AbstractIntroduction Ovarian cancer is a chemosensitive tumour, but two thirds of women have a recurrence during the fol-low-up, even after an optimal surgical debulking followed by chemotherapy with a platinum and a taxane compound. Cytotoxic drugs are used in a second- or third-line set-ting but tumour progression is the rule. Also patients with the same histology achieve different outcomes in terms of survival. We decided to study gonadotropin and steroid re-ceptors and to consider if these histological markers could select patients with different prognosis.Materials and methods In our study we have measured by immunohistochemistry oestrogen, progestin and gonado-tropin-releasing hormone receptors (Gn-RHRs) in paraffi n-embedded ovarian cancer tissue in a sample of 62 con-secutive patients with advanced ovarian cancer treated with surgery and adjuvant chemotherapy. Descriptive methods, a survival analysis (Kaplan–Meier) and a Cox regression analysis were done.

Results Oestrogen receptors (ERs) were positive in 65% of patients and the same positivity was obtained for progestin receptors (PRs), with 74% showing some positivity for Gn-RHR receptors. Maximal cytoreduction and ERs, but not gonadotropin receptors, were independently associated with overall survival, with better survival for oestrogen-negative tumours. No association was established for pro-gression-free survival.Conclusions We can conclude that ER status in our series is an independent prognostic factor for ovarian cancer with better survival for oestrogen-negative receptor tumours. PRs could also have a prognostic role in association with ERs.

Keywords Ovarian cancer · Prognostic factor ·Oestrogen receptor · Gonadotropin-releasinghormone receptor

Introduction

Advanced epithelial ovarian cancer is still a lethal disease, despite radical surgery followed by chemotherapy with a platinum compound and a taxane [1, 2]. Several prognostic factors have been clearly identifi ed such as stage, histologi-cal grade and the size of residual tumour after the initial surgery [3–6], but there are no biological markers to pre-dict long-term survival or to be implemented as a therapeu-tic target. Even though the ovarian gland is under strong infl uence from oestrogen, progestin and gonadotropin hor-mones, epidemiological studies have not found any rela-tionship between the use of hormonal replacement therapy [7–10] or the use of fertility drugs and ovarian cancer inci-dence. Oestrogen, progestin and gonadotropin receptors are

L. Alonso (�) · A. Sánchez-Muñoz · E. Torres · B.I. Pajares ·S. Leefl ang · C. BahaMedical Oncology ServiceHospital Universitario Virgen de la VictoriaCampus de Teatinos, s/nES-29010 Málaga, Spaine-mail: ljalonso@uma.es

E. GallegoPathology DepartmentHospital Universitario Virgen de la VictoriaMálaga, Spain

F.J. GonzálezFundación IMABISMálaga, Spain

Clin Transl Oncol (2009) 11:748-752DOI 10.1007/s12094-009-0437-4

R E S E A R C H A R T I C L E S

Gonadotropin and steroid receptors as prognostic factorsin advanced ovarian cancer: a retrospective study

Lorenzo Alonso · Elena Gallego · Francisco Jesús González · Alfonso Sánchez-Muñoz ·Esperanza Torres · Bella Isabel Pajares · Stephanie Leefl ang · Camelia Baha

Received: 30 March 2009 / Accepted: 14 September 2009

Clin Transl Oncol (2009) 11:748-752 749

expressed in ovarian cancer tissue and they play a role as a stimulus for angiogenesis, growth and interference with an apoptotic mechanism [11–14].

The percentage of response to hormonal manipulation is around 20% in ovarian cancer, but most of the studies have been implemented in heavily treated patients. Dif-ferent approaches have been used such as gonadotropin agonist or tamoxifen and aromatase inhibitors [15–20]. No signifi cant association between tumour response and hormone receptor expression has been demonstrated, but at least one experimental study has established an association between endocrine response and insulin-like growth factor binding proteins [21].

In this study, we measure the status of oestrogen recep-tor (ER), progestin receptor (PR) and gonadotropin-releas-ing hormone receptor (Gn-RHR) in a group of advanced disease ovarian cancer patients to evaluate the possible role of these markers as prognostic factors.

