Distribution of estrogen and progesterone receptors isoforms in endometrial cancer

Approximately 70–80% of sporadic endometrial carcinomas are distinguished as type I carcinomas, which is the most common malignancy in female reproductive tract and defined as EC. Well-described EC prognosis includes stage of disease at the time of diagnosis, histologic type, degree of tumor differentiation, depth of myometrial invasion and lymphovascular space invasion. The glandular epithelium from which the cancer arises is hormone responsive, expressing both PRs and ERs [1]. EC often develops from endometrial hyperplasia, which is attributed to prolonged exposure to estrogen in the absence of (unopposed) sufficient progesterone [2], and is often well differentiated and noninvasive or superficially myoinvasive, rarely producing metastases and expressing ER [3]. Whereas early-stage, well differentiated EC usually retain expression of both receptors, advanced stage, poorly differentiated tumors often lack one or both of these receptors, which has been correlated in many studies with a poor prognosis [4, 5]. The majority of estrogen-dependent carcinomas occur during the post-menopausal period, when active sex steroids are not produced by the ovaries. Therefore, in-situ estrogen metabolism has a crucial role in the development and progression of EC in this period [6]. Both estrogen and progesterone exert their effect through intra-and extra-nuclear receptors. ER exists in 2 main forms, ER-α and ER-β, encoded by separate genes, ESR1 and ESR2 respectively, binding the same estrogen response elements (EREs) and regulate similar sets of genes [7]. However, ER-α and ER-β has a distinct pattern of expression in the tissues [8], which varies during cellular proliferation and differentiation [9]. ER-α is required for the basic development of estrogen sensitive tissues and ER-β is required for organization and adhesion of epithelial cells and hence for differentiated tissue morphology and its functional maturation [10].

PR has been implicated in the development of endometrial cancer, as well. The single-copy PR gene uses separate promoters and translational start sites to produce two isoforms, PR-A and -B [11], which are in fact two functionally distinct transcription factors [12], mediating their own response genes and physiological effects with little overlap [13]. The physiological roles of progesterone in the regulation of endometrial tissue are, in general, considered to antagonize estrogen-mediated cell proliferation and to induce cellular differentiation [14, 15]. Loss of total PR expression was found in well and in poorly differentiated EC, and was related to PR-A [16‐18]. Highly malignant forms of endometrial, cervical and ovarian cancers have been correlated with over-expression of PR-B [19, 20]. Another examined marker in this study is Ki67, a widely used nuclear marker expressed during all active phases of the cell cycle, but absent from resting cells (G0) [21], and therefore its expression is examined in order to assess proliferative activity. High Ki-67 expression was found in various types of endometrial carcinomas [22] and correlates with histological grade, depth of myometrial invasion and risk of recurrence [23‐25]. In the current study, the common examination of receptors profile in the epithelial cells of the tumor was under focus by the evaluation of ER and PR isoforms expression as well as Ki67 in the stromal cells and the epithelial glands of EC specimens. Profile of expression was correlated to the tumor histological depth.

Molecular tumor classification, which includes PR and ER expression, is an integral part of the disease characteristics. The presence of steroid receptors ER-α, PR-A and PR-B has been quantitatively associated with histologic differentiation [30, 31], response to therapy [32] and metastatic potential [33]. ER-α expression was found to be decreased in EC [18, 34] and is further decreased as EC grading is advanced [35‐38]. In correlation, our results demonstrate significantly reduced expression of ER-α in both glands and stroma of endometrioid tumor in relation to non-malignant endometrial tissue (Figure 1). The expression of ER-α is lower in the stroma than in the glands of EC, indicating that stroma cells are significantly more affected than the epithelial cells. ER-β quantification faced technical problems and therefore was not assessed in the current study. Loss of ER suggests an advanced molecular pathology of the tumor with the deregulation of signaling pathways. Common deregulation courses include PTEN inactivation by mutation [39], de novo methylation of ER-α gene and aberrant methylation of CpG islands [1]. These epigenetic alterations occur in a wide variety of tumors [8, 40‐44], including endometrial cancer [36, 45].

