Discrepant expression and prognostic role of epidermal growth factor receptor and icotinib radiosensitization for nasopharyngeal carcinoma CURRENT STATUS: POSTED

Background: Epidermal growth factor receptor (EGFR) expression was observed in many tumors. Icotinib (IH), an EGFR tyrosine kinase inhibitor, could enhance the radiosensitivity, but few studies have focused on nasopharyngeal carcinoma (NPC). Materials and Methods : One hundred fifteen NPC lesions and 30 nasopharyngitis lesions were enrolled for EGFR expression evaluation with immunohistochemical staining. The correlation of EGFR expression and clinical parameters was analyzed. Survival analysis was calculated using univariate and multivariate analysis. A radioresistant NPC cell line, CNE-2R, was established with a gradient irradiation schedule of 60 Gy from CNE-2. Cell viability of CNE-2/2R was detected using microculture tetrazolium assay and colony-forming assay. The IH radiosensitization effect and EGFR expression were also evaluated in CNE-2/2R. Results: High EGFR expression was more frequently detected in NPC lesions (73/115) than chronic nasopharyngitis lesions (5/30, p =0.000). EGFR expression was correlated with tumor stage ( p =0.000) and tumor-node-metastasis (TNM) stage ( p =0.006). EGFR expression was an independent prognostic factor of disease-free survival ( p =0.047) and overall survival ( p =0.016). The optical density value of CNE-2R was higher than CNE-2 on days 1, 2, 4, and 6 after radiation of 4 Gy ( p <0.005). CNE-2 colony numbers clearly decreased compared with CNE-2R ( p <0.05), and IH enhanced the radiosensitizing effect of CNE-2R with an apparently lower survival fraction (1.414 times, p <0.05). Higher EGFR expression was observed in CNE-2R cells and decreased accordingly when exposed to IH. Conclusion: High EGFR expression was more frequently detected as a poor prognostic factor in NPC patients. Higher radioresistance and EGFR expression were observed in CNE-2R, and IH could reduce it. These results provided a new theoretical basis for clinical application of icotinib radiosensitization in NPC radiotherapy. for clinical application


Introduction
Nasopharyngeal carcinoma (NPC) is a common malignant tumor with regional distribution in southern China with an annual incidence of 3.0 per 100,000 persons [1]. Radiotherapy with or without chemotherapy is the predominant treatment modality for NPC, especially local advanced NPC. It is well-known that distant metastasis and local failure contribute largely to poor outcome [2,3]. Residual irradiated NPC cells with high radioresistance are the most common cause. In addition, chemotherapy drugs are regarded as radiosensitizers in the concurrent chemoradiotherapy of local advanced NPC [4,5]. However, severe therapeutic side effects, including myelosuppression and mucositis lead to reduced treatment tolerance and quality of life [6]. Thus, it is urgent to explore effective therapeutic strategies for increasing radiosensibility with fewer side effects.
Epidermal growth factor receptor (EGFR) is a transmembrane glycoprotein that belongs to the ErbB receptor tyrosine kinase family, which enhances cancer cell proliferation, differentiation, metastasis, and angiogenesis [7]. EGFR overexpression has been demonstrated in NPC, which involves chemo-/radioresistance, malignant progression, and poor survival, making it possible to block the EGFR signal pathway to improve the radiosensitivity [8,9]. EGFR inhibitors include monoclonal antibodies and small molecule tyrosine kinase inhibitors. For the former, Huang et al. demonstrated that concurrent administration of nimotuzumab and celecoxib cooperatively enhanced the cytotoxicity and radiosensitivity of NPC cells (CNE-2) in vitro [10]. Clinical studies have shown that cetuximab or nimotuzumab, combined with radiotherapy can enhance the efficacy of radiotherapy with a better safety profile [11][12][13].
Icotinib hydrochloride (IH), a small molecule inhibitor of EGFR, has shown radiosensibility in several malignant tumors. Zhang et al. found that pretreatment with IH reduced clonogenic survival, inhibited EGFR activation, and increased radiation-induced apoptosis of lung cancer cells [14]. Similar results were also observed in colorectal cancer and cervical cancer cells [15,16]. However, very little is known about the radiosensitization of IH on NPC. In our previous study, IH (125 mg/day) plus intensity-modulated radiation therapy (IMRT) in patients with locally advanced NPC had an acceptable safety profile and was well tolerated [17]. Hence, we check EGFR expression and its correlations with clinical pathological features in NPC patients, and then explore IH radiosensitization efficacy to NPC cells. Province. All the enrolled patients signed an informed consent form for the potential use of biopsy specimens for scientific research. To protect their privacy, only de-identified patient data were used.

