Hormone–receptor expression status of epithelial ovarian cancer in Ibadan, South-western Nigeria

Introduction Epidemiological evidence strongly suggests that steroid hormones are implicated in the pathogenesis of ovarian cancer. Estrogen receptor (ER) and Progesterone receptor (PR) are prognostic indicators for a number of epithelial tumors and may play the same role in ovarian cancers. This study aims to evaluate the expression of ER and PR in epithelial ovarian cancer (EOC) in an African population and compare it with other prognostic factors such as age, International Federation of Gynaecology and Obstetrics (FIGO) stage, grade and histological subtype. Methods Ninety cases of histologically confirmed EOC were reviewed. Immunohistochemistry was used to assess their ER and PR expression status and was then compared with other demographic variables using statistical methods, with level of significance set at p < 0.05. Results 30.2% and 8.3% of serous and mucinous carcinomas respectively were ER positive while 41.2% and 22.5% of both tumour types were PR positive. One of the two endometrioid carcinomas showed PR expression but neither were positive for ER. The only case of Brenner tumour in the series was ER positive but negative for PR. There was a significant association between ER and the histological subtypes (p = 0.042) while no significant association was found between PR expression and histological subtypes (p = 0.650). No significant association was found between hormone receptor status, age and stage of the EOC. Conclusion The study showed a lower ER expression in serous carcinoma compared to large cohorts from developed countries. Future translational studies could be used to determine response of EOC to endocrine therapy.


Introduction
Ovarian cancer causes more than 140,000 deaths worldwide every year and is the most lethal gynaecological malignancy in developed countries [1]. In the USA, this neoplasm ranks second among gynaecological cancers, yet it is by far the most lethal one, accounting for more than 15,000 deaths annually [2,3]. Ovarian cancer is often known as "the silent killer," its early detection is difficult because the ovaries are deep within the pelvis and initial symptoms are often ambiguous. The cancer goes undiagnosed until after the disease is far advanced and has spread throughout the abdomen or to distant organs. Nearly 75% of cases are diagnosed at an advanced stage, despite great efforts to develop reliable screening and prevention strategies [4,5]. Majority of patients with ovarian carcinoma eventually succumb to disease due to development of treatment resistance. The introduction of cisplatinbased chemotherapy was initially thought to show improvement in the survival of patients with ovarian cancer, the percentage of recurrent disease is still high even in those patients who achieve a complete response to chemotherapy, so that more than 80% of patients with advanced stage of disease in Africa die within 5 years [2,6,7]. More effective treatment regimens and better early detection tools must be developed to improve morbidity and reduce mortality in ovarian cancer patients [3,7,8]. This has led to the development of "targeted" oncologic therapies that might ultimately be more effective and less toxic. Steroid hormone receptors expression in epithelial ovarian cancers has been proposed to have therapeutic and prognostic relevance, as is the case in breast cancer [7]. At present, the prognostic characterisation of ovarian cancer patients, based on clinicopathological parameters such as stage, histology, grade and residual tumour after surgery, seems to be inadequate, since patients with similar clinicopathological characteristics often experience different clinical outcome. Therefore, the identification of biological factors related to tumour aggressiveness could be relevant in order to identify patients with different prognosis and chance to respond to chemotherapy, thus allowing the selection, at the time of initial diagnosis, of high risk patients needing more aggressive therapy or alternative treatment and a closer follow-up [3,8].
Among the biological parameters proposed as possible prognostic factors in ovarian cancer, much attention has been focused on endocrine factors and especially on steroid hormones and their receptors (estrogen and progesterone receptors). Epithelial tumours are most frequently associated with nulliparity, early menarche, late age at menopause and a long estimated number of years of ovulation [9,10]. Different theories have been proposed as causal relationship between ovarian stimulation and neoplasia including Fathalla's incessant ovulation hypothesis which was initially thought to explain the pathogenesis of the tumour [11]. Recently however studies have shown that ovarian carcinomas arise via two main separate pathways, the high grade serous tumours are now known to arise from the fallopian tube fimbrial precursor lesions implanted on the ovarian surface [12][13][14]. Low grade tumours such as endometrioid carcinomas are believed to arise from areas of endometriosis [12][13][14]. Steroid hormones, primarily oestrogen and progesterone have been implicated in ovarian carcinogenesis.
Oestrogens are major regulators of growth and differentiation in normal ovaries. Loss of heterozygosity at the 11q23. 3 [18]. There has been no work done on the expression of ER and PR in ovarian cancers seen in females in Nigeria. This study aims to contribute to the body of knowledge by evaluating the expression of ovarian cancers in females seen in our environment. Breast tissue was used as a positive control. Slides were reviewed by two of the authors (MAA and AOO) and discordant scores were resolved through a consensus scoring. The agreement among the observers was 95%. Only nuclear staining of the tumour cells was considered a positive expression for ER and PR ( Figure 1). Grading of nuclear ER and PR staining was performed using an immunoreactive H-scoring system obtained by the product of intensity of immunostaining (none = 0 (negative); 1-25% = 1+ (weak); 26-50%=2+ (moderate); > 50% = 3+(strong)). The ER and PR positivity was defined as ≥ 1% tumour cell nuclei (i.e. encompassing weak, moderate and strong nuclear staining) [22].

Methods
The data obtained were subjected to statistical analysis using

Results
The age range of patients from which the surgical specimens were obtained was from 16 years to 82 years with a mean age of 52.2 ± 12.6 years. The peak age of occurrence of epithelial ovarian cancers was in the fifth decade of life. There was only one case of epithelial ovarian cancer seen in the first two decades of life. Serous carcinomas was the most common histological subtype and was seen in 63 (70.0%) cases, followed by mucinous carcinoma, which accounted for 24 (26.7%)cases, There were 2 (2.2%) endometrioid carcinomas and only 1(1.1%) case of malignant Brenner tumour (Table 1). The poorly differentiated (grade 3) serous carcinomas were more commonly seen while the reverse is the case in mucinous carcinomas with grade 1 tumours being more common (  23,24,29,32,33]. In the current study, a significant positive correlation was found between tumour grade and PR expression, with 60% of PR positive epithelial ovarian cancers being grade 3 (poorly differentiated) tumours, and 10% being grade 1 (well differentiated) tumours. This finding is in agreement with some studies [29,38,39] [26,39]. There was no association between PR expression and FIGO stage. These findings are in agreement with some reports in the literature [29,39]   Progesterone receptor was found to be more expressed in grade 3 tumours than grade 1 tumours whereas there was no significant association between the grade and estrogen expression.

Competing interests
The authors declare no competing interest.

Authors' contributions
Mustapha Akanji Ajani conceived and designed the study. All authors carried out the research and contributed to the experimental design and preparation of the manuscript. All authors were involved in the revision of the draft manuscript and have agreed to the final content.

Acknowledgments
We appreciate Ajagboye for the technical assistance he rendered with the slides used for this study and Abayomi Odetunde who worked tirelessly in making sure the immunohistochemical staining on the cases was successfully carried out.