Endocrine Abstracts

Introduction: Because of the increase of brain image explorations, the prevalence of pituitary tumours has increased a lot. Although most of them have an indolent behaviour, some behave aggressively, demanding a lot of resources for their management. Therefore, it is very important to identify them as soon as possible. The recent WHO 2017 classification of pituitary tumours gives insights into improving their identification using the immunostaining (IS) of transcription factors (TF) but does not include severity indexes. Therefore, a clinicalpathological classification is necessary. In 2013, Trouillas et al proposed a clinical-pathological classification that considered pituitary tumour subtypes along with radiological characteristics of the tumours. However, the identification of the pituitary tumour subtypes

was based on the IS of pituitary hormones that gives an inacceptable percentage of Null Cell and unusual plurihormonal tumours. The aim of the present study is to classify a series of pituitary tumours coming from a PTCOE, identified according to WHO 2017 recommendations, and using the clinicopathological classification proposed by Trouillas et al 2013.

Methods: A retrospective cohort study including 180 patients with pituitary tumors surgically treated (NES) in a PTCOE from 2013 to 2020. We collected the following information from the electronic medical records: function, IS features (pituitary hormones, Transcription Factors, MIB15 and p-53 IS and mitotic count), and MRI tumour characteristics: size, sinus invasion, optic chiasma displacement and T2 value. We have defined aggressiveness based on the presence of tumour invasiveness defined by pre-operative MRI/endoscopic intraoperative examination plus a Ki67 index ≥ 3%. We have analyzed the relative risk and adjusted RR for all variables through the Chi-squared test and logistic regression, respectively.

Results: From the 180 recruited patients we have excluded 64 patients due to lack of adequate MRI follow-up. Age 59 } 16 years, 52% males. Of the 116 patients included for analyses, 75% were non-functioning tumors and 35% were functioning. Of functioning tumors, 7.5% were CTs, 13.9% STs, 5.5% PRLs and 1% TTs. Of NFP tumors: 34.3% were silent gonadotroph tumors (SGT), 8% silent corticotroph tumors (SCT), 1% silent somatotroph tumors (SST), 1% silent lactotroph tumors (SLT) and 1.5% silent tirotroph tumors (STT). Recurrence rate in non-functioning and functioning pituitary tumors were 17.9% and 9.7% respectively. A higher risk of post-surgical recurrence was found in tumors with sinus invasion (RR 2.13 (IC95%: 1.6-2.7; P< 0.001)), a T2 ratio higher than 2 (RR 2.7 (IC95% 1.7-4.4, P< 0.001)). No statistical significance (P>0.05) was found for Ki-67 index >3%, however when combine this last one with sinus invasion (Trouillas IIb) in multivariant analyses a higher degree of recurrence was found (RR 42 (IC95%: 3.0-586); P< 0.05). Regarding non-functioning invasive silent CT and GT, 60% and 15.2% showed recurrence, with relative risk of 4 (IC 95% 1.2-13.2) and 1.5 (IC 95% 1.2-1.9), respectively. (P>0.05).

Conclusions: As expected tumors with high proliferation indexes and sinus invasion show worse behaviors and therefore should be managed more frequently and carefully. Therefore, the clinical pathological classification of pituitary tumors proposed by Trouillas et al is an important tool which should be used by all PTCOEs. As expected, SCT behaved more aggressively than SGT, highlighting the importance of a correct typification of the different pituitary tumors subtypes, using whenever is possible the IS of TF.