Materials And
On hundred fifteen primary NPC paraffin-embedded specimens were collected from the two hospital tissue banks between January 2011 and October 2017 with the approval of the Ethics Committee.
Chronic nasopharyngitis lesions (n = 30) were collected between August 2017 and October 2017. All the patients were pathologically diagnosed with immunohistochemistry, and NPC patients with distant metastasis were excluded at the time of diagnosis. According to the 7th edition of the American Joint Committee on Cancer staging system, tumor-node-metastasis (TNM) staging score was determined.
All the patients received standard treatment with radiotherapy, chemotherapy, or chemoradiotherapy. There was also no another malignant disease history for the selected patients.
The median age of 115 patients was 54 years (ranging from 30 to 82 years), and other detailed characteristics are summarized in Table 1. The average follow-up was 53.7 ± 25.1 months (range: 10.6-99 months).

Immunohistochemical staining and score for EGFR
Paraffin-embedded slides of 4 µm were dewaxed in xylene and rehydrated through a graded series of ethanol. Antigen retrieval was performed by boiling the slides with citrate buffer (pH = 6) at 120 °C for 5 min, then blocked by normal goat serum. Internal peroxidase activity was blocked using a 3% hydrogen peroxide solution at room temperature for 15 min. After that, the samples were incubated with antibodies against EGFR (18986-1-AP, Proteintech Group, Wuhan, China) at 4 °C overnight. After washing with 0.01 M of phosphate-buffered saline (PBS) solution, the samples were incubated with labeled goat anti-rabbit secondary antibody (SA00001-2, Proteintech Group, Wuhan, China) for 60 min at 37℃. Following washing with PBS again, the samples were stained with diaminobenzidine solution, counterstained with hematoxylin, and mounted with glycerol gelatin.

Generation of radioresistant subclone
When CNE-2 cells reached approximately 60% confluence, the gradient irradiation schedule was initiated: The flask surface was covered by tissue equivalent filler with a thickness of 1.5 cm and then subjected to 2 Gy using Siemens clinical linear accelerator (Primus H, 6MV-X, dosage rate of 200 cGy/min, source skin distance 100 cm, and irradiation field of 10*10 centimeter). Following radiation exposure, cells were further cultured in a new flask. When they reached 60% confluence again, cells were then irradiated to 2 Gy a second time. These procedures were repeated three times with a gradient model of 2, 4, 6, and 8 Gy to the total dose of 60 Gy in 12 months, and the surviving cells were cultured and named CNE-2R. Another flask of CNE-2 cells was cultured synchronously as a control cell line with no radiation exposure.

Microculture tetrazolium (MTT) assay
CNE-2/2R cells in the exponential phase were seeded and incubated overnight in 96-well micro-titer plates at a density of 5,000 cells/well with a volume of 100 ul/well. CNE-2/2R cells were irradiated with the dose of 4 Gy according to the above irradiation scheme. MTT assay was performed to test optical density (OD) value on days 1, 2, 4, and 6 after irradiation. Briefly, CNE-2/2R cells were treated with 20 µl/well of 3-(4, 5-diethyl-2-thiazolyl)-2, 5-diphenyltetrazolium bromide (5 mg/ml) for 4 h at 37˚C, resulting in the generation of formazan crystals. After aspirating the medium, DMSO (150 µl/well) was added to dissolve the formazan for 10 min, and then OD value was measured in a microplate reader (Safe Heart, SHE 3000) at a wavelength of 490 nm. Each group had three repeated wells on different days.
CNE-2 cells in the exponential phase were seeded in 96-well microtiter plates at a density of 10,000 cells/well and incubated overnight. CNE-2 cells were then treated by IH with different final concentrations (0.013, 0.025, 0.05, 0.1, 0.2, and 0.4 mg/ml) in a total volume of 100 ul/well. After 24 h of IH addition, MTT assay was performed, and the percentage inhibition rate was calculated as: (1 -OD value of experimental group/control group OD) x100%. The IC20 (20% inhibitory concentration) was selected as the subsequent experimental concentration.

Western Blot assay
The CNE-2, CNE-2R, and CNE-2R + IH groups in the exponential phase were treated with or without IH for 24 h. The medium containing IH or DMSO was substituted with complete growth medium, and then cells were irradiated with 4 Gy. After 24 h, the protein lysates of the three cell groups were harvested and centrifuged, and then the supernatants were collected. Protein was separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes. After blocking with 5% protease-free bovine serum albumin for 1 hour, the membranes were incubated with EGFR antibody (18986-1-AP, Proteintech Group, Wuhan, China) overnight at 4℃. Following washing with tris-buffered saline with Tween (TBST) three times, the membranes were continually incubated with secondary antibody labeled with horseradish peroxidase.
The bands of EGFR were finally visualized. GAPDH was used as an internal control to balance equal loading.

Statistical analysis
Statistical analysis was performed with PASW Statistics V18.0 software. All quantitative data are represented as mean values (± SEM), and statistical differences between groups were analyzed using the t-test. GraphPad Prism software V6.0 was used for graphing the data. The correlation between EGFR expression and clinical parameters was calculated with the Pearson's chi-squared (x 2 ) test.
Survival analysis was conducted according to the Kaplan-Meier method and Cox proportional hazards regression analysis. P < 0.05 was considered to be significant.

