A morphological study was carried out on pituitary adenomas removed from 13 normoprolactinemic and 9 hyperprolactinemic acromegalic patients whose hormonal dynamics had been carefully investigated. Double immunocytochemical labeling with the protein-A-gold electron microscopic technique was used to detect the presence of GH and PRL in the adenomas. Two morphological patterns were found; 11 adenomas contained cells positive only for GH, and 11 contained a variable proportion (from 10-98%) of cells positive for PRL. The great majority of cells positive for PRL also were positive for GH and so were actually mammosomatotrophic cells. Among the normoprolactinemic patients, no cells containing PRL were found in the tumors from 8 patients, and 10-26% of the cells of the tumors of the other 5 patients contained PRL. Two thirds of the hyperprolactinemic patients had tumors containing mammosomatotrophs (18-80%) with or without the concomitant presence of mammotrophs (0-18%). A positive correlation was found between the serum PRL levels and the percentage of mammosomatotrophs. No significant differences in GH secretory responses to TRH, dopamine, GHRH, and SRIH were found between patients having tumors with or without cells positive for PRL. We conclude that 1) the frequency of mammosomatotrophs in adenomas from acromegalic patients is higher than that previously estimated using different immunocytochemical methods; and 2) serum GH responses to TRH and dopamine are not strictly related to the presence of mammosomatotrophs and/or mammotrophs within the tumor.
It is debated if acromegalic patients have an increased risk to develop malignancies. The aim of the present study was to assess the standardized incidence ratios (SIRs) of different types of cancer in acromegaly on a large series of acromegalic patients managed in the somatostatin analogs era. It was evaluated the incidence of cancer in an Italian nationwide multicenter cohort study of 1512 acromegalic patients, 624 men and 888 women, mean age at diagnosis 45 ± 13 years, followed up for a mean of 10 years (12573 person-years) in respect to the general Italian population. Cancer was diagnosed in 124 patients, 72 women and 52 men. The SIRs for all cancers was significantly increased compared to the general Italian population (expected: 88, SIR 1.41; 95% CI, 1.18-1.68, P < 0.001). In the whole series, we found a significantly increased incidence of colorectal cancer (SIR 1.67; 95% CI, 1.07-2.58, P = 0.022), kidney cancer (SIR 2.87; 95% CI, 1.55-5.34, P < 0.001) and thyroid cancer (SIR 3.99; 95% CI, 2.32-6.87, P < 0.001). The exclusion of 11 cancers occurring before diagnosis of acromegaly (all in women) did not change remarkably the study outcome. In multivariate analysis, the factors significantly associated with an increased risk of malignancy were age and family history of cancer, with a non-significant trend for the estimated duration of acromegaly before diagnosis. In conclusion, we found evidence that acromegaly in Italy is associated with a moderate increase in cancer risk.
The influence of beta-adrenergic blockade by oral propranolol on the variability of GH responses to GHRH and on GH responsiveness to repeated GHRH administrations was investigated. Eight normal volunteers underwent three tests on three separate occasions. Each test consisted of two administrations of 80 micrograms GHRH at 2-h intervals without other medication (test 1) or combined with oral administration of 80 mg propranolol 90 min before the first (test 2) or the second GHRH injection (test 3). In test 1 GH levels increased significantly after the first, but not the second GHRH bolus (net incremental area under the curve [nAUC], mean +/- SD: 1453 +/- 974 and 178 +/- 309 micrograms.l-1.(120 min)-1, respectively). In test 2 basal GH secretion was not influenced by propranolol administration, whereas the GH response to the first GHRH injection was significantly greater than in test 1 (2327 +/- 1814 micrograms.l-1.(120 min)-1; p less than 0.05). However, individual subjects showed the same variability of GH response as in test 1. The GH response to the second GHRH bolus remained negligible. In test 3 administration of propranolol 90 min before the second GHRH bolus led to a clear GH increase (690 +/- 1002 micrograms.l-1.(120 min)-1), not significantly different from the GH response to the first bolus (1796 +/- 1375 micrograms.l-1.(120 min)-1). However, only 4 subjects showed a marked restoration of the GH responsiveness to the second GHRH administration. In conclusion, oral administration of propranolol is able to increase GH responsiveness to GHRH without changing the great individual variability. The response to a repeated GHRH stimulation is only partially restored by propranolol.