Materials and methods

Patients and study design

We conducted a retrospective study in 62 patients diag-nosed with advanced epithelial ovarian cancer (stages IIB through IV) with fi rst diagnosis between the years 2001 and 2007. The sample for the study was obtained using consecutive patients treated in a single institution.

All patients had a histological sample of the tumour and were treated with maximal cytoreduction under the rules established in ovarian cancer surgery when technically possible. After surgery, patients received chemotherapy mainly with a platinum compound alone or in combination with a taxane. No borderline tumours or recurrence as fi rst diagnosis were included. We retrieved information directly from the medical records including demographic variables of the patients, tumour variables and other possible fac-tors related to the prognosis. ERs, PRs and Gn-RHRs were measured retrospectively by immunohistochemistry. The cut-off point to differentiate ER positive (ER+) from ER negative (ER–) tumours was established at the 5% percent-age of stained cells.

Immunohistochemical staining

Consecutive sections of paraffi n-embedded ovarian can-cer tissue were cut, mounted on xylane-coated slides, deparaffi nised in xylene and rehydrated in a graded se-ries of alcohols. Heat-induced epitope retrieval was done with the pressure cooker method in TBS buffer (pH 7.6). Endogenous peroxidase activity was blocked by incuba-tion in 1% hydrogen peroxide in methanol for 30 min. Immunohistochemical assays with the antibodies mouse monoclonal anti-human ER (clone 1D5, prediluted, Dako,

Copenhagen, Denmark), mouse monoclonal anti-human progesterone receptor (clone PgR 636, prediluted, Dako, Copenhagen, Denmark) and mouse monoclonal anti-human Gn-RHR (clone A9E4, diluted 1:5 overnight, No-vocastra, Newcastle, UK) were done with an automated staining system (Tech Mate Horizon, Dako, Copenhagen, Denmark) using the Dako Real EnVision method to detect the antibody. Negative controls were routinely carried out by substitution of the primary antibody by Tris-buffered saline.

Statistical analysis

Initially a descriptive analysis was done, using mean values for quantitative variables and frequencies for qualitative variables. Univariate analysis was done using a two-by-two table analysis, categorising quantitative variables and under the Chi-squared distribution with a level of signifi cance p<0.05. A survival analysis was also done using the Ka-plan–Meier method and the log-rank test for signifi cance. A Cox regression analysis (Backward, Wald, inclusion criteria p<0.05, removal p>0.10) was also done to detect the independent contribution of each variable in the model to survival. The variables initially included were age, stage, histological grade, type of surgery, ER, PR, Gn-RHR sta-tus, histology and menopausal situation.

Results

Patient characteristics

The median follow up for the whole series was 27 months. Among the 62 patients included, 45 (73%) were still alive at the closing time of the study and 35 patients (56%) were disease-free. Median overall survival was 58 months and median disease-free survival 28 months. Maximum follow-up was 77 months.

One third of the patients were treated with maximal cytoreduction. Two thirds of the patients (68%) were post-menopausal. Table 1 shows the patient characteristics.

Immunohistochemistry: gonadotropin, oestrogen and pro-gestin receptors

The majority of tumours (65%) were ER positive and the same percentage were PR positive. The expression of some kind of positivity (gradation 1, 2 or 3 vs. 0) for gonadotro-pin receptor was high, with 74% of tumour samples ex-pressing this histological marker. No association was found between menopausal status or histological type and status of ERs or PRs. Table 2 shows patients characteristics ac-cording to ER/PR status.

750 Clin Transl Oncol (2009) 11:748-752

Prognostic factors

The variables associated with overall survival were type of surgery, stage and ER status. No variables were signifi -cantly associated with disease-free survival. Figure 1a and

b shows the overall survival and progression-free survival curves for ER status. Progestin or gonadotropin receptor status was not associated with overall survival or progres-sion-free survival. Figure 2 shows the overall survival plot considering together the different combinations of ER and PR status, with better survival for the combination ER(–) PR(+), posing the question of a modulating role for PR on ER activity. In the multivariate analysis the variables independently associated with overall survival were type of surgery and ER status, with a probability of higher survival for patients with ER-negative tumours in relation to ER-positive tumours. The level of signifi cance and the coeffi -cients for these two variables are displayed in Table 3.

