Specimens of tissues were obtained from 108 NPC patients and 14 nasopharyngitis patients who underwent nasopharyngeal biopsy at the Department of Otolaryngology, Head Neck Surgery, The Second Xiangya Hospital of Central South University (Changsha, China) from January 2002 to January 2004. The NPC biopsy specimens were from patients who were diagnosed with primary NPC, and the noncancerous nasopharyngeal biopsy samples were from patients who were suspected of having NPC originally, but afterwards were pathologically and clinically proven to be clear. There were 87 male and 21 female patients, and their median age was 56 years (range, 19–86 years). The diagnoses for all patients were confirmed by histopathologic examination. The NPC tissue specimens were histologically confirmed and classified according to WHO classification into keratinizing carcinoma (type 1, 8 cases), differentiated non-keratinizing carcinoma (type 2, 30 cases) and undifferentiated non-keratinizing carcinoma (type 3, 70 cases). All patients were re-staged according to the 2010 American Joint Committee on Cancer (AJCC) criteria. There were 2 cases in stage I (T1N0M0 2 cases), 12 cases in stage II (T1N1M0 2 cases, T2N0M0 6 cases, T2N1M0 4 cases), 33 cases in stage III (T1N2M0 6 cases, T2N2M0 12 cases, T3N0M0 6 cases, T3N1M0 1 case, T3N2M0 8 cases) and 61 cases in stage IV (T1N3M0 5 cases, T2N3M0 22 cases, T2N1M1 1 case, T2N2M1 1 case, T2N3M1 2 cases, T3N3M0 13 cases, T3N2M1 1 case, T3N3M1 2 cases, T4N0M0 2 cases, T4N1M0 2 cases, T4N2M0 1 cases, T4N3M0 4 cases, T4N1M1 1 case, T4N2M1 3 cases and T4N3M1 1 case). Demographic details and tumor characteristics are listed in Table I.

All tissue specimens were fixed in 10% neutralized formalin and embedded in paraffin. Seventeen cases of NPC tissues among the 108 cases and 14 cases of noncancerous nasopharyngeal tissues were cut into 2 portions. One portion was fixed in 10% neutralized formalin and embedded in paraffin blocks for immunohistochemistry. The other portion was frozen immediately and stored in liquid nitrogen for RNA and total protein extraction.

The study was approved by the Central South University Ethics Committee. Informed consent was obtained from all of the patients. All specimens were handled and made anonymous according to the ethical and legal standards.

All patients received a complete course of radiotherapy with or without chemotherapy. Irradiation was administered primarily with megavoltage (6–10 MV) X-rays from a linear accelerator. The dose was 1.8–2.0 Gy/fraction, with 5 fractions/week. All patients were evaluated by nasopharyngoscope at 10–14 weeks after completion of radiotherapy. Patients were followed up regularly after treatments. The follow-up period was defined as the interval between the date of treatment completion and the date of either the patient's death or the last follow-up. Deaths were treated as censored cases. Clinical examinations, including an indirect mirror examination of the nasopharynx with or without a nasopharyngoscope, were performed during subsequent follow-ups. Chest X-ray, liver echo, whole body bone scans, computed tomography and magnetic resonance imaging (MRI) were performed only when clinically indicated. All disease-free patients had a minimum follow-up of 60 months after the completion of treatment. Locoregional or distant recurrence was confirmed by histopathological or radiological criteria.

Specimens were obtained before therapy, fixed in 10% neutral buffered formalin, and embedded in paraffin. Serial 5-μm sections were prepared. Slides were deparaffinized in xylene, rehydrated in graded alcohol, and incubated in 0.3% hydrogen peroxide in methanol to quench the endogenous peroxidase activity for 30 min. The sections were subjected to microwave heat-induced antigen retrieval in 0.01 M citrate buffer, pH 6.0, at high power twice for 7 min each. Nonspecific binding sites were blocked by a 20-min incubation with normal horse serum.

The sections were incubated for 2 h at room temperature with primary rabbit polyclonal antibody detecting TrkB (1:100) followed by the addition of HRP-labeled goat anti-rabbit polymers. The streptavidin-biotin-peroxidase complex tertiary system (Boster, Wuhan, China) was used according to the manufacturer's instructions. All slides were visualized by applying 3,3′-diaminobenzidine tetrahydrochloride for 2 min and then counterstained with hematoxylin. Negative controls were non-immune rabbit IgG at the same dilution as for the primary antibody.