Abstract. Circulating GH consists of several molecular size species with different biological activity. A reduced sensitivity of some monoclonal antibodies towards high-molecular weight GH variants has been reported. The aim of the present work was to evaluate the molecular size species of circulating GH using Sephadex G-100 gel filtration chromatography in acromegalic patients and in normal subjects employing both RIA and an immunoradiometric assay for all GH determinations. In 6 normal subjects, studied under GHRH stimulation, little GH was 69.8±6% (mean ± sd ), big GH (44 kD) 26.4±6% and big-big GH (>80 kD) 2.8±4%, in IRMA, with a good correspondence with RIA results (70.8±8, 27.0±4, and 3.2±2%, respectively). In 13 untreated acromegalic patients, studied in basal conditions, the little form constituted 76.2±7%, the big form 18.3±4%, which is significantly lower than in normals (p<0.05), and the big-big form 5.5±7%. Similar results were obtained with RIA. A clear elevation of big-big GH (21% for both in IRMA, and 15.7 and 27.8% in RIA) was found in 2 patients with IGF-I levels lower than expected on the basis of mean GH concentrations. The study was extended to an additional acromegalic patient, previously operated and irradiated on, characterized by discrepant serum GH levels in RIA (4.6 μg/l), and in IRMA (1.4 μg/l), and by normal IGF-I levels. Serum GH showed a lack of parallelism to standard GH in RIA, but not in IRMA. RIA immunoreactivity was almost completely composed (92%) of a high molecular weight GH form (>90 kD), not recognized by IRMA. All IRMA immunoreactivity eluted with a K av corresponding to 19–50 kD. In conclusion: a. the three main molecular size isomers of serum GH are similarly recognized by IRMA and RIA methods in normal subjects. b. in acromegaly, both quantitative and qualitative modifications of the GH chromatographic profile may be present. In particular, increased amounts of big-big forms, whether or not recognized by monoclonal antibodies, have been observed. Their lower bioactivity, suggested by the normal or lower than expected IGF-I levels, can account for the discrepancy between serum GH levels and the clinical picture or IGF-I levels sometimes observed in acromegaly.
Somatic mutations in the alpha-chain (alpha s) of the stimulatory regulatory protein of adenylyl cyclase (Gs) causing constitutive activation of the enzyme have been identified in a subset of human GH-secreting pituitary adenomas. This study reports on the differences between acromegalic patients bearing tumors without (group 1; n = 51) or with (group 2; n = 29) this alteration. No difference in age, sex, clinical features, duration of the disease, or cure rate was observed between the two groups. By contrast, group 2 patients had higher basal GH levels than group 1. Moreover, a significant difference in sellar morphology was found; group 2 patients more frequently showed sellas of normal size (grade I) than group 1. Hypersecretory activity of group 2 tumors was also apparent at electron microscopy; contrary to those of group 1, cells of group 2 tumors were densely granulated and showed prominent rough endoplasmic reticulum and Golgi complex. With respect to group 1, group 2 patients were less responsive to GH-releasing hormone, while they were more sensitive to somatostatin- and dopamine-induced GH inhibition. These results suggest that patients with constitutively active adenylyl cyclase have hyperactive tumors; the sensitivity of these tumors to inhibitory agents (somatostatin and dopamine), possibly counteracting the expression of activating mutations, might explain the low rate of tumor growth.
The recent availability of a Tyr3-substituted octreotide (SDZ 204–090) for radioiodination has allowed somatostatin (SRIH) receptor binding to be studied in vivo, and receptor-positive tumors of different origins to be visualized with a γ-camera. This prompted us to investigate whether this compound could be used for external imaging of functionless pituitary adenomas displaying SRIH receptors. Eight patients with functionless pituitary adenomas, three patients with acromegaly, and three with macroprolactinoma were injected iv with 123I-labeled Tyr3-octreotide and then scanned with a γ-camera. Positive scans were obtained in the three acromegalics and in two of the eight patients with functionless pituitary tumors. The patients with macroprolactinoma had negative scans. The diagnosis of functionless pituitary adenomas was confirmed by light and electron microscopic examination as well as immunocytochemical studies. In vitro binding of [125I]Tyr11-SRIH to cell membranes was evaluated in four functionless and three GH-secreting adenomas removed from seven of the patients. All of the GH-secreting as well as one of the four functionless adenomas had high affinity SRIH-binding sites, without differences in number or affinity, whereas SRIH-binding sites were not detected in the others. Positive scans were observed only in patients bearing tumors with high affinity SRIH-binding sites. In conclusion, [123I]Tyr3-octreotide appears to be a promising tool for singling out, in vivo, patients with functionless pituitary tumors displaying SRIH receptors who might potentially benefit from octreotide treatment.