Discussion

The percentage of tumours with ER expression (65% in our study) and PR is similar to the data published in the literature [22, 23]. In our study 74% of tumour samples show positivity for Gn-RHR, which is in accordance with previous studies where the expression of Gn-RHR type I is around 80% in human ovarian epithelial tumours [24, 25].

The fi nding of a clear benefi t in survival for patients treated with maximal cytoreduction has been established in ovarian cancer as the main prognostic factor [26]. Less common is the finding of a better survival for patients with ER-negative epithelial ovarian cancer in relation to ER-positive tumours, as we found in our patients. This could be a result obtained by chance or related to a differ-ent behaviour of tumour cells depending on the ER status. Previous studies have evaluated the possible prognostic role of ER in ovarian cancer. Fujimoto et al. [27], in a se-ries of 60 patients, found a better survival for patients with low expression of ER alfa-mRNA in relation to patients with high expression. Münstedt et al. [28] studied 186

Table 1 Clinicopathological characteristics of the patients

Variables N %

Patients 62 Age (median, years) 56 Histological type Serous papillary 45 72 Mucinous 3 5 Clear cell 3 5 Undifferentiated 11 18FIGO stage IIB–III 49 79 IV 13 21Grade 1 5 8 2 18 29 3 39 63Type of surgery Optimal 22 36 Non-optimal 40 64Chemotherapy Carboplatin/taxol 35 56 Carboplatin 3 5 Others 24 39ER Negative 22 35 Positive 40 65PR Negative 22 35 Positive 40 65Gn-RHR Negative 16 26 Positive 46 74

Table 2 Patient characteristics in relation to ER/PR status

Variables ER(–) PR(–) ER(+) PR(+) ER(+) PR(–) ER(–) PR(+)

N (%) N (%) N (%) N (%)

Age <55 4 (33) 15 (50) 4 (40) 8 (80) >55 8 (67) 15 (50) 6 (60) 2 (20)Histology Serous 6 (50) 26 (86.5) 5 (50) 7 (70) Others 6 (50) 4 (13.5) 5 (50) 3 (30)Grade (1+2) 5 (41) 14 (46.5) 3 (30) 1 (10) 3 7 (59) 16 (53.5) 7 (70) 9 (90)Gn-RHR Positive 9 (75) 19 (63) 5 (50) 6 (60) Negative 3 (25) 11 (37) 5 (50) 4 (40)Surgery Optimal 4 (33) 13 (43) 1 (10) 4 (40) Non-optimal 8 (67) 17 (57) 9 (90) 6 (60)

Clin Transl Oncol (2009) 11:748-752 751

patients with ovarian cancer and found a better survival for PR-expressing tumours, but this benefi t was only main-

tained in the group of ER-negative patients, with the best outcome for the combination ER(–) PR(+). In contrast, the MALOVA study [29] found a better prognosis for ER(+) tumours but low potential malignant tumours were also included. Recently, Arias-Pulido et al. found, in a series of 89 patients with ovarian cancer including both initial and advanced disease, a superior clinical outcome for tumours with ER(–) PR(+) expression [30].

The rationale for a worse prognosis for ER(+) tumours could be related to the increase in proliferation, reduc-tion of apoptosis and modulation of activity of EGFR and HER2/neu that oestrogen stimulus produces at the molecu-lar level [22, 31]. There is evidence that oestrogens can be produced intratumour after an in situ aromatisation and its mitogenic activity could be independent from circulating oestrogen [32].

The association between ER status and survival raises the possibility of different subsets of ovarian cancer pa-tients with different biological behaviour but similar histol-ogy or different response to cytotoxic drugs. The pattern of ER and PR expression has been clearly associated in breast cancer with mutation in the BRCA1 gene, with prognostic and therapeutic implications [33]. Moreover, population studies in ovarian cancer have confi rmed the association in

Fig. 1a Overall survival for status of ER. b Interval-free disease for status of ER

Time (months)

Pro

babi

lity

of s

urvi

val

A B

Fig. 2 Overall survival for subsets of oestrogen and progesterone status receptors

