Somatotroph Adenoma with Dual Transcription Factor Expression

A 20-year-old male presented with evidence of gigantism/acromegaly. Endocrinological investigations identified elevated growth hormone levels and a failed glucose tolerance test. Imaging revealed a macroadenoma expanding the sella with encroachment on the optic chiasm and cavernous sinuses. Trans-sphenoidal resection was undertaken and a gross total removal was achieved. Histopathological features were typical of a densely granulated somatotroph adenoma with abundant growth hormone expression, scattered prolactin expression and sparse examples of fibrous bodies. Unexpectedly, the adenoma not only expressed PIT-1 but also SF-1 transcription factors. This finding suggests that the adenoma may have been pluripotent. The prognostic significance of this finding is uncertain although the patient is stable from an endocrinological and imaging perspective approximately one-year post-op. A pituitary adenoma of this nature has not been previously reported. The recent literature on atypical transcription factor expression patterns and revisions to the classification of pituitary adenomas will be reviewed.Learning ObjectivesAppreciate the rarity of dual transcription factor expression in pituitary adenomasRationalize the use of transcription factor characterization in the revised WHO classification of pituitary adenomas

Pituitary Somatotroph Adenoma Cell-Derived Exosomes: Characterization of Novel Non-Hormonal Functions

Exosomes, small extracellular vesicles carrying lipids, proteins, DNA and RNA, enable intercellular communication. Pituitary-derived exosomes have not been well validated, and as no human pituitary cell lines are available, we characterized exosomes derived from rat somatotroph tumor cells (GH1 and GH3). Rat FR and H9C2 cells were used as non-pituitary controls. Exosomes were isolated from serum-free culture supernatants by combining ultrafiltration and ultracentrifugation to eliminate hormone contamination. Derived exosomes were analyzed by NanoSight to visualize, size, and count particles. Exosomal proteins were extracted and exosome markers including TSG101, ALIX, CD63, HSP70, HSP90 detected by Western Blot. The exosome inhibitor GW4869 (10 µM, 30 h) reduced exosome release (up to 81%), whereas treating cells with hydrocortisone (0.1 µM, 72 h) increased exosome production (up to 42%) in GH1 and GH3 cells. Exosomal shuttle RNA characterized by RNA-Seq showed distinct pituitary vs non-pituitary exosome RNA profiles. Selected miRNAs assessed in exosomes and corresponding cells by qRT-PCR validated exosomal RNA-seq and suggested that miRNA signatures in exosomes and in respective cells of origin were concordant. Next, we explored downstream signaling of GH1-derived exosomes (GH1-exo) in vitro and in vivo and studied biological actions in normal hepatocytes and in malignant cells. As evidenced by mRNA-seq, GH1-exo distinctly altered signaling pathways in rat primary hepatocytes, vs pathways elicited by GH or PRL (0.5 µg/mL, 24 h). GH1-exo, FR-exo or vehicle were intravenously injected to 4-week-old female Wistar rats twice weekly for 4 weeks (5*109 exo/200 g, n=3), and livers dissected for mRNA-seq. Among GH1-exo specifically regulated genes, EIF2AK/ATF4, involved in cAMP responses and amino acid biosynthesis, were attenuated. In hepatocytes, GH1-exo suppressed up to 65% of nascent protein synthesis and reduced forskolin (10 µM)-stimulated cAMP activity by 19%, while GH (0.01-1 µg/mL) did not affect this pathway. Notably, GH1-exo also attenuated malignant cell motility. Both GH1-exo incubation or GH1 cell co-culture (48 h) suppressed migration, invasion and wound healing of HCT116 cancer cells by up to 70%. In contrast, treatment with rGH (0.5 µg/mL) increased HCT116 motility. Intravenous administration of GH1-exo (1010 exo/mouse, twice a week for 5 weeks) decreased metastatic tumor volume by 40% in nude mice harboring splenic HCT116 implants (5*105 cells/mouse, n=10), and especially abrogated hepatic metastases. mRNA-seq of GH1-exo treated HCT116 cells vs controls indicated dysregulated p53 and MAPK pathways, which may partially explain mechanisms underlying motility attenuation. The results elucidate novel biological actions of somatotroph adenoma cell-derived exosomes and suggest exosomes as non-hormonal messengers produced by pituitary tumors.

