Clinicopathological Significance of ATRX Expression in Nasopharyngeal Carcinoma Patients: A Retrospective Study

Background: Nasopharyngeal carcinoma (NPC) is the most common head and neck squamous cell carcinoma in south China. Radiation technology improves the local control rates in early NPC. However, the distant metastases are still the main cause of treatment failure. Thus, to find biomarkers for prognosis will help to enhance the survival of NPC. ATRX is a chromatin remodeling protein localized in the nucleus. Deletion or mutation of ATRX gene has been demonstrated in a variety of malignancies. However, the significance of ATRX expression in the prognosis of NPC remains unclear. Methods: Tumor tissues from 227 NPC patients diagnosed in the Second Xiangya Hospital of Central South University from 2011 to 2016 were selected. Immunohistochemistry was used to detect the ATRX expression level of the tumor tissue. Chi-square test was used to analyze the relationship between ATRX expression and clinical characteristics such as age, sex, T stage, N stage and clinical stage. Kaplan-Meier method was used for survival analysis, and log-rank was used to compare the difference in survival rate. Results: There were 53 patients with negative ATRX expression, accounting for 24.2% of the total group. ATRX expression was not significantly associated with age, sex, N stage, clinical stage, and progression-free survival (PFS) (P>0.05). However, patients with negative ATRX expression had earlier T staging (P=0.045) and a higher 5-year overall survival (84.9% vs 66.9%, P=0.022). Conclusions: Loss of ATRX expression may contribute to better prognosis in patients with NPC.


Introduction
Nasopharyngeal Carcinoma (NPC) is the most common head and neck squamous cell carcinoma in south China and southeastern Asia [1,2]. Due to its radiosensitive behavior, radiation (RT) and chemoradiotherapy (CRT) has become the cornerstone of the treatment for early or locoregionally advanced diseases [3,4]. However, about 25-30% of the patients will suffer from disease relapse or metastasis, and the prognosis of advanced patients is poor. The biological behavior of NPC varied dramatically in individual patients. Therefore, seeking for biomarkers for survival prediction, patient stratification and treatment adaptation is urgently needed [5].
Alpha thalassemia/mental retardation X-linked (ATRX) is a protein containing an ATPase/helicase domain, which belongs to the switch/sucrose nonfermentable (SWI/SNF) family of chromatin remodeling proteins [6]. It is found to be involved in cell cycle-dependent phosphorylation, which regulates its nuclear matrix and chromatin association and suggests its involvement in gene regulation at interphase and chromosomal segregation in mitosis [7]. It has been demonstrated that ATRX is frequently mutated in a variety of mesenchymal tumors Ivyspring International Publisher including gliomas [8], neuroendocrine neoplasms [9], and sarcomas [10]. ATRX mutation leads to decreased ATRX protein expression, and results in tumor genome instability, higher tumor mutation burden, and thus leading to increased sensitivity to chemotherapy, radiation therapy and immunotherapy agents [11,12]. However, the data on the incidence of ATRX expression and clinical significance in epithelial carcinomas is limited. It is reported that ATRX loss predicts good prognosis in cervical carcinoma and hepatocarcinoma [13,14]. And ATRX mutation results in increased immune checkpoint inhibitor (ICI) sensitivity in NSCLC [12]. However, the significance of ATRX expression in the prognosis of NPC remains unknown.
In the present study, we evaluated the ATRX expression in NPC patients via immunohistochemistry and analyzed the correlation between ATRX expression and clinicopathological characteristics and patient's prognosis, and demonstrated that ATRX deficiency is correlated with earlier T stage and longer overall survival in NPC patients.

Patients enrollment and clinical data
Patients who were diagnosed with NPC at the Second Xiangya Hospital of Central South University (Hunan, China) between 2011 and 2016 were retrospectively reviewed. And patients without complete clinical data or unavailable paraffin blocks were excluded. Finally, a total of 227 patients were enrolled in this study. Two pathologists (HH Wang and SQ Fan) independently reviewed all H&E staining and IHC staining slides. And all the clinical and pathological characteristics were obtained from electronic medical records system of the hospital. Approval for using the patient material in this study was obtained from the Ethics Committee of the Second Xiangya Hospital of Central South University. All procedures were conducted in accordance with the declaration of Helsinki. Tissue samples were obtained from the Department of Pathology of the Second Xiangya Hospital of Central South University.

