Prognostic indicators in pituitary tumors

Biomarker Discovery for Immunotherapy of Pituitary Adenomas: Enhanced Robustness and Prediction Ability by Modern Computational Tools

Pituitary adenoma (PA) is prevalent in the general population. Due to its severe complications and aggressive infiltration into the surrounding brain structure, the effective management of PA is required. Till now, no drug has been approved for treating non-functional PA, and the removal of cancerous cells from the pituitary is still under experimental investigation. Due to its superior specificity and safety profile, immunotherapy stands as one of the most promising strategies for dealing with PA refractory to the standard treatment, and various studies have been carried out to discover immune-related gene markers as target candidates. However, the lists of gene markers identified among different studies are reported to be highly inconsistent because of the greatly limited number of samples analyzed in each study. It is thus essential to substantially enlarge the sample size and comprehensively assess the robustness of the identified immune-related gene markers. Herein, a novel strategy of direct data integration (DDI) was proposed to combine available PA microarray datasets, which significantly enlarged the sample size. First, the robustness of the gene markers identified by DDI strategy was found to be substantially enhanced compared with that of previous studies. Then, the DDI of all reported PA-related microarray datasets were conducted to achieve a comprehensive identification of PA gene markers, and 66 immune-related genes were discovered as target candidates for PA immunotherapy. Finally, based on the analysis of human protein⁻protein interaction network, some promising target candidates (GAL, LMO4, STAT3, PD-L1, TGFB and TGFBR3) were proposed for PA immunotherapy. The strategy proposed together with the immune-related markers identified in this study provided a useful guidance for the development of novel immunotherapy for PA.

Galectin-3 expression in pituitary adenomas as a marker of aggressive behavior

The purpose of this retrospective study was to investigate the role of galectin-3 (LGALS3) expression in predicting the recurrence and the progression potential of prolactin (PRL) and adrenocorticotropic hormone (ACTH)-producing pituitary adenomas and its correlation with the RUNX1 and RUNX2 transcription factors involved in the regulation mechanism of LGALS3 expression. Clinical, neuroradiologic, and follow-up data from 92 pituitary adenomas, including 59 PRL cell adenomas and 33 ACTH-functioning pituitary adenomas, were collected. The LGALS3 expression was analyzed by both immunohistochemistry and quantitative real time-polymerase chain reaction, whereas RUNX1 and RUNX2 were analyzed by quantitative real time-polymerase chain reaction only. The data obtained indicated that invasive growth with suprasellar extension, Ki-67 labeling index, and LGALS3 immunohistochemical and/or LGALS3 messenger RNA levels are the most important histologic features for assessing a high risk of progression or recurrence of PRL- and ACTH-functioning pituitary adenomas. Multivariate Cox regression analysis assessed LGALS3 immunohistochemical positivity in at least 30% of neoplastic cells and/or LGALS3 messenger RNA positivity (P < .001) as strong predictive factors of recurrence/tumor progression followed by a Ki-67 labeling index greater than 3% (P = .019) in the 81 cases in which follow-up data were available. In addition, a significant correlation between LGALS3 and RUNX1 expression levels (P = .0435) was found. This retrospective immunohistochemical and molecular study demonstrated that LGALS3 expression appeared to be a predictive factor of the aggressive behavior of PRL- and ACTH-functioning pituitary adenomas, and its expression was correlated with RUNX1 expression levels.

Analysis of IMP3 expression in normal and neoplastic human pituitary tissues

Insulin-like growth factor II mRNA-binding protein 3 (IMP3) is an oncofetal protein highly expressed in fetal tissue and malignant tumors but rarely found in adult benign tissues. In various tumors, IMP3 expression is correlated with increased tumor aggressiveness and reduced overall survival. To our knowledge, IMP3 expression has not been investigated in pituitary tumors. We analyzed the immunohistochemical expression of IMP3 in five normal pituitary tissues and 75 pituitary tumors (64 adenomas and 11 carcinomas) to determine if specific tumor types expressed IMP3 and if there were differences in IMP3 expression between adenomas and carcinomas. Immunohistochemical analysis showed that IMP3 was positive in four (80%) normal pituitaries with focal stain in a subset of normal anterior pituitary cells. IMP3 was expressed in 31% (20/64) of adenomas and in 36% (4/11) of carcinomas. A slightly higher level of IMP3 expression was observed in PRL-GH-TSH adenomas compared to the other types of pituitary adenomas. Expression of IMP3 was not significantly higher in carcinomas than in adenomas (p = 0.737). RT-PCR and Western Blotting supported the heterogeneous expression of IMP3. These results indicate that IMP3 is expressed both in normal and in neoplastic pituitary gland tissues without significant differences in expression levels in pituitary carcinomas.

Pathobiology of Pituitary Adenomas and Carcinomas

OBJECTIVE

To examine relationships between pituitary tumors and lesion size, invasiveness, resectability, deoxyribonucleic acid ploidy, cell cycle profile, mitotic activity, and immunoreactivity for MIB-1, proliferating cell nuclear antigen (PCNA), p27Kip1, and p53.

PATIENTS AND METHODS

One hundred fifty-three adenomas of most pathological subtypes, including 20 medically treated and prolactin and growth hormone-containing tumors, as well as 10 premetastatic tumors and 13 pituitary carcinomas, were studied.

RESULTS

Significant (P < 0.05) differences were noted between functional versus nonfunctional adenomas (percent aneuploidy, percent S phase, p27Kip1 labeling indices [LI], male sex, tumor size, and frequency of visual disturbance); Cushing's versus silent adrenocorticotropin adenomas (percent hypertetraploidy, p53 LI, tumor size, visual disturbance, and resectability); untreated versus medically treated prolactin cell adenomas (MIB-1 LI, p53 LI, and resectability); untreated versus medically treated growth hormone-containing adenomas (percent diploidy, percent S phase, MIB-1 LI, p53 LI, and p27 LI); untreated prolactin cell adenomas versus premetastatic tumors (percent hypertetraploidy, PCNA LI, p53 LI, invasiveness, and resectability); untreated growth hormone-containing adenomas versus premetastatic tumors (percent diploidy, percent S phase, PCNA LI, p53 LI, invasiveness, and resectability); Cushing's adenomas versus premetastatic tumors (percent diploidy, percent hypertetraploidy, percent S phase, MIB-1 LI, p53 LI, tumor size, invasiveness, visual disturbance, and resectability); Nelson's adenomas versus premetastatic tumors (p53 LI, tumor size, invasiveness, and resectability); silent adenomas as a whole versus nonfunctional adenomas (percent nondiploid, percent S phase, invasiveness, and respectability); silent adrenocorticotropin adenomas I and II versus silent adenoma Subtype III (invasiveness); silent adrenocorticotropin adenoma Subtypes I and II versus premetastatic tumors (MIB-1 LI and invasiveness); silent adenoma Subtype III versus premetastatic tumors (PCNA and p53 LI); and premetastatic tumors versus metastatic pituitary carcinomas (MIB-1 LI).

CONCLUSION

Only trends toward differences were noted between Cushing's versus Nelson's adenomas and between prolactinomas of reproductive female patients versus those of menopausal female patients and male patients. Too few "atypical adenomas" were encountered to permit their comparison with premetastatic tumors, but our results suggest that most pituitary carcinomas arise by malignant transformation from adenomas.