Table 3 Final model for Cox regression analysis

Variable Explanation Signifi cance (p) Hazard ratio CI 95%

ER Negative vs. positive 0.006 9.95 (1.9–51)Type of surgery Optimal vs. non-optimal 0.009 8.95 (1.7–46)Age <55 vs. >55 0.69 1.25 (0.4–3.9)PR Negative vs. positive 0.20 0.98 (0.96–1)Gonadotropin receptor Negative vs. positive 0.66 0.87 (0.48–1.57)Histology Serous vs. others 0.91 0.92 (0.23–3.6)Stage III vs. IV 0.69 1.36 (0.28–6.5)Menopause Pre vs. post 0.08 0.30 (0.07–1.19)

752 Clin Transl Oncol (2009) 11:748-752

terms of better survival for tumours with BRCA1 mutations [34], posing the question of different subtypes of ovarian cancer based not only on morphological aspects.

We can conclude that in our series of advanced epithe-lial ovarian cancer, ER status is an independent prognostic factor, with a better survival for ER-negative tumours. This could express a different biological behaviour and different response to treatment, probably related to biological vari-

ables in the tumour, raising the possibility of different sub-sets of ovarian cancer that share the same histology. In the near future a prospective study will be implemented trying to describe an association between ER, PR and genetic or proteomic biomarkers of proliferative activity specifi c to ovarian cancer.Confl ict of interest The authors declare that they have no confl ict of interest relating to the publication of this manuscript.

References1. Piccart MJ, Bertelsen K, James K et al (2000)

Randomized intergroup trial of cisplatin-paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: three-year results. J Natl Cancer Inst 92:699–708

2. McGuire WP, Hoskins WJ, Brady MF et al (1996) Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 334:1–6

3. Ozols RF, Bundy BN, Greer BE et al (2003) Phase III trial of carboplatin and paclitaxel com-pared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 21:3194–3200

4. Bristow RE, Tomacruz RS, Armstrong DK et al (2002) Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. Clin Oncol 20:1248–1259

5. Hoskins WJ, Bundy BN, Thigpen JT et al (1992) The infl uence of cytoreductive surgery on recur-rence-free interval and survival in small-volume stage III epithelial ovarian cancer: a Gynecologic Oncology Group study. Gynecol Oncol 47:159–166

6. Hoskins WJ, McGuire WP, Brady MF et al (1994) The effect of diameter of largest residual disease on survival after primary cytoreductive surgery in patients with suboptimal residual epithelial ovari-an carcinoma. Am J Obstet Gynecol 170:974–980

7. Riman T, Dickman PW, Nilsson S et al (2002) Hormone replacement therapy and the risk of in-vasive epithelial ovarian cancer in Swedish wom-en. J Natl Cancer Inst 94:497–504

8. Folsom AR, Anderson JP, Ross JA (2004) Es-trogen replacement therapy and ovarian cancer. Epidemiology 15:100–104

9. Riman T, Dickman PW, Nilsson S et al (2002) Risk factors for invasive epithelial ovarian cancer: results from a Swedish case-control study. Am J Epidemiol 156:363–373

10. Sit AS, Modugno F, Weissfeld JL et al (2002) Hormone replacement therapy formulations and risk of epithelial ovarian carcinoma. Gynecol On-col 86:118–123

11. Leung PC, Cheng CK, Zhu XM (2003) Multi-fac-torial role of GnRH-I and GnRH-II in the human ovaries. Mol Cell Endocrinol 202:145–153

12. Minegishi T, Kameda T, Hirakawa T et al (2000) Expression of gonadotropin and activin receptor messenger ribonucleic acid in human ovarian epi-thelial neoplasms. Clin Cancer Res 6:2764–2770

13. Furui T, Imai A, Tamaya T (2002) Intratumoral level of gonadotropin-releasing hormone in ovar-ian and endometrial cancers. Oncol Rep 9:349–352

14. Konishi I, Kuroda H, Mandai M (1999) Review: gonadotropin and development of ovarian cancer. Oncology 57[Suppl 2]:45–48

15. Du BA, Meier W, Luck HJ et al (2002) Chemo-therapy versus hormonal treatment in platinum- and paclitaxel-refractory ovarian cancer: a ran-domised trial of the German Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) Study Group Ovarian Cancer. Ann Oncol 13:251–257

