| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Medicine (S.J.H., S.F., P.H.B.), Royal Victoria Infirmary, Newcastle upon Tyne, NE1 4LP; Department of Medicine (P.E.H.), Kings College Medical School, London, SE5 9TJ; Department of Medicine (W.F.K.), Middlesbrough General Hospital, Middlesbrough, TS5 5AZ; and University Department of Pathology (A.M.M.), Glasgow Royal Infirmary, Glasgow, G4 OSF
Address correspondence and requests for reprints to: S.J. Hurel, Department of Medicine, Royal Victoria Infirmary, Queen Victoria Road, Farmington Place, Newcastle-Upon-Tyne, United Kingdom NE1 4LP.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The molecular basis for such aggressive tumors is uncertain. A variety
of mutations have been associated with pituitary tumors (4). The best
characterized genetic change is the gsp mutation, which
occurs in 3440% of somatotroph adenomas (5, 6), although
gip mutations have also subsequently been described (7, 8).
Mutations in the ras oncogene are relatively rare. A
mutation in H-ras has been described in a highly invasive
prolactinoma, and H-ras mutations have been described in
metastatic deposits of pituitary carcinoma, including a PRL-secreting
carcinoma, but not in the primary tumors (9, 10). A point mutation in
the
isoform of protein kinase C has been described in highly
invasive pituitary tumors (11). The p53 tumor suppressor gene mutation
is the commonest specific gene alteration associated with major human
neoplasia, but abnormal p53 expression appears to occur infrequently in
human pituitary adenomas (10, 12) other than corticotrophic adenomas
(13). c-myc is overexpressed in a wide variety of pituitary
adenomas, and overexpression of c-fos, although occurring
much less frequently, may have a pathogenic role in the development of
corticotroph adenomas (14). Loss of heterozygosity at the
retinoblastoma susceptibility gene (Rb) has been reported in malignant
and invasive prolactinomas (15). It remains unknown, however, whether
any specific mutation or combination of mutations predispose tumors to
develop a particularly aggressive phenotype. We report a case of
metastatic prolactinoma, describe the clinical course, treatments
employed, and subsequent immunocytochemical analysis.
| Case History |
|---|
|
|
|---|
A magnetic resonance imaging (MRI) scan demonstrated an extensive tumor
invading the ethmoidal sinuses, left orbit, temporal fossa, pons,
maxillary antrum, pterygo-palatine fossa, jugular nodal chain, and
thoracic bodies (Fig. 1
). An In-111
pentetreotide scan showed abnormal uptake in the anterior of skull and
the angle of the jaw (Fig. 2
). The PRL
estimation of 114,000 mU/L was confirmed. Cabergoline was commenced and
increased to 8 mg/week in divided doses coincident with a moderate
decrease (30%) in her PRL level to 74,000 mU/L. Interestingly, when
first treated at this high dose, she developed a transient paranoid
delusional state with some features of dermatozoonosis, which resolved
within 2 days of decreasing the dose. Octreotide was commenced via a
continuous sc infusion and maintained at 800 µg/day. Within 24 h
her facial pain resolved completely. She was given three cycles of
carboplatin (440 mg infusion over 30 min) and etoposide (100 mg/day
orally for 5 days). Serial PRL estimations and MRI scans showed no
clear benefit from the treatment. Six months after commencing
chemotherapy, following a severe headache presumably caused by tumor
infarction or hemorrhage, she died peacefully at home. Autopsy
examination was not performed.
|
|
| Results |
|---|
|
|
|---|
Paraffin sections obtained from the first hypophysectomy and the
lymph node biopsy were immunostained for a range of known
proto-oncogenes. Antibodies to RB protein, c-jun, WAF-1,
p53, and c-myc were purchased from Novocastro
Laboratories (Newcastle, UK). RB protein, c-jun, and
WAF-1 were demonstrated using a peroxidase-labeled streptavidin
technique, and c-myc was demonstrated using an indirect
peroxidase method. Sections underwent microwave pretreatment for RB
protein, WAF-1, and c-myc and trypsinization for
c-jun. Controls were omission of primary antibody as
negative and inclusion of known positives. The results are shown in
Fig. 3
. Widespread nuclear staining was
demonstrated for p53 tumor suppressor gene (5075%), WAF-1
(4050%), Rb protein (5075%), and c-myc (4060%)
in both tissues. There was no positive staining for
c-erb-2 in any tissue. The original tumor stained
negative for c-jun, but the secondaries were positive.
Few cells in the metastatic tumor showed positivity with MIB-1
antibody, which demonstrates ki-67, a nuclear protein expressed during
the cell cycle. The lymphoid tissue showed the expected pattern of
staining.
|
| Discussion |
|---|
|
|
|---|
The specific DA2 antagonist cabergoline appeared to have no advantage over bromocriptine; although serum PRL estimations decreased by 30%, which may represent dedifferentiation, there was no decrease in tumor size. Cabergoline did, however, precipitate an acute paranoid crises in the form of dermatozoonosis, i.e. a sensation of animals crawling over the skin. This interesting monosymptomatic hypochondriasis was first described by Ekbom (16) as a consequence of presenile neuronal degeneration and was subsequently seen in association with high doses of bromocriptine (5). In this latter case, as in ours, the psychotic symptoms resolved on reduction in the dose of dopamine agonist.