Intensity-modulated Radiotherapy for Pituitary Somatotroph Adenomas

Objective To summarize our experience in the treatment of pituitary somatotroph adenomas by fractionated intensity-modulated radiotherapy (IMRT), describe the treatment outcomes, and determine predictors. Methods and Materials Patients with pituitary somatotroph adenoma treated by IMRT in our institution from August 2009 to January 2019 were reviewed. A total of 113 patients (37 male) were included in this study. The median age was 33 years (range 12-67 years). A total of 112 patients had not achieved complete remission after surgery, and 1 patient was treated by radiotherapy (RT) alone because she refused to surgery. The median growth hormone level was 8.6 ng/mL (range 2-186 ng/mL) and the median insulin-like growth factor (IGF)-1 level was 732 ng/mL (range 314-1485 ng/mL) pre-RT. The radiation doses to clinical target volume were usually 50-56 Gy in 25 to 30 fractions and to gross tumor volume were 60.2 Gy in 28 fractions while simultaneous integrated boost-IMRT used. After RT, the patients were followed up with endocrine testing every 6 to 12 months and magnetic resonance imaging annually. Endocrine complete remission was defined as a normal sex- and age-adjusted IGF-1 level without any pituitary suppressive medications. The outcomes including endocrine remission and new hypopituitarism after RT were recorded. The median follow-up time was 36 months (range 6-105.5 months). Results The endocrine complete remission rates of IGF-1 at 1, 2, 3, and 5 years were 6%, 22.8%, 48.6%, and 74.3%, respectively. The median time to complete remission was 36.2 ± 3.8 months. The tumor control rate was 99% during the follow-up. The overall incidence of RT-induced hypopituitarism was 28.3% at the last follow-up. Univariate and multivariate analysis demonstrated that tumor sizes before RT, pre-RT IGF-1 level, and age significant predicted the endocrine remission. Conclusions IMRT is a highly effective treatment for pituitary somatotroph adenoma. Endocrine remission rate, tumor control rate, the median time to remission and hypopituitarism incidence are similar to stereotactic radiosurgery. Age and IGF-1 level before RT were significant predictive factors in endocrine remission.

MON-317 Prognostic Factors and Mortality According to the Structural and Functional Classification of Acromegaly

Background: acromegaly is a systemic disorder caused by overproduction of growth hormone (GH) where most common cause is a pituitary somatotroph adenoma. Different prognostic factors have been previously reported, such as adenoma granulation, and p21 and somatostatin receptor type 2 (SSTR2) immunohistochemistry expression. Considering these factors, three different acromegaly subtypes were proposed. Type 1 tumors are dense granulated microadenomas with higher expression of p21 and SSTR2, with better prognosis. In contrast, type 3 are sparsely granulated, highly invasive macroadenomas, with less expression of SSTR2 and p21, and worst prognosis with less responsiveness to treatment. Type 2 adenomas showed intermediate prognosis. Objective: to determine independent prognostic factors according to somatotroph adenoma subtypes in patients with acromegaly. Material and methods: this is a comparative, observational, longitudinal study. We classified patients according to p21, SSTR and tumor granulation as previously reported. Then, bivariate and multivariate using binary logistic and Cox-regression analyses were performed. Results: 222 patients (52% women and 48% men, mean age 40+-13 years old) with confirmed diagnosis of acromegaly were classified in the three acromegaly types. Mean age at diagnosis was significantly different among groups, with 45, 41 and 38 years old, in types 1 to 3, respectively (p 0.010). All patients with type 1 (n=47) showed microadenomas, and type 2 (n=72) and 3 (n=103) macroadenomas (p <0.001). Invasiveness was present in 11% in type 1, 0% in type 2, and 100% in type 3 (p <0.001). GH and the upper limit number IGF-1 index was significantly higher in in type 3 vs. type 1 (p <0.001). Hypertension (43% vs. 34% vs. 23%), cardiovascular/cerebrovascular disease (44% vs. 28% vs. 28%), diabetes (47% vs. 29% vs. 24%), hypertriglyceridemia (54% vs. 24% vs. 22%), and high LDL-c (43% vs. 30% vs. 27%) were significantly higher in type 3 vs. type 2 vs type 1 acromegaly patients, respectively (all p<0.05). Using multivariate Cox regression analysis (adjusted for neurosurgery and radiotherapy) we identified that type 3 acromegaly showed worst prognosis with higher probability of remaining active (2LogR=652.6, chi square=17.7, p=0.001), and higher mortality (2LogR=489, chi square=43, p<0.001) than type 2 vs. type 1 acromegaly patients at last follow-up. Independent parameters related with disease activity were acromegaly types 2 and 3 (OR=2.7; IC95% 1.9-6.7, p=0.005), and radiotherapy (OR=0.36, 0.18-0.70, p=0.003). Conclusions: Type 3 acromegaly patients showed higher frequency of comorbidities, disease activity, and risk of mortality, adjusted for treatment, than type 2 and type 1 patients. References 1. Cuevas-Ramos, D., Carmichael, J. D., Melmed, S. (2015). A Structural and Functional Acromegaly Classification. JCEM, 100(1), 122–131.doi:10.1210/jc.2014-2468