The 44 paired samples of ccRCC and normal adjacent tissues (ADTs) were collected from patients who underwent radical nephrectomy at the Department of Urology, Hefei Hospital Affiliated to Anhui Medical University, between November, 2007 and December, 2012. The ADT samples were collected at a distance of 2.0 cm from visible ccRCC lesions and were all properly maintained until reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis. For immunohistochemical analysis of the PTTG1 protein, a total of 192 paraffin-embedded pathologically verified ccRCC samples were collected. All the patients had undergone radical nephrectomy performed at the Department of Urology, Hefei Hospital Affiliated to Anhui Medical University, between 2000 and 2013. The histological and clinical diagnoses of the tumors in all the patients were performed by the Department of Pathology in the same hospital. The characteristics of the 192 patients are summarized in Table I. Patient survival data were obtained via telephonical communication and data on clinical characteristics were obtained from the medical records. Tumor stage was reclassfied based on the 2011 Union for International Cancer Control TNM classification of malignant tumors and nuclear grading was performed according to the Fuhrman classification (11).

This study was approved by the Ethics Committee of Anhui Medical University and all the patients provided written informed consent.

RT-qPCR was performed as previously described (12). The corresponding primer sequences were as follows: PTTG1: sense primer, 5′-AAAGCTCTGTTCCTG CCTCA-3′; and reverse primer, 5′-GAGAGGCACTCC ACTCAAGG-3′. GAPDH : sense primer, 5′-GGAGTCCAC TGGCGTCTTCACC-3′; and reverse primer, 5′-GAGGAG TGGG TGTCGCTGTTG-3′. The relative expression levels of PTTG1 were normalized to the geometric mean of GAPDH (internal control gene). The data were analysed via the comparative threshold cycle method (13).

Immunohistochemistry was performed to investigate PTTG1 expression in the 44 paired ccRCC and normal tissue samples. This was also implemented in the 192 ccRCC samples. All the procedures were performed as previously described (12). The sections were incubated with the monoclonal rabbit anti-human PTTG1 antibody (cat. no. sc-5843, dilution 1:400; Abcam, Cambridge, MA, USA).

The stained sections were reviewed by two independent pathologists. Scoring was based mainly on color intensity and extent. The proportion of cells expressing PTTG1 varied between 0 and 100% and the intensity of staining ranged from weak to strong. The proportion of PTTG1-expressing tumour cells was scored at low magnification on a scale of 0–5 (0, no positive cells; 1, 0–5%; 2, 6–25%; 3, 26–50%; 4, 51–75%; and 5, 76–100% positive cells). The intensity score was determined at high magnification on a scale of 0–3 (0, negative; 1, weakly positive; 2, moderately positive; and 3, strongly positive staining) and the final score was calculated by multiplying the two parameters, with scores of 0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 12 and 15. The optimal cut-off values for PTTG1 levels were set by measuring the heterogeneity in overall survival rates using the log-rank test method. Low expression was defined as a total score of <5 and high expression as a total score of ≥5. Thus, the stained sections were divided into two different groups by PTTG1 expression level.

Paired-sample t-tests were used in the RT-qPCR and immunohistochemical assays to analyze the significance of the differences in mRNA and protein expression level between ccRCCs and the ADTs. The correlation of PTTG1 expression with clinicopathological characteristics was assessed by the χ² test. Survival curves were plotted according to the Kaplan-Meier method and compared by the log-rank test. A multivariate analysis according to Cox's proportional hazards regression model adjusted for clinicopathological factors (age, gender, tumor size, Fuhrman grade, TNM stage and PTTG1 expression) was performed to identify the clinical variables that were independently correlated with overall survival. The statistical analysis was performed using the SPSS 17.0 package (SPSS Inc., Chicago, IL, USA) and P<0.05 was considered to indicate a statistically significant difference.

RT-qPCR was performed to measure the expression of PTTG1 mRNA in 44 ccRCC tumor tissues and normal ADTs. Compared to normal tissues, 37 ccRCC tumor tissue samples exhibited a higher expression at the mRNA level (paired-sample t-test, P<0.001; Fig. 1).

Immunohistochemistry was applied to assess the expression and subcellular localization of the PTTG1 protein in 44 paraffin-embedded ccRCC and paired normal ADT samples. PTTG1 staining was present mainly in the nuclei and cytoplasm (Fig. 2). In normal renal tissue, PTTG1 protein expression was negative (29/44, score=0) or low (15/44, score ≤5). The PTTG1 protein expression in the 44 tumor tissue samples was higher compared to that in the paired ADTs (paired-sample t-test, P<0.001; Fig. 3).

To further assess the correlation between PTTG1 expression and various clinicopathological parameters, a further immunohistochemical analysis was performed in 192 ccRCC samples. As shown in Table I, low expression of PTTG1 (score ≤4) was exhibited by 79 of the 192 tumor samples, whereas high expression (score ≥5) was exhibited by 113 samples. Increased expression of PTTG1 in tumor samples was correlated with T stage (χ² =10.816, P=0.004), N classification (χ² =4.674, P=0.031), metastasis (χ² =4.169, P=0.041), recurrence (χ² =7.903, P=0.005) and Fuhrman grade (χ² =14.719, P=0.002), while associations with age (χ² =0.560, P=0.454) and gender (χ² =1.819, P=0.177) were not identified. High expression of PTTG1 was observed in 69.5, 62.7 and 77.3% of stage T1, T2 and T3/4 ccRCC cases, respectively (P=0.004, χ² test). High expression of PTTG1 was observed in 69.4 and 53.8% of stage N0 and N1/2 ccRCC cases, respectively (P=0.041, χ² test). High expression of PTTG1 was observed in 74.2 and 54.6% of ccRCC cases with and without metastasis, respectively (P=0.009, χ² test). Finally, high expression of PTTG1 protein was observed in 80.0 and 54.1% of ccRCC cases with and without recurrence, respectively (P=0.005, χ² test).

To further investigate the prognostic value of PTTG1 expression in ccRCC, Kaplan-Meier analysis and the log-rank test were applied to assess the association between PTTG1 expression level in ccRCC and prognosis. The level of PTTG1 expression was found to be correlated with the overall survival of ccRCC patients. Patients with higher PTTG1 expression (PTTG1+) exhibited poorer survival rates compared to those with lower expression (PTTG1-). In the group of PTTG1+ patients, the mean and median survival time was 59.722 and 47 months, respectively; however, the mean and median survival time in the PTTG1-group was 88.285 and 79 months, respectively. The log-rank test revealed that the survival rates were significantly different between the two groups (log-rank; P=0.011, Fig. 4A). Similarly, patients without regional lymph node involvement (N0) exhibited a better prognosis compared to those with regional lymph node involvement (N+) (log-rank, P=0.024; Fig. 4B).

In addition, the multivariate Cox regression analysis indicated that PTTG1 expression (P=0.027) and N stage (P=0.020) were independent prognostic factors for the overall survival of ccRCC patients (Table II).