Patients and tissue specimens

The inclusion criteria for patients were as follows: (1) a pathologically confirmed diagnosis of cervical squamous cell cancer and an International Federation of Obstetrics and Gynecology (FIGO) stage between IB1 and IIB; (2) initial treatment including radical hysterectomy and bilateral pelvic lymph node dissection were performed at Sun Yat-sen University Cancer Center between January 2009 and December 2011; (3) no preoperative anti-cancer therapy; and (4) no secondary cancer. Paraffin-embedded cervical squamous cell cancer specimens were obtained from the Pathology Department. Normal cervical epithelial samples (to be used as normal controls) were collected from patients who had benign uterine tumors and needed a hysterectomy during the same period. Patients’ hospital records were reviewed to obtain demographic data, including age, serum level of squamous cell carcinoma (SCC) antigen, tumor size and stage, surgical procedures, pathologic report, adjuvant therapy, and follow-up information.

Postoperative adjuvant radiotherapy was recommended for patients with the following pathologic risk factors: positive lymph nodes, deep cervical stromal invasion, positive margin, and/or lymphovascular space invasion. Postoperative chemotherapy was recommended for patients with positive lymph nodes or lymphovascular space invasion. The sequential chemotherapy regimens were as follows: irinotecan at 60–80 mg/m² on days 1 and 8 or paclitaxel at 175 mg/m² and cisplatin at 60–75 mg/m² on day 1. These doses were repeated every 3 weeks for 4 cycles. The concurrent chemotherapy regimen was cisplatin at a dose of 35–40 mg/m² weekly during radiotherapy. The total radiation dose was approximately 45–50 Gy with a fractionation of 1.8–2.0 Gy daily. The irradiation fields were mainly the pelvic cavity field with two exceptions of a para-aortic extended field.

Ten fresh cervical squamous cell cancer specimens (from 6 patients with FIGO stage IB1 tumor and 4 patients with FIGO stage IIA1 tumor; 2, 3, and 5 of the 10 patients had grade 3, 2, and 1 tumors, respectively) and their matched adjacent non-tumorous cervical tissues were collected immediately after resection and were immersed in RNAlater (Ambion, Austin, TX, USA) to prevent RNA degradation. The samples were stored at 4 °C overnight and then frozen at −80 °C before RNA and protein extraction for quantitative reverse transcription polymerase chain reaction (qRT-PCR) and Western blotting analysis.

qRT-PCR

Total RNA was extracted from frozen tissues using Trizol reagent (Invitrogen, Carlsbad, CA, USA) in accordance with the manufacturer’s instructions. A Nanodrop Spectrophotometer (ND-1000; Thermo Scientific, Wilmington, DE, USA) was used to assess the concentration and quality of the extracted total RNA. According to the manufacturer’s instruction, cDNA was synthesized using 2 μg RNA and M-MLV Reverse Transcriptase (Promega, Fitchburg, WI, USA). We used the ABI 7900HT Real-time PCR system (Life Technologies, Carlsbad, CA, USA) to perform gene amplification with the following reaction conditions: 94 °C for 5 min; 40 cycles at 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 50 s for primer extension; and 72 °C for 10 min. The housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. The primers for qRT-PCR were designed using the primer 5.0 software (PREMIER Biosoft International, Palo Alto, CA, USA) and had the following sequences: 5′-TACTTTGGGGAGGTCTTCGAG-3′ for forward PTK6 primer and 5′-TGCCGCAGCTTCTTCATG-3′ for reverse PTK6 primer; 5′-CTCCTCCTGTTCGACAGTCAGC-3′ for forward GAPDH primer and 5′-CCCAATACGACCAAATCCGTT-3′ for reverse GAPDH primer. The comparative cycle threshold (Ct) value was measured and used for data analysis.

Western blotting analysis

Total protein was extracted from frozen samples using Radio-Immunoprecipitation Assay Lysis Buffer, and the total protein concentration was measured using the Bicinchoninic Acid Protein Assay Kit (Bio-Rad, Hercules, CA, USA). Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and then electro-transferred onto a polyvinylidene difluoride (PVDF) membrane (Bio-Rad). The blotted membranes were blocked in 5% skim milk diluted in phosphate buffer solution supplemented with 0.1% Tween (PBST) for 1 h. After blocking the macromolecular antigen, the blotted membranes were incubated with anti-rabbit PTK6 antibody (1:1000 dilution; Proteintech, Wuhan, Hubei, China) followed by horseradish peroxidase (HRP)-conjugated secondary antibody (1:2000 dilution; Cell Signaling Technologies, Danvers, MA, USA) for 1 h. GAPDH protein (1:2000 dilution; Proteintech) was used as a loading control.

