Gene expression profiles separate endometriosis lesion subtypes and indicate a sensitivity of endometrioma to estrogen suppressive treatments through elevated ESR2 expression

Abstract

<h3>Background</h3><p>Endometriosis is a common, gynaecological disease characterised by the presence of endometrial-like cells growing outside the uterus. Lesions appear at multiple locations, present with variation in appearance, size and depth of invasion. Despite hormones being the recommended first-line treatment, their efficacy, success and side effects vary widely amongst study populations. Current, hormonal medication for endometriosis is designed to suppress systemic oestrogen. Whether these hormones can influence the lesions themselves is not yet clear. Evidence of hormone receptor expression in endometriotic lesions and their ability to respond is conflicting. A variation in their expression, activation of transcriptional co-regulators and the potential to respond may contribute to their variation in patient outcomes. Identifying patients who would benefit from hormonal treatments remain an important goal in endometriosis research.</p><h3>Methods</h3><p>Using gene expression data from endometriosis lesions including endometrioma (OMA, <em>n</em> = 28), superficial peritoneal lesions (SUP, <em>n</em> = 72) and deeply infiltrating lesions (DIE, <em>n</em> = 78), we performed principal component analysis, differential gene expression and gene correlation analyses to assess the impact of menstrual stage, lesion subtype and hormonal treatment on the gene expression.</p><h3>Results</h3><p>The gene expression profiles did not vary based on menstrual stage, but could distinguish lesion subtypes with OMA significantly differentiating from both SUP and DIE. Additionally, the effect of oestrogen suppression medication altered the gene expression profile in OMA, while such effect was not observed in SUP or DIE. Analysis of the target receptors for hormonal medication indicated <em>ESR2</em> was differentially expressed in OMA and that genes that correlated with <em>ESR2</em> varied significantly between medicated and non-medicated OMA samples.</p><h3>Conclusions</h3><p>Our results demonstrate of the different lesion types OMA present with strongest response to hormonal treatment directly through <em>ESR2</em>. The data suggests that there may be the potential to target treatment options to individual patients based on pre-surgical diagnoses.</p>