16. Paskeviciute L, Roed H, Engelholm A (2002) No rules without exception: a long-term complete remission observed in a study using a LH-RH agonist in platinum-refractory ovarian cancer. Eur J Cancer 138[Suppl 6]:S73

17. Balbi G, Piano LD, Cardone A et al (2004) Sec-ond-line therapy of advanced ovarian cancer with GnRH analogs. Int J Gynecol Cancer 14:799–803

18. Perez-Gracia JL, Carrasco EM (2002) Tamoxifen therapy for ovarian cancer in the adjuvant and advanced settings: systematic review of the litera-ture and implications for future research. Gynecol Oncol 84:201–209

19. Bowman A, Gabra H, Langdon SP et al (2002) CA125 response is associated with estrogen re-ceptor expression in a phase II trial of letrozole in ovarian cancer: identifi cation of an endocrine-sensitive subgroup. Clin Cancer Res 8:2233–2239

20. Papadimitriou CA, Markaki S, Siapkaras J et al (2004) Hormonal therapy with letrozole for re-lapsed epithelial ovarian cancer. Long-term results of a phase II study. Oncology 66:112–117

21. Walker G, MacLeod K, Williams ARW et al (2007) Insulin-like growth factor binding proteins IGFBP3, IGFBP4, and IGFBP5 predict endocrine responsiveness in patients with ovarian cancer. Clin Cancer Res 13:1438–1444

22. Lindgren P, Bäckström T, Mählck CG et al (2001) Steroid receptors and hormones in relation to cell proliferation and apoptosis in poorly differentiated epithelial ovarian tumors. Int J Oncol 19:31–38

23. Lee P, Rosen DG, Zhu C et al (2005) Expression of progesterone receptor is a favorable prognostic

marker in ovarian cancer. Gynecol Oncol 96:671–677

24. Völker P, Gründker C, Schmidt O (2002) Expres-sion of receptors for luteinizing hormone-releas-ing hormone in human ovarian and endometrial cancers: frequency, autoregulation, and correlation with direct antiproliferative activity of luteinizing hormone-releasing hormone analogues. Am J Ob-stet Gynecol 186:171–179

25. Imai A, Ohno T, Lida K et al (1994) Gonadotro-pin-releasing hormone receptor in gynecologic tumors. Frequent expression in adenocarcinoma histologic types. Cancer 74:2555–2561

26. Hoskins WJ, Bundy B, Thigpen JT et al (1992) The infl uence of cytoreductive surgery on recur-rence-free interval and survival in small-volume stage III epithelial ovarian cancer: a Gynecologic Oncology Group study. Gynecol Oncol 47:159–166

27. Fujimoto J, Alam SM, Jahan I et al (2007) Clini-cal implication of estrogen-related receptor (ERR) expression in ovarian cancers. J Steroid Biochem Mol Biol 104:301–304

28. Münstedt K, Steen J, Knauf AG et al (2000) Ste-roid hormone receptors and long term survival in invasive ovarian cancer. Cancer 89:1783–1791

29. Hogdall EV, Christensen L, Hogdall CK et al (2007) Prognostic value of estrogen receptor and progesterone receptor tumor expression in Dan-ish ovarian cancer patients: from the ‘MALOVA’ ovarian cancer study. Oncol Rep 18:1051–1059

30. Arias-PulidoH, Smith HO, Joste NE et al (2009) Estrogen and progesterone receptor status and outcome in epithelial ovarian cancers and low ma-lignant potential tumors. Gynecol Oncol 114:480–485

31. Bookman MA (2005) Is there still a role for hor-monal therapy? Int J Gynecol Cancer 15[Suppl 3]:291–297

32. Li YF, Hu W, Fu SQ et al (2008) Aromatase inhibitors in ovarian cancer: is there a role? Int J Gynecol Cancer 18:600–614

33. Brekelmans CT, Seynaeve C, Menke-Pluymers M et al (2006) Survival and prognostic factors in BRCA1-associated breast cancer. Ann Oncol 17:391–400

34. Chetrit A, Hirsh-Yechezkel G, Ben-David Y et al (2008) Effect of BRCA1/2 mutations on long-term survival of patients with invasive ovarian cancer: the national Israeli study of ovarian can-cer. J Clin Oncol 26:20–25