Immunohistochemistry

Paraffin-embedded cervical squamous cell cancer specimens were cut into 5-mm thick sections and then dewaxed with xylene. After rehydration in gradient alcohol solutions, these sections were submerged in ethylene diamine tetraacetic acid (EDTA) buffer and microwaved in a cooker at high pressure for 12 min followed by low pressure for 13 min for antigen retrieval. To quench endogenous peroxidase and non-specific binding, these sections were incubated in 3% hydrogen peroxide for 10 min followed by submergence in fetal bovine serum for 30 min at room temperature. The tissue sections were incubated with anti-PTK6 antibody (1:500 dilution) at 4 °C overnight followed by HRP-conjugated secondary antibody after washing with phosphate-buffered saline (PBS) five times (5 min each time). Subsequently, the visualization signal was developed with diaminobenzidine tetrahydrochloride (DAB) for 1 min and counterstained with hematoxylin.

The results of immunohistochemical (IHC) staining were scored independently by two pathologists regardless of patients’ clinical features. The IHC score was determined by both staining intensity and proportion of positively stained cancer cells. Intensity of staining score was categorized as 0, no staining; 1, weak staining; 2, moderate staining; and 3, strong staining. The percentage of stained tumor cell was scored as 0, <10%; 1, ≥10% to <25%; 2, ≥25% to <50%; 3, ≥50% to <75%; and 4, ≥75%. The final immunoreactivity score (IRS) was the product of staining intensity score and percentage score and ranged from 0 to 12. Cut-off values for PTK6 expression were based on the median of all products. An optimal cut-off value was determined as follows: ≤6 indicates low PTK6 expression, and >6 indicates high PTK6 expression.

Statistical analysis

Statistical Product and Service Solutions (SPSS) software package (version 17.0, SPSS Inc., Chicago, IL, USA) was used to perform statistical analysis. The Chi square test or Fisher’s exact test was used to determine the relationship between PTK6 expression levels and the clinicopathologic features of patients with cervical squamous cell cancer. Patients were followed up by either outpatient visit or telephone survey before January 31, 2015, and recurrence was diagnosed based on clinical and laboratory assessments. The overall survival (OS) of patients was defined as the time from initial surgery until the date of death or the last follow-up. The progression-free survival (PFS) of patients was determined as the time from initial surgery to recurrence or progression. Kaplan–Meier analysis was employed to plot survival curves, and the log-rank test was used to compare the differences between the survival curves. The prognostic effects of clinicopathologic variables were identified by univariate and multivariate Cox proportional hazards regression analysis. A P < 0.05 was considered statistically significant.

Clinical features of included patients with early-stage cervical squamous cell cancer

PTK6 mRNA and protein expression levels in early-stage cervical squamous cell cancer tissues

qRT-PCR results showed that the mRNA level of PTK6 in cancer tissues was higher than that in adjacent non-tumorous cervical tissues, and the range of fold increase was 3.6–11.9 (Fig. 1a). Similarly, Western blotting showed that the cancer tissues had higher levels of PTK6 protein expression compared with matched non-tumorous tissues (Fig. 1b).

PTK6 protein expression in cervical cancer tissues and normal cervical epithelia

PTK6 expression levels were detected with IHC staining in 150 cervical cancer specimens and 10 normal cervical epithelia specimens (Fig. 2). Of the 150 patients, 69 (46.0%) had high cytoplasmic PTK6 expression levels, and 81 (54.0%) showed low cytoplasmic expression levels. No PTK6 protein expression was detected in the 10 normal cervical epithelia samples.

Relationship between PTK6 expression and clinicopathologic characteristics of patients with early-stage cervical squamous cell cancer

PTK6 expression was associated with tumor grade in patients with early-stage cervical squamous cell cancer (P = 0.020); no significant association was detected between PTK6 expression level and age, tumor size, FIGO stage, deep cervical stromal invasion, pelvic lymph node metastasis, SCC antigen level, and lymphovascular space invasion (all P > 0.05) (Table 1).

Association of PTK6 expression and patient survival

The 5-year OS rate of the 150 patients was 90.7%. The 5-year OS rate was significantly lower in patients with high PTK6 expression than in those with low PTK6 expression (81.3% vs. 96.2%, P = 0.008) (Fig. 3a); the 5-year DFS rate was similar between these two groups (79.7% vs. 81.5%, P = 0.834) (Fig. 3b).