28). Indeed, the role of sst2 and sst5 in modulating cell proliferation and hormone secretion in pituitary adenomas is not completely understood. In fact, in in vitro cultures of human nonfunctioning pituitary adenomas, sst2-selective agonists inhibited hormone secretion without affecting cell viability, whereas sst5-selective agonists did not influence secretion but, unexpectedly, promoted cell viability (29). Conversely, sst5-selective agonists have been shown to inhibit cell proliferation in GH-secreting pituitary adenomas (8). Therefore, SSAs may differentially affect hormone secretion and cell proliferation, depending on not only the specific activated receptor subtype but also the specific target cell (29).
In our case the expression of sst5 was higher than all other receptors, including D2R and sst2. This patient responded to octreotide, experiencing a significant tumor shrinkage, whereas GH levels were almost unchanged after more than 5 months of treatment, although IGF-I declined. This apparent discrepancy might be explained by the fact that in acromegaly when GH levels are greater than 4 but less than 10 µg/liter, even mild variation of GH levels might be accompanied by significant plasma IGF-I changes (30, 31). In addition, a direct effect of octreotide on IGF-I at the level of the liver cannot be ruled out (32). Moreover, the dissociation between the antihormonal and the antiproliferative effects of octreotide was confirmed during the in vitro experiments, supporting the hypothesis that sst2 and sst5 may play different roles in GH secreting adenomas, depending on the level of expression. In particular, the reduced expression of sst2, compared with sst5, may determine a sort of resistance to the GH-lowering activity of SSAs. Conversely, the increased expression of sst5 may account for the significant tumor shrinkage. In fact, our in vitro data showed that all testing substances interacting with sst5 displayed a similar potency in inhibiting cell proliferation and were more effective than SSAs, mainly interacting with sst2. Although an antiangiogenic activity of octreotide may have contributed to the tumor shrinkage, these in vitro findings support the direct effect of octreotide on tumor cells.
Recently, Casarini et al. (33) reported a case of dissociation between hormone control and tumor growth, suggesting that the high level of sst3 might explain this phenomenon (via sst3-mediated apoptosis). Conversely, in our case sst3 mRNA was very low, and the antiproliferative activity of octreotide could have been mediated by sst5 (via cell cycle arrest).
The use of SS/dopamine chimeric ligands may overcome the resistance to SSAs. In fact, when tested in vitro, both sst2/D2R and sst2/sst5/D2R chimeric compounds were significantly more potent in inhibiting GH secretion than SSR subtype-selective agonists. Conversely, these chimeric compounds were not significantly different from SS; however, the sst2/sst5/D2R chimera was more potent than the sst2/D2R chimera in inhibition of cell proliferation, suggesting an important role of the sst5 activation. Therefore, patients resistant to conventional treatments could benefit from new chimeric compounds, depending on their tumor receptor profile. This observation emphasizes the importance of SSR subtype characterization in pituitary adenomas, which may allow us to target tumors displaying specific receptor patterns with proper drugs.
In conclusion, this case confirms that biochemical response in acromegaly is a poor predictor of tumor size reduction. Moreover, the in vitro study proves that a dissociation between antisecretive and antiproliferative effects of SSAs may occur and that mechanisms other than those involved in controlling hormone secretion may operate in controlling tumor growth. The new insights in pathophysiology of SSRs indicate that different receptor profiles on GH-secreting adenomas may affect the clinical response, depending on the specific subtypes expressed.
| Footnotes |
|---|
Disclosure Statement: E.R., F.B., A.R., F.M., D.F., P.D., J.-L.R., G.Z., R.S., A.S., and P.J. have nothing to declare. M.D.C. is employed by Biomeasure Inc./IPSEN.
First Published Online February 20, 2007
Abbreviations: ß-Gus, ß-Glucuronidase; D2R, dopamine 2 receptor; MRI, magnetic resonance imaging; SS, somatostatin; SSA, SS analog; SSR, SS receptor; sst, SSR subtype.
Received September 22, 2006.
Accepted February 13, 2007.
| References |
|---|
|
|
|---|
-subunit and chromogranin a secretion and on cell viability in human nonfunctioning pituitary adenomas in vitro. J Clin Endocrinol Metab 89:51815188This article has been cited by other articles:
![]() |
D. Ferone, W. W. de Herder, R. Pivonello, J. M. Kros, P. M. van Koetsveld, T. de Jong, F. Minuto, A. Colao, S. W. J. Lamberts, and L. J. Hofland Correlation of in Vitro and in Vivo Somatotropic Adenoma Responsiveness to Somatostatin Analogs and Dopamine Agonists with Immunohistochemical Evaluation of Somatostatin and Dopamine Receptors and Electron Microscopy J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1412 - 1417. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. F Taboada, R. M Luque, L. V. Neto, E. d. O Machado, B. C Sbaffi, R. C Domingues, J. B Marcondes, L. M C Chimelli, R. Fontes, P. Niemeyer, et al. Quantitative analysis of somatostatin receptor subtypes (1-5) gene expression levels in somatotropinomas and correlation to in vivo hormonal and tumor volume responses to treatment with octreotide LAR Eur. J. Endocrinol., March 1, 2008; 158(3): 295 - 303. [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 |