Despite the presence of somatostatin receptors, as demonstrated by the positive pentetreotide scan, octreotide had no significant effect on disease progression. This agrees with previous observations that tumor growth suppression may not occur with octreotide despite a high level of uptake when using radiolabeled ligand (17). However, octreotide did have a rather favorable effect on the patients facial pain. Somatostatin messenger RNA has been identified in high density within the pain-modulating caudal region of the spinal trigeminal nucleus and in the substantia gelatinosa of the dorsal horns within both the cervical and lumbar regions (18). This has led to the suggestion that spinal injections of somatostatin analogs may be of benefit in the management of intractable pain of malignancy (19). In humans, intraarticular injection of somatostatin has been shown to reduce painful osteoarthritis of the knee (20). We have seen a similar phenomenon in a patient with a malignant phaeochromocytoma that also had large uptake of octreotide. There was no reduction in tumor size, but a dramatic analgesic effect occurred with octreotide therapy. Furthermore, octreotide is well recognized as having a rapid analgesic effect on headache in acromegalic patients. Despite the lack of effect of octreotide, pentetreotide scanning may still have a role in determining the extent of local and metastatic spread in patients with prolactinomas.
Walker et al. (1) treated three patients with procarbazine, etoposide, and lomustine, resulting in dramatic reductions in serum PRL concentration suggestive of either response to treatment or dedifferentiation with tumor progression. There was, however, no clear benefit in terms of life expectancy. After consultation with our local oncologists, it was felt that carboplatin and etoposide offered the best chance of success with the minimum of adverse effects in our patient. It seems unlikely, however, that they had any real benefit on the progression of disease.
We demonstrated immunocytochemical staining for c-myc, p53, WAF-1, and Rb protein in both the primary and secondary tumor. The original pituitary tumor did not stain for c-jun, but the recurrent tumor was positive. We have subsequently carried out further studies on genomic DNA but have identified no mutations other than loss of heterozygosity within the Rb region as described previously (15). The underlying oncogenic mechanisms occurring in pituitary neoplasia are poorly understood. X-chromosome inactivation has demonstrated that most pituitary tumors are monoclonal, implying the occurrence of an initial genetic event (21). The best-characterized genetic change occurring in pituitary adenomas is in the gsp oncogene (5, 6), although no mutations were present in our case. Positive nuclear staining for p53 cannot normally be detected because of its short half-life. The presence of WAF-1, however, suggests that there is functional p53 present in these cells. It may be that this reflects mechanisms of p53 inactivation other than mutation, resulting in loss of control of cell proliferation and apoptosis function. It is interesting that c-jun activating protein-1 protein was increased in the tumor metastases but not in the primary tumor. These data suggest that the activating protein-1 proteins c-myc and c-jun may have had a specific role in progression to metastatic spread similar to that suggested for H-ras (10). Pei et al. (15) reported allelic loss in the region of the Rb gene in 13 aggressive pituitary tumors, seven of which were metastatic. Immunohistochemical analysis did, however, reveal the presence of Rb protein (which may not reflect functional protein), and it appears that loss of heterozygosity in the Rb region of chromosome 13q may provide a marker for other tumor suppression factor(s) that are of central importance in the development of metastatic pituitary tumors. Overall, the data point to multiple potential oncogenic mutations in the development of this tumor. Increased numbers of mutations, perhaps secondary to radiotherapy (22, 23), in particular loss of heterozygosity, appear to be associated with increasingly aggressive tumor behavior. It may be that particular mutations, e.g. ras or jun are associated with malignant transformation.
In summary, the management of pituitary carcinoma remains very unsatisfactory. Specific oncogenic mutations may be associated with the progression of the disease, and in the future gene therapy may be possible to correct for the loss of tumor suppressor function.
Received April 10, 1997.
Revised May 28, 1997.
Accepted June 3, 1997.
| References |
|---|
|
|
|---|
chain
of Gs and stimulate adenylyl cyclase in human pituitary tumours. Nature. 340:692696.[CrossRef][Medline]
and
Gi2
mutations in clinically non-functioning pituitary
tumours. Clin Endocrinol (Oxf). 41:815820.[Medline]
-protein kinase-C. J Clin Endocrinol Metab. 77:11251129.[Abstract]
This article has been cited by other articles:
![]() |
M. Kars, F. Roelfsema, J. A Romijn, and A. M Pereira Malignant prolactinoma: case report and review of the literature. Eur. J. Endocrinol., October 1, 2006; 155(4): 523 - 534. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. Gillam, M. E. Molitch, G. Lombardi, and A. Colao Advances in the Treatment of Prolactinomas Endocr. Rev., August 1, 2006; 27(5): 485 - 534. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. A. Kaltsas, P. Nomikos, G. Kontogeorgos, M. Buchfelder, and A. B. Grossman Diagnosis and Management of Pituitary Carcinomas J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 3089 - 3099. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. N. Farrow, A. P. Bradford, J. J. Tentler, and A. Gutierrez-Hartmann Structural and Functional Analysis of the Differential Effects of c-Jun and v-Jun on Prolactin Gene Expression Mol. Endocrinol., October 1, 2004; 18(10): 2479 - 2490. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Petrossians, W. De Herder, D. Kwekkeboom, G. Lamberigts, A. Stevenaert, and A. Beckers Malignant Prolactinoma Discovered by D2 Receptor Imaging J. Clin. Endocrinol. Metab., January 1, 2000; 85(1): 398 - 401. [Full Text] |
||||
![]() |
Y. Cheng, I. Zhizhin, R. L. Perlman, and D. Mangoura Prolactin-induced Cell Proliferation in PC12 Cells Depends on JNK but Not ERK Activation J. Biol. Chem., July 21, 2000; 275(30): 23326 - 23332. [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 |