Determining the endometrial phenotype of women with abnormal uterine bleeding

INTRODUCTION: Abnormal Uterine Bleeding (AUB) affects at least 1-in-3 reproductive-aged women and has a debilitating impact on quality of life. It is a symptom which can be caused by several possible conditions. These conditions have been grouped together under the International Federation of Obstetrics and Gynecology (FIGO) AUB classification system, and can be divided into structural causes of AUB and non-structural causes of AUB. Two common structural causes of AUB, fibroids and adenomyosis, impact the myometrium, yet, the mechanism by which the endometrial phenotype might be modified, with consequent AUB, is unknown. Research to-date has predominantly investigated the cellular and molecular changes occurring within the fibroid or adenomyotic lesion, however, there is lack of evidence related to the changes occurring in the endometrium due to the presence of these myometrial conditions. Fibroids and adenomyosis are sex-steroid dependant conditions, and lines of evidence suggest there is an impaired progesterone response in the endometrium from women with adenomyosis, however, there is a dearth of evidence related to the endometrium from women with fibroids. Women with AUB, do not always respond to therapeutic agents for AUB, often ligands for the progesterone receptor, further suggesting an impaired responsiveness to progesterone and progesterone regulated pathways. HYPOTHESIS: The endometrium from women with myometrial causes of AUB (fibroids or adenomyosis), is phenotypically different to the endometrium from women without identifiable causes of AUB, notably with aberrations in the responsiveness to progesterone. RESEARCH QUESTIONS: 1) Are there features consistent with impaired responsiveness to progesterone in the endometrium from women with myometrial causes of AUB? 2) Does the phenotype of the endometrium from women with myometrial causes of AUB differ from the endometrium from women with no identifiable causes of AUB? 3) Can phenotypic differences in the endometrium from women with myometrial causes of AUB be detected using endometrial derived stromal and epithelial cells in vitro? METHODS: Endometrial samples from a Female Reproductive Tissue Tract Resource (University of Edinburgh: REC Approvals: 20/ES/0119; 19/SS/0102) were categorised according to patient’s clinical history, menstrual cycle stage (confirmed by circulating oestradiol and progesterone levels at time of biopsy and histology), presence of fibroids not in contact with the endometrium (AUB-L), adenomyosis (AUB-A), fibroids and adenomyosis (AUB-L&A) or no identifiable cause of AUB (AUB-E; used as a comparator). All other identifiable causes of AUB were excluded. Endometrial samples were divided into proliferative phase (AUB-L, n=15, AUB-A n=6, AUB-L&A n=4, AUB-E n=24) and secretory phase (AUB-L, n=19, AUB-A n=4, AUB-L&A n=5, AUB-E, n=19). Quantitative-RT-PCR assessed sex steroid receptors: PR, PRB, ESR, AR and candidate progesterone-regulated genes (FOXO1, HAND2, IL15, IGFBP1, HOXA10, 17ßHSD2 and FOXM1). Immunohistochemistry evaluated spatiotemporal immunolocalisation, on a selection of samples, of sex steroid receptors (PR, PRB, ERα, AR) as well as Ki67, 17ßHSD2, and HOXA10 in AUB-L, AUB-A, AUB-L&A, and AUB-E samples. A randomly selected subgroup of samples (proliferative phase AUB-L n=6, AUB-A n=6, AUB-E n=6; secretory phase AUB-L n=6, AUB-A n=4, AUB-E n=6) were investigated using bulk RNA sequencing (Lexogen Quantseq platform). Data were interrogated for differential expression and pathway analyses. Data were validated using a separate set of endometrial samples. Finally, the component cells of the endometrial tissue from carefully characterised patients (AUB-L n=1, AUB-E n=2) were cultured in vitro to assess the phenotype of endometrial stromal cells (ESCs), endometrial epithelial organoids (EEOs) and ESCs and EEOs in co-culture. ESCs were exposed to progesterone and incrementally increasing doses of levonorgestrel (0.1ng/ml – 500ng/ml), a widely used treatment for AUB and PR ligand, to study the progesterone responsiveness. Analysis focused on the comparison of data derived from endometrium from women with myometrial causes of AUB with the data derived from endometrium from women with no identifiable causes of AUB (AUB-E). RESULTS: Endometrium from women with AUB associated with fibroids: Using a candidate approach, the gene expression of all sex steroid receptors (PGR (p=0.03), PRB (p=0.02), ESR (Of the most significantly differentially expressed pathways, signalling involving the sex steroid receptors ESR and PGR, was highlighted to be upregulated, whilst the regulation involving insulin like growth factor binding proteins (IGFBPs) or VEGF (marker of angiogenesis) were downregulated. Immunohistochemistry, confirmed expected spatial-temporal immunolocalisation of AR in AUB-L. However, PR, PRB, and ERα secretory phase immunolocalisation was demonstrated in the glandular epithelium in AUB-L, higher than the immunolocalisation in the secretory phase endometrium from women with AUB-E. Finally, a pre-clinical in vitro model of the endometrium derived from women with AUB (AUB-L and AUB-E) was developed via the growth and expansion of ESCs and EEOs, both in isolation and in co-culture. The protocol to expand ESCs and EEOs was found to be heavily reliant on the phenotype of the patient from which the cells originated. ESCs derived from endometrium from women with AUB, in isolation, demonstrated physiological responses to sex steroid hormones and a PR ligand, levonorgestrel. Endometrium derived from women with AUB associated with adenomyosis: Proliferative phase endometrium from women with AUB-A showed no change to sex steroid receptor mRNA expression, however 17ßHSD2 was upregulated (p=0.046). Progesterone receptor and 17βHSD2 protein immunolocalisation was increased in the proliferative phase. Secretory phase PGR (p=0.03), PRB (p=0.01), ESR (p=0.04) and AR (p<0.01) were upregulated in endometrium from women with AUB-A, whilst PRB, ERα and Ki67 immunolocalisation were increased in the secretory phase. Differential expression analysis demonstrated that there were no significant differentially expressed genes when comparing the proliferative phase endometrium from women with adenomyosis to women with AUB-E, however, in the secretory phase, two genes were significantly upregulated and two genes were significantly downregulated. DISCUSSION AND CONCLUSION: Well-characterised endometrium from women with AUB-L and AUB-A revealed altered immunolocalisation of progesterone-regulated genes and sex-steroid receptor (PR +/-PRB) protein immunolocalisation in the secretory phase underpinning likelihood of impaired progesterone response. In women with AUB-L, perturbations in progesterone-regulated pathways mediated via altered localisation of PR/PRB in endometrial cells, may indicate a modified “endometrial phenotype”. The comparisons made within this PhD reveal a likely secondary endometrial disorder in the endometrium derived from women with AUB associated with myometrial causes. These findings are clinically relevant, and may underlie the altered response to PR ligands in the treatment of AUB.