Association of EGFR expression and clinical parameters in NPC patients
The association between EGFR expression and clinical parameters of NPC patients was shown in    Abbreviation: EGFR, Epidermal growth factor receptor; HR, hazard ratio; CI, confidence interval; TNM, tumor-node-metastasis. Table 3 The main parameters of cell survival curves of CNE-2 and CNE-2R treated irradiation with or without icotinib.
MTT assay was performed to detect the effect of IH on cell viability and IC20 value (0.04 mg/ml) was obtained for subsequent experiments (Fig. 3D). In clonogenic assay, the radiosensitivity difference between CNE-2/2R was obvious (Table 3 and Fig. 4A). In response to radiation, there was a dosedependent reduction of clonogenic cell survival. CNE-2 colony numbers decreased compared with CNE-2R (p < 0.05), and IH enhanced the radiosensitizing effect of CNE-2R with an apparently lower survival fraction (1.414 times, p < 0.05).

EGFR expression profiles in CNE-2/2R cells.
In RT-PCR assay, the EGFR expression level was 1.343 times higher in CNE-2R cells than CNE-2 cells (p < 0.05). In western blot assay, GAPDH was used as an internal control. Compared with CNE-2 cells, EGFR protein expression was higher in CNE-2R cells (Fig. 4B). When IH was added, EGFR expression was decreased accordingly. These results demonstrated that EGFR expression was higher in the radioresistant cell lines than the parental cell line and IH could reduce EGFR expression.

Discussion
Although NPC is sensitive to radiotherapy, local recurrence and distant metastasis remain the leading causes resulting in treatment failure, which is attributed primarily to the inherent and acquired radioresistance. Enhancing the radiosensitivity with chemotherapy drugs is a conventional method for locally advanced NPC treatment [4,5,18]. However, severe side effects may affect and even interrupt the chemoradiotherapy procedure. A number of studies have demonstrated that EGFR inhibitors can improve the radiosensitivity in many malignant tumors [14-16, 19, 20]. Few studies have focused on the radiosensibility of IH in NPC, which is the aim of the study based on our previous study [17].
EGFR expression was presented in 83.5% of NPC patients, in contrast to 23.3% of chronic nasopharyngitis patients, with a significant statistical difference. Further analysis showed that EGFR expression was significantly associated with TNM stage and T stage, but not gender, age, tumor differentiation, and lymph node status. In univariate and multivariate Cox proportional regression analysis, two parameters (EGFR expression and TNM stage) and three parameters (EGFR expression, age and TNM stage) were identified as independent prognostic factors for DFS and OS, respectively, while EGFR expression was the common factor. Zhang et al. reported that EGFR expression was 89.5% in 96 NPC patients and correlated with T stage [21]. Zafar et al. reported 86.35% of EGFR expression in head and neck squamous cell carcinoma [22].
In addition, EGFR expression is an independent prognostic factor, three-year survival rates were lower in EGFR-positive (75.8%) patients than in EGFR-negative patients (91.2%). A meta-analysis including 15 3.35]) were higher than 2, indicating a strong prognostic value of EGFR [23].
Another meta-analysis including 22 studies produced similar results [8]. Hashmi et al. showed a significant association of EGFR expression with tumor stage and DFS [24]. Inhibiting EGFR expression might be feasible to reverse the radioresistance of NPC cells.
Radioresistant cell lines can be established via cumulative exposure to irradiation [25][26][27]. In our study, the radioresistant cell line CNE-2R was successfully obtained using a gradient irradiation schedule with accumulated doses of 60 Gy. Compared with CNE-2, CNE-2R cells had stronger ability of proliferation and colony forming, which means a higher radioresistance. There was a higher EGFR expression in CNE-2R cells than CNE-2 cells. When CNE-2R was exposed to IH (IC20 value, 0.04 mg/ml) plus irradiation, EGFR expression and colony-forming ability were reduced. These results demonstrated that IH improved the radiosensitivity of CNE-2R. Ma et al. evaluated the effect of IH on radiosensitivity and showed that IH more effectively sensitized colorectal cancer cells (HT29 and HCT116) to radiation [15]. Wang et al. reported IH can significantly inhibit cell proliferation, redistribute cell cycle, enhance apoptosis and impair DNA damage response of Hela S3 cells following radiation [16].
In conclusion, high EGFR expression is a poor prognostic factor for NPC patients, CNE-2R cells obtained by gradient irradiation are distinguished from the CNE-2 cells in viability, radiosensitivity, and EGFR expression. IH is capable of enhancing the radiosensitivity of CNE-2R. The study provides a new theoretical basis for clinical application of IH radiosensitization in NPC radiotherapy.

Conclusions
High EGFR expression was more frequently detected as a poor prognostic factor in NPC patients.
Higher EGFR expression and radioresistance were observed in radioresistant cell line CNE-2R, which was reduced by IH. These results provided a new theoretical basis for clinical application of IH radiosensitization in NPC radiotherapy.   The three cohorts were exposed to different dose and continuously incubated until macroscopic colony (>50 cells per colony) appeared. Survival fraction curve was conducted; (B) expression level of EGFR protein in CNE-2/2R cells was detected by western blot for the three treatment groups.