Few markers of ovarian cancer prognosis have been established, perhaps because potential subtype associations are missed in studies including patients with all histopathologic subtypes. In a study reported in The Lancet Oncology, Weiva Sieh, MD, PhD, Assistant Professor of Health Research and Policy at Stanford University, and colleagues assessed the prognostic effect of progesterone receptor (PR) and estrogen receptor (ER) status among nearly 3,000 women with invasive epithelial ovarian cancer.1 They found that tumor ER and PR are prognostic markers for endometrioid and high-grade serous ovarian cancers.
In the study, 12 studies participating in the Ovarian Tumor Tissue Analysis consortium contributed tissue microarray sections and clinical data from women who had been diagnosed with invasive serous, mucinous, endometrioid, or clear-cell carcinomas of the ovary. For a patient to be eligible, tissue microarrays, clinical follow-up data, age at diagnosis, and tumor grade and stage had to be available. PR and ER status was assessed by central immunohistochemistry analysis done by blinded pathologists. PR and ER staining was defined as negative (< 1% tumor cell nuclei), weak (1% to < 50%), or strong (≥ 50%).
A total of 2,933 women with invasive epithelial ovarian cancer were included in the analysis, consisting of 1,742 with high-grade serous carcinoma, 110 with low-grade serous carcinoma, 207 with mucinous carcinoma, 484 with endometrioid carcinoma, and 390 with clear-cell carcinoma.
PR and ER Expression
PR and ER expression differed among ovarian cancer subtypes. The proportion of tumors that stained positive (weak or strong) for PR was highest for endometrioid carcinoma (67%) and low-grade serous carcinoma (58%), intermediate for high-grade serous carcinoma (31%), and lowest for mucinous carcinoma (17%) and clear-cell carcinoma (8%).
More tumors stained positive for ER than for PR in all subtypes, with the proportion of ER-positive tumors being highest for low-grade serous carcinoma (87%) and high-grade serous carcinoma (81%) and lowest for mucinous carcinoma (21%) and clear-cell carcinoma (20%). For ER-positive tumors, coexpression of PR was most likely for endometrioid carcinoma (82%) and least likely for high-grade serous carcinoma (34%) and clear-cell carcinoma (32%).
The proportion of tumors that were ER- or PR-positive or both was highest for low-grade serous carcinoma (91%), high-grade serous carcinoma (84%), and endometrioid (82%) tumors and lowest for mucinous carcinoma (23%) and clear-cell carcinoma (21%).
Associations With Survival
Both PR expression (log-rank P < .0001) and ER expression (log-rank P < .0001) were associated with improved disease-specific survival in endometrioid carcinoma. PR expression (log-rank P = .0006) but not ER expression was associated with improved disease-specific survival in high-grade serous carcinoma. There were no significant survival differences for low-grade serous or mucinous carcinomas; ER expression seemed to be associated with worse survival in clear-cell carcinoma, but the association was not significant after adjustment for other clinical covariates.
In analysis adjusting for site, age, stage, and grade, strong (hazard ratio [HR] = 0.71, P = .0061) but not weak PR expression was associated with significantly improved high-grade serous carcinoma survival, with similar associations observed after adjustment for residual disease in the subset of patients for whom such data were available. Strong PR expression was associated with improved low-grade serous carcinoma survival (HR = 0.39, P = .019), but the association was no longer significant after adjustment for residual disease.
Both weak (HR = 0.61, P = .047) and strong (HR = 0.38, P = .0005) expression of PR and weak (HR = 0.40, P = .0041) and strong (HR = 0.41, P = .0001) expression of ER were associated with improved endometrioid carcinoma survival, with similar associations observed after adjustment for residual disease. Overall, positive hormone receptor expression (weak or strong PR or ER expression or both) was associated with improved endometrioid carcinoma survival (HR = 0.33, P < .0001). The best prognostic model for endometrioid carcinoma (identified by a comparison of model likelihoods) combined women with hormone receptor–positive tumors (PR or ER or both) into the low-risk group.
A summarized by the authors, “We have shown that patients with endometrioid carcinoma who have hormone-receptor-positive tumors are less likely to die from their disease than are those with hormone-receptor-negative tumors. Furthermore, we established that patients with [high-grade serous carcinoma] whose tumors stain strongly positive for PR have improved survival compared with those whose tumors stain negative or weakly positive for PR.”
They further noted that the magnitude of these effects is similar to the protective effect of germline BRCA mutations on ovarian cancer survival and that the effects were stronger than the observed associations of grade and extent of residual disease with survival in endometrioid carcinoma.
The authors concluded, “PR and ER are prognostic biomarkers for endometrioid and high-grade serous ovarian cancers. Clinical trials, stratified by subtype and biomarker status, are needed to establish whether hormone-receptor status predicts response to endocrine treatment, and whether it could guide personalized treatment for ovarian cancer.” ■
Disclosure: For full disclosures of the study authors, visit www.thelancet.com.
1. Sieh W, Köbel M, Longacre TA, et al: Hormone-receptor expression and ovarian cancer survival: An Ovarian Tumor Tissue Analysis consortium study. Lancet Oncol 14:853-862, 2013.