Immunohistochemical staining and semi-quantitative scores
Immunohistochemical staining for ATRX was performed with 4-μm-thick sections from paraffin blocks using the MaxVision HRP polymer anti-rat IHC kit. Each section was deparaffinized and rehydrated, and high-temperature antigen retrieval was achieved for all antibodies by heating the samples in 0.01 M citrate buffer in a domestic microwave oven at full power (1000 W) for 15 mins. Besides internal positive control, positive control slides were included in every experiment. Methanol containing 3% H2O2 was applied for fifteen minutes to inactivate endogenous peroxidase. The slides were incubated with a rabbit monoclonal antibody against human, a 1:200 dilution of the primary antibody to ATRX (Sigma, Sigma-Aldrich, St. Louis, MO, USA) at 4 °C overnight. Then, all the slides were rinsed with PBS three times for 5 mins each. After the incubation with secondary antibody, the visualization signal was conducted with 3,3′-diaminobenzidine tetrachloride. Subsequently, all slides were counterstained with hematoxylin. The specificity of the antibody was determined with matched IgG isotype antibody as a negative control. Tissue samples were evaluated independently by HHW and SQF who were blinded to the clinicopathological data, at 200× magnification light microscopy. Nucleus staining of the tumor cells was assessed. Representative microscopic photos are shown in Figure 1. According to the proportion of nucleus staining of the tumor cells, the ATRX expression was classified as negative and positive. ATRX staining loss of tumor nucleus was defined as ATRX negative while ATRX positive was defined as that ATRX is expressed in at least 1% of nuclei in the tumor cells. The concordance rate between the two pathologist is 97%.

Statistical analysis
Statistical analysis was performed using SPSS software V.22 (IBM, Armonk, USA). The χ² test was used to evaluate the association between ATRX expression and clinicopathological parameters including Age, Gender, T stage, N stage and clinical stage according to AJCC 8 th staging system. Progression-free survival (PFS) and overall survival (OS) curves were plotted using Kaplan-Meier method, and the log-rank test was used to compare the differences. P<0.05 was considered as statistically significant.

Patient characteristics
The clinicopathological characteristics of the patients are summarized in Table 1

Correlations between ATRX expression and clinicopathological factors
NPC patients with ATRX expression, defined as ATRX positive staining in at least 1% of nuclei in the tumor cells, consist of 76.7% (174/227) of the cohort. And 23.3% (53/227) of the cohort had ATRX loss. The association between ATRX expression and the clinicopathological factors are shown in Table 2. ATRX loss was significantly correlated with earlier T stage (p=0.045). Statistical significance was not found in sex, age, ECOG, smoking history, hemoglobin level, treatment mode, chemotherapy agent type, N stage and AJCC stages.

Impact of ATRX expression on the prognosis of NPC patients
As shown in Figure 2, NPC patients with ATRX loss had significantly longer overall survival than those with ATRX expression (84.9% vs 66.9%, P=0.022). No prognostic significance between ATRX expression and progression free survival (PFS) was observed (73.1% vs 65.9%, p=0.124).

Discussion
In the present study, we investigated the expression of ATRX in NPC and its correlation with clinicopathological characteristics and prognosis of NPC patients. ATRX protein expression was absent in 24.2% of NPC patients, and ATRX loss is correlated with earlier T stage and longer OS in NPC patients.
ATRX is a member of the SWI/SNF chromatin remodeling superfamily which is involved in remodeling and stabilizing genome [15]. It is reported that SWI/SNF genes are found to be mutated in almost 25% of malignancies, which have prognostic significance [16]. Moreover, expression changes of specific SWI/SNF chromatin remodeling proteins are also implicated as prognostic biomarkers [16]. However, the consequences of mutations or expression alterations on prognoses are highly context-specific [16]. ATRX mutation is mostly seen in mesenchymal malignances, including glioma, sarcoma, and neuroendocrine tumors, while the report on the incidence and prognostic significance on epithelial malignancies is rare. Loss of ATRX results in impaired genome damage repair system, which leads to genetic instability, microsatellite instability (MSI), and more aggressive phenotype [17]. At the same time, it also sensitizes cancer cells to treatment agents, including DNA-damaging agents, radiotherapy [11,18], and immune checkpoint inhibitors [12,19]. The prognostic significance of ATRX mutation on cancer patients varies between researches. It is reported that ATRX mutation is a poor prognostic factor in pancreatic neuroendocrine tumors (pNETs) [20], hepatic angiosarcomas [21], and leiomyosarcoma [22,23], while it is found to be a good prognostic factor in glioma [24], and cervical cancer [13]. Such