Does Pituitary Tumour Apoplexy Have an Infective Aetiology?

Abstract

Pituitary tumour apoplexy (PA) is a life-threatening clinical syndrome resulting from a haemorrhage and/or an infarction of a pre-existing pituitary adenoma. In a recent study, PA subjects demonstrated dysbiosis in the sinus microbiota, whereby known respiratory tract pathogens were present. The absence of such bacteria from nonapoplectic controls could suggest a potential role of the microbiota in precipitating this condition. This doctoral thesis focuses on the microbiota of the sphenoid sinus in pituitary apoplexy (PA) subjects versus non-apoplectic controls with non-functional pituitary adenoma (NFPA) using culture-independent analysis of sinus tissue biopsy specimens. The three data chapters in this thesis attempt to address the aim of better understanding the role of the microbiota in pituitary tumour apoplexy. In a study to investigate the microbiota of the sphenoid sinus, sinus mucosa biopsies were collected from NFPA subjects (n=36) during endoscopic trans-sphenoidal surgery for resection of a pituitary adenoma. Samples were investigated using next- generation sequencing of the V4 region of the 16s rRNA gene using a Miseq platform. Taxonomic analysis of the consensus sequences indicated the predominance of the Firmicutes with members of the staphylococci most represented at genus level in terms of relative abundance. Analysis of similarity (Adonis) indicated that there were no significant ordination clusters in sinus microbiota datasets according to collected metadata. Alpha diversity metrics suggested these sites to harbour complex microbiota profiles, (mean=541 OTUs). An understanding of the microbiota of this site has important implications when developing model systems more representative of the sinuses, allowing the future investigation of community dynamics. In order to support pre-clinical investigations, an in vitro model was designed to be representative of the sinuses in adenoma patients. A combination of the MSD and artificial airway surface media were applied to support obligate anaerobes. Microcosms were generated from a single, fresh tissue biopsy specimen given COVID-19 related restrictions but permitted the piloting of this system to support complex sinus microcosms. Viable plating on selective media combined with amplicon sequencing from three independent runs permitted the testing of inter-model stability, but also intra-model dynamic compositional stability. Overall, systems were found to climax at about 3 days and achieved dynamic stability for the remainder of the test period. Most of the culturable consortia were dominated by anaerobic (facultative and/or obligate) species, as evidenced by differential count data (8 log10 CFU per filter and per ml). Homogeneity of variance testing indicated variation in one run, with remaining runs supporting the continuous maintenance of sinus microcosms. Analysis of Similarity (Adonis) demonstrated significant clustering according to model run category following FDR adjustment (FDR P= 0.001). The compositional similarity of the microcosms based on viable counting may differ to that determined by deep sequencing analysis and as such, a combination of approaches is advisable for future validation studies. The development of a reproducible model system will permit the investigation of community dynamics within the sinuses and a better understanding of the underlying aetiological mechanisms of pituitary apoplexy. Here, we have applied a combination of next generation sequencing and MR-imaging to better understand the microbiota of the sinuses in patients with and without apoplexy of the pituitary gland. In order to investigate a potential bacterial aetiology for pituitary apoplexy, sphenoid sinus tissue biopsies were analysed from patients presenting with non-functioning pituitary adenomas (NFPA) and pituitary apoplexy and subjected to DNA extraction and PCR amplification using primer pairs specific for the V4 region of the 16S rRNA gene. 7/7 apoplexy patients exhibited signs of sphenoid sinus mucosal thickening (SSMT) with different grades ranging from (1- 13mm), in contrast, none of the control cases showed such changes. Known risk factors of PA were only observed in 2 (28%) apoplexy patients. A significant clustering of OTUs was observed on ordination when categorised as apoplexy and non-apoplexy (FDR p value = 0.009**). Differential abundance testing according to this category suggested significant differences in the relative abundance of staphylococci, corynebacteria and propionibacteria. In conclusion, further studies are needed using larger sample sizes. Quantitative PCR could be applied in the sinus tissue samples to allow absolute determination of sinus microbial communities in such patients, along with immunohistochemistry investigations to determine the probability of the presence of sinus pathogens within tissues obtained from apoplexy patients.