Sections were evaluated in a blinded manner and scored by two independent experienced pathologists. Expression levels of TrkB protein were assessed by observing the incidence and staining intensity (i.e., color strength) of immuno-positive cells. Staining intensity was scored as: 0, negative; 1, weak; 2, strong. The incidence of positive cells was scored as: 1, 0–50%; 2, 51–75%; 3>75%. Tumors were categorized into three groups based on the final staining point: low expression (scored 0–3), and high expression (scored 4–6) (11).

The Epstein-Barr virus (EBV)-negative, highly differentiated NPC cell line HNE-1 and the EBV-negative, lowly differentiated NPC cell line HNE-2 were established by the Cancer Research Institute, Xiangya School of Medicine, Central South University (12). The NPC cell line C666-1 which consistently expressed EBV was established from undifferentiated nasopharyngeal carcinoma by the Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, University of Hong Kong (Hong Kong, China) (13). The NPC cell lines CNE-1 and CNE-2 were established from the primary tumors of patients with well-differentiated squamous cell NPC and poorly differentiated squamous cell NPC, respectively, by the Chinese Academy of Medical Sciences. The highly metastatic NPC cell line 5-8F and the nonmetastatic NPC cell line 6-10B were established from the NPC cell line SUNE-1 by the Cancer Research Institute of Sun Yatsen University, Guangzhou, China (14). The above 7 NPC cell lines were kindly provided by Cancer Research Institute, Xiangya School of Medicine, Central South University. The nontransformed nasopharyngeal epithelium cell line (NP-69) was established from the nasopharynx of a human by the Department of Anatomy, Faculty of Medicine, University of Hong Kong (15). This cell line was purchased from the Cell Experiment Center of Shanghai, China. All cell lines were maintained as monolayer cultures in Dulbecco's modified Eagle's medium (DMEM)/F12 medium (1:1) supplemented with 10% fetal bovine serum (FBS), 100 IU/ml penicillin and 100 IU/ml streptomycin at 37°C in a humidified atmosphere with 5% CO₂.

Total RNA of the tissue specimens, the 7 NPC cell lines and the nontransformed nasopharyngeal epithelium cell line (NP-69) was isolated by Simply P Total RNA Extraction kit (Bioer Technology, Inc., Hangzhou, China). Total RNA (1 μg) was reverse transcribed by High-Capacity cDNA Reverse Transcription kits (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. The primers for TrkB and GAPDH were synthesized by Sangon Biological Engineering Technology and Services Co., Ltd. (Shanghai, China) as follows: TrkB F, 5′-AGTCGCAGATGCTGCATATTAG-3′ and R, 5′-TAGAATGTCCAGGTAGACC-3′; GAPDH F, 5′-CTGCAGCATC TTCTCCTTCC-3′ and R, 5′-CAAAGTTGTCATGGATGACC-3′. The reverse transcription product (1 μl) was amplified by PCR using the following conditions: 95°C for 5 min; followed by 32 cycles at 95°C for 30 sec, 58°C for 30 sec, and 72°C for 30 sec; and an extension at 72°C for 5 min. The PCR product (8 μl) was then electrophoresed on 1.5% agarose gel. TrkB mRNA expression levels were quantified by Quantity One software (Bio-Rad Technical Service Department, Hercules, CA, USA) and represented as the densitometric ratio of the targeted gene to GAPDH. PCR experiments were repeated three times.

Total protein lysates were harvested from the tissue specimens and cell lines by lysis buffer (20 mM Na₂PO₄, 150 mM NaCl, 1% Triton X-100, 1% aprotinin, 1 mM phenylmethylsulfonyl fluoride, 100 mM NaF and 2 mM Na₃VO₄). Proteins (50 μg) were separated by polyacrylamide gel electrophoresis on a 10% sodium dodecyl sulfate-polyacrylamide gel and transferred to polyvinylidene difluoride membranes. Membranes were incubated with the primary antibody overnight at 4°C. On the next day, the membranes were washed with phosphate-buffered saline (PBS) three times and then the membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit secondary antibodies (all the antibodies were obtained from Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). Finally, the proteins were detected by enhanced chemiluminescence (ECL) procedure. TrkB protein expression levels were quantified by Quantity One software and represented as the densitometric ratio of the targeted protein to β-actin. Western blot experiments were repeated three times.

Data are expressed as means ± SD. Student's two-tailed t-test was used to compare TrkB expression between cancer and noncancerous nasopharyngeal tissues. The protein expression and clinicopathological parameters were compared using the χ² test. Survival analysis was undertaken using the Kaplan-Meier method, and curves were compared using the log-rank test. Multivariate analysis, using the Cox's proportional hazards model, was applied to identify which factors were independent indicators for prognosis. All statistical evaluations were performed using the Statistical Package for the Social Sciences version 10.0 software program (SPSS Inc., Chicago, IL, USA). All tests were two-tailed, and a P-value of <0.05 was considered to indicate a statistically significant result.

The 108 NPC tumor samples were initially tested for TrkB protein expression by immunohistochemistry, and then the level of TrkB protein expression was correlated with clinicopathological parameters. Positive TrkB immunostaining was predominantly diffusely distributed throughout the cytoplasm of NPC cells (Fig. 1). Of all the samples tested, 65 (60.2%) cases displayed low TrkB protein expression (Fig. 1C and D) and 43 (39.8%) cases displayed high TrkB protein expression (Fig. 1E and F). The χ² test showed that the TrkB expression level was significantly associated with NPC tumor size (T classification) (P=0.015), lymph node metastasis (N classification) (P=0.009), AJCC clinical stage (P=0.007). However, we did not find a significant association between TrkB expression level and age, gender and distant metastasis (M classification). The relationship between clinicopathological characteristics and TrkB expression level in individuals with NPC are summarized in Table II.

Seventeen freshly collected clinical NPC tissues and 14 freshly collected normal nasopharyngeal mucosal tissues were further investigated by RT-PCR and western blot analysis. The results of RT-PCR showed that the TrkB relative mRNA expression level was upregulated in the NPC tissues when compared to that in the nasopharyngeal mucosal tissues (1.1±0.64 vs. 0.38±0.27, P=0.000<0.01; Fig. 2A and C). Consistent with the mRNA data, the TrkB relative protein expression level in the NPC tissue was significantly higher when compared with that in the nasopharyngeal mucosal tissues (0.52±0.24 vs. 0.26±0.18, P=0.003<0.01; Fig. 2B and C).

In previous experiments, we found that both TrkB mRNA and protein expression levels were upregulated in NPC tissues and were correlated with a number of clinicopathological parameters. The expression pattern of TrkB in human NPC cell lines was further investigated by RT-PCR and western blot analysis to verify the role of TrkB in vitro. As shown in Fig. 3, both mRNA and protein expression of TrkB in all the 7 NPC cell lines was significantly higher when compared to that in the normal nasopharyngeal epithelium mucosal NP-69 cell line. Moreover, compared to the other NPC cell lines, the 5-8F cell line, which has high metastatic potential, displayed the highest TrkB mRNA and protein expression.

During the follow-up period of 5 years, 6 patients were lost to follow-up or died from other causes. Fifty-five patients were still alive without recurrence, and the remaining 47 patients had recurrent tumors. Thirty-nine of these 47 patients with recurrent disease died of locoregional recurrence or distant metastases within 5 years.

According to the TrkB protein expression levels, 108 patients were divided into 2 groups: 65 patients in the low expression group and 43 patients in the high expression group. In the Kaplan-Meier patient survival analysis, NPC patients in the high expression group had both reduced overall survival and disease-free survival when compared with patients in the low expression group (overall survival rate, 48.8 vs. 73.8%, P=0.001<0.05; disease-free survival rate, 41.9 vs. 66.2%, P=0.001<0.05) (Fig. 4). In the multivariate Cox regression analysis, high TrkB expression [risk ratio (RR)=1.863, P=0.013], poor N classification (RR=5.216, P=0.028) and M classification (RR=2.359, P=0.049) were identified as independent prognostic factors for overall survival. Moreover, these 3 clinicopathologic parameters were identified as independent prognostic factors for disease-free survival (TrkB expression: RR=1.600 and P=0.033; N classification: RR=6.609 and P=0.011; M classification: RR=2.544 and P=0.025). The other clinicopathological parameters including age, gender, histological type, T classification and AJCC clinical stage that were examined did not provide any independent prognostic information (Table III).