PR expression of either one or both of the two PR isoforms was found to be reduced or absent in endometrial cancer [16‐18, 46], mostly lower for the higher histological grade [47‐49] and inversely correlates with myometrium invasion [50, 51]. Our results demonstrate that PR-A shows the exact same pattern of expression as ER-α in the gland and stroma cells, as well as in the different portions of EC specimens (Figure 2). It is well documented in the literature that the transcription of PR gene is induced by estrogen and inhibited by progesterone in the majority of estrogen responsive cells, so the expression of ER and PR is considered to be coordinated [27, 37, 52, 53]. As described, we found significantly and differentially altered expression of sex steroid receptors in superficial and deep sections of the specimens. Previous reports, which support our findings, describe total protein expression in the tissue. Our findings, describing the expression of PR-A and ER-α in the stroma and epitheial cells in EC solely, is implicated in the mitogenic response of epithelial cells to estrogen, which is mediated indirectly by stromal ER [54]. A model for this assumption was demonstrated by co-culture of non-expressing ER stroma cells and ER-positive epithelial cells [55]. No epithelial proliferation in response to estrogen was detected in this model, or in a model of pure epithelial cultures, an induction observed in co-cultures of normal uterine stroma and epithelial cells [55]. Evidently estrogen induced epithelial proliferation requires an ER-positive stroma. Response of uterine epithelial cells to progesterone was also found to be mediated by stromal PR [56]. This mediated operation between the cells may be implicated and result in the altered pattern of receptors expression in the transformed cells, found in our study. In addition to the well-known growth inhibiting effect of progesterone, it plays an important role in regulating invasive properties of endometrial cancer cells. A correlation was found between decreased PR expression in EC tumors and the expression of E-cadherin and myometrial invasion [57, 58]. An extensive myometrial invasion may be a progeny of epithelial to mesenchymal transition (EMT) which is highly implicated in EC tumors invasive characteristics [59, 60]. Therefore, the reduced expression of ER-α and PR-A in the tumor cells, particularly the significantly reduced expression in the stroma cells, may indicate an invasive characteristics of the tumor, as described for ER-α [61], and the deep portion of the tumor is of special interest. These findings in both superficial and deep portions of the tumor stand against the extra-tumoral portion of the tumor, which was found to be affected as well, but to a lower extent. PR-B quantification showed reduced expression in the epithelial glands of superficial and deep portions of EC (Figure 3). Supporting our findings, PR-B promoter was previously found to be methylated in endometrial carcinoma [62] and the loss of expression was referred to as an independent prognostic factor for cause-specific survival in high risk patients [63]. The significantly high expression of PR-B in the extra-tumoral portion of the malignant specimens may imply a certain protective reaction opposing invasive properties of the tumor cells. In a study conducted by Balmer NN et al. [64], in which tumoral- and extra-tumoral- portions were examined by immunohistochemistry, resembling the current study methodology, PR-B expression was found to be significantly higher in carcinoma-associated nonmalignant endometrium compared to endometrial carcinoma. Zafran et al. [65] found that a state of PR-B dominance, like in the cell line HEC-1A, was less invasive than cell lines that PR-A is the predominantly expressed variant. PR-A may be associated with a cell- and promoter specific repression of PR-B [66] and imbalance in PR-A to PR-B ratio is frequently associated with carcinogenesis [67]. The relative over-expression of PR-B, which is referred to as an endometrial estrogen agonist [68], without transcriptional repression by PR-A, as shown in our findings, may also be related to the metastatic potential and partially cause deviation from sex steroidal dependency in endometrial cancers [33]. Our results show higher expression of Ki-67 in the malignant tissue than in the nonmalignant, as seen in previous studies [22‐25, 69]. A wide score range of Ki67 expression was found in the non-malignant biopsies. These results correlate with the expression of Ki67 in normal cyclical endometrium, in which Ki-67 staining is intense and diffused in the proliferative phase, but decreases dramatically in the early and mid-secretory phase.

In the current study, we have showed the importance of referring to steroid receptors profile in the stroma as well as the epithelial cells. The problem in attaining a consensus regarding assessment of endometrial carcinomas was recently discussed [70, 71] and updating the pathologist biomarkers panel was shown to be useful in characterizing EC tumors and in patients prognosis [72‐74]. Studies of steroid receptors pattern of expression help in understanding their mechanism of action in target tissues, and could be helpful in defining biologically different subgroups and therapeutic efficacy. We have found that the ratio of ER-α and PR-A expressions in the epithelial glands and the stroma of EC biopsies has a distinct values in different portions of the tumor. These findings may serve in the marker panels of the pathologist in order to improve diagnostic reproducibility. It should be noted that this study has focused on a small and limited group of biopsies. Further analysis in large scale study may contribute to the understanding of ER and PR isoforms expression in EC, and a possible use of ER-α and PR-A relative expression as a clinical tool.