DNA Methylation of Tumor Suppressor Genes in Pituitary Neuroendocrine Tumors.

CONTEXT: Epigenetic alterations may play a role in the development and behavior of pituitary neuroendocrine tumors (PitNETs). OBJECTIVE: To evaluate the effect of methylation of tumor suppressor genes (TSGs) on their gene expression and on the behavior of PitNETs. MATERIAL AND METHODS: We used methylation-specific multiplex ligation-dependent probe amplification and quantitative real-time PCR techniques to analyze the DNA-promoter hypermethylation and gene expression of 35 TSGs in 105 PitNETs. We defined functionality, size, and invasiveness of tumors according to their clinical manifestations, Hardy's classification, and MRI invasiveness of the cavernous sinus, respectively. RESULTS: We observed different methylation patterns among PitNET subtypes. The methylation status of TP73 correlated negatively with its gene expression in the overall series (P = 0.013) and in some subtypes. MSH6 and CADM1 showed higher methylation frequency in macroadenomas than in microadenomas in the overall series and in corticotroph PitNETs (all P ≤ 0.053). ESR1 and RASSF1 were more highly methylated in noninvasive than in invasive tumors in the overall series (P = 0.054 and P = 0.031, respectively) and in the gonadotroph subtype (P = 0.055 and P = 0.050, respectively). ESR1 and CASP8 appeared more hypermethylated in functioning than in silent corticotroph tumors (P = 0.034 and P = 0.034, respectively). CONCLUSIONS: DNA methylation of TSGs has a selective effect on their gene expression and on the growth and invasiveness of PitNETs. Its involvement in their functionality is biased because all silent operated tumors are macroadenomas, whereas all operated microadenomas are functioning ones. Therefore, the subtypes of PitNETs should be considered different entities.

Is it time to consider the expression of specific-pituitary hormone genes when typifying pituitary tumours?

The aim of the present study is to check whether we can replicate, in an independent series, previous results showing that the molecular study of pituitary-specific gene expression complements the inmunohistochemical identification of pituitary neuroendocrine tumours. We selected 112 patients (51 (46.4%) women; mean age 51.4±16 years; 102 macroadenomas (91.9%), 9 microadenomas (8.1%)) with complete clinical, radiological, immunohistochemical and molecular data from our data set of pituitary neuroendocrine tumours. Patients were different from those previously studied. We measured the expression of the pituitary-specific hormone genes and type 1 corticotrophin-releasing hormone and arginine vasopressin 1b receptors, by quantitative real-time polymerase chain reaction using TaqMan probes. Afterwards, we identified the different pituitary neuroendocrine tumour subtypes following the 2017 World Health Organization classification of pituitary tumours, calculating the concordance between their molecular and immuhistochemical identification. The concordance between molecular and immunohistochemical identification of functioning pituitary neuroendocrine tumours with the clinical diagnosis was globally similar to the previous series, where the SYBR Green technique was used instead of TaqMan probes. Our results also corroborated the poor correlation between molecular and immunohistochemical detection of the silent pituitary neuroendocrine tumour variants. This discrepancy was more remarkable in lactotroph, null-cell and plurihormonal pituitary neuroendocrine tumours. In conclusion, this study validates the results previously published by our group, highlighting a complementary role for the molecular study of the pituitary-specific hormone genes in the typification of pituitary neuroendocrine tumours subtypes.