Formalin-fixed, paraffin-embedded ovarian tissues were harvested from 110 cases of primary epithelial ovarian carcinoma and 24 benign ovarian tumor surface epithelium samples. The patients with ovarian tumors underwent surgery from January 2003 to December 2007 at the Affiliated Hospital of Nantong University (Nantong, China). The 10 healthy individual ovary samples were also harvested for comparison. The clinical stage of all the patients with EOC was determined using the FIGO staging system (23); 64 cases were low stage tumors (I–II) and 46 were high stage tumors (III–IV). Neither preoperative radiation nor chemotherapy had been received by the patients with EOC. Clinical information on each patient, including age, histological type, grade based on the World Health Organization (WHO) criteria (24), FIGO stage and tumor size, was collected from medical records, including from a 5-year follow-up. The mean age of the patients was 54.5 years (range, 29–78 years). The protocol of the present study was approved by the Institutional Review Board at the Affiliated Hospital of Nantong University. Informed written consent was obtained for each patient.

EOC fresh-frozen at −80°C and paired normal ovarian epithelium samples (n=16 of each) were obtained from January 2011 to August 2012 at the Affiliated Hospital of Nantong University. The mean age of the patients was 34.5 years (range, 23–64 years). TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) was used to extract total RNA from the frozen samples. Total RNA was then reverse transcribed using a RevertAid™ First Strand cDNA Synthesis kit (Fermentas; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol at 42°C for 30 min. For RT-qPCR, the analysis of mRNA levels was performed using SYBR-Green Reagents (Toyobo Life Science, Osaka, Japan) using an iQ5 Multicolor Real-time PCR Detection System, and all mRNA levels were normalized to GAPDH (n=3). The FOXA1 primer sequences were as follows: Forward, 5′-GTTGAAGACTCCAGCCTCCTC-3′ and reverse, 5′-CTGCCCAGAACATCATCCCT-3′. GAPDH primers sequences were as follows: Forward, 5′-CGGAGTCAACGGATTTGGTCGTAT-3′ and reverse, 5′-AGCCTTCTCCATGGTGGTGAAGAC-3′ (Invitrogen; Thermo Fisher Scientific, Inc.). GAPDH mRNA levels were used as an internal control following melt curve analysis (25). Amplification conditions were as follows: Taq activation at 94°C for 2 min; 35 cycles of 94°C for 20 sec, 58°C for 20 sec andelongation at 72°C for 30 sec.

The 4 µm sections were deparaffinized using a graded ethanol series (Xylene 3 times for 3 min, then 100, 95 and 70%, 3 times for 2 min each), and 0.3% hydrogen peroxide was used to block endogenous peroxidase activity at room temperature for 15 min. For antigen retrieval, the sections were placed in 10 mM citrate buffer (pH 6.0) and heated to 121°C in an autoclave for 20 min. Goat serum (10%; Sangon Biotech Co., Ltd., Shanghai, China) was used to block non-specific reactions for 1 h at room temperature. The sections were then rinsed in phosphate-buffered saline (pH 7.2) and incubated with anti-human FOXA1 antibody (dilution, 1:200; cat. no. ab23738; Abcam, Cambridge, MA, USA) overnight at 4°C. Then, the sections were incubated with secondary antibody (dilution 1:1,000; cat. no. ab205718; Abcam, Cambridge, MA, USA) at room temperature for 30 min. Sections incubated without antibody were used as the negative control. All slides were processed using the peroxidase anti-peroxidase method (Dako; Agilent Technologies, Inc., Santa Clara, CA, USA). The sections were counterstained with hematoxylin, dehydrated, and coverslips were added subsequent to rinsing with water. The stained sections were observed under a phase contrast microscope (magnification, ×400). All immunostained sections were assessed blindly. High-power fields (n=5) for each specimen were randomly selected to assess FOXA1 expression, and nuclear staining was observed under high-power magnification. To determine the mean percentage of immunostained cells, >500 cells were counted. The analysis was repeated twice in half of the samples to decrease the likelihood of technical errors; the results of these repeats were similar. If the nuclei were stained >5% and the plasma was not, FOXA1 protein expression was considered positive. Staining of <5% of the cells was judged as negative, 5–30% as weak (low expression), 31–70% as moderate and >70% as strong (high expression).

SPSS 18.0 software (SPSS, Inc., Chicago, IL, USA) was used to perform statistical analysis. Mean ± standard deviation was used to represent the results. To compare clinicopathological data, the Fisher's exact test, the χ² test, and two-sample t-tests were used. Overall survival time was defined as the interval between primary surgery and patient mortality or the final follow-up. Survival curves were plotted using the Kaplan-Meier method. The association between clinical characteristics and survival time in patients with EOC was assessed using the log rank test. P<0.05 was considered to indicate a statistically significant difference.

In the present study, total RNA was extracted from paired, frozen EOC and normal ovarian epithelium tissues, and subjected to RT-qPCR to measure FOXA1 mRNA expression, normalized to GAPDH mRNA level. FOXA1 expression was increased in EOC specimens compared with the corresponding normal tissue (Fig. 1A and B). The mean relative expression of FOXA1 mRNA in EOC and the corresponding benign ovarian tumor surface epithelium tissues was 1.389±0.1225 and 1.029±0.0626, respectively, indicating a significant difference (P=0.014; Fig. 1C).

To confirm the increased expression of FOXA1 in EOC tissues compared with non-cancerous tissues, immunohistochemical staining was performed. FOXA1 expression was primarily observed in the nuclei and not inthe cytoplasm (Fig. 2). As revealed by immunohistochemical analysis, FOXA1 expression was upregulated in EOC. Only 10.0% of the normal ovarian tissues demonstrated moderate or strong positive expression of FOXA1: ~20.8 and 73.6% of the benign ovarian tumor surface epithelium tissues and EOC specimens, respectively, revealed moderate or strong positive expression of FOXA1 (Table I). Furthermore, the percentage of FOXA1 positive cells increased with an increasing WHO grade in EOC tissue: 84.8% of high-grade (III–IV) and 65.6% of low-grade (I–II) OC tissues exhibited moderate or strong expression of FOXA1 (Table I). Therefore, the total expression of FOXA1 in EOC tissues was significantly increased compared with non-cancerous tissues (P<0.001), and increased with increasing tumor grade (P=0.024) (Table II). The results indicated that the mean value of FOXA1 immunoreactivity in benign ovarian tumor surface epithelium tissues was not significantly differentcompared with normal ovarian tissues (P=0.450).

To determine the association between FOXA1 expression and clinicopathological characteristics in EOC, the clinical data of 110 patients with EOC were analyzed (Table II). Increased expression of FOXA1 in EOC was significantly associated with the tumor WHO grade (P=0.024) and differentiation status (P=0.003). There was no significant association between FOXA1 expression and age, histological subtype, tumor diameter or location. Furthermore, increased FOXA1 expression in non-cancerous epithelial ovarian tissues was not significantly associated with the clinical characteristics of EOC (data not shown). However, the overall survival time of the patients in the FOXA1 low expression and high expression groups differed significantly (P=0.013) (Fig. 3). The patients with low expression of FOXA1 exhibited increased overall survival time compared with those with high expression of FOXA1, indicating that FOXA1 expression may exhibit prognostic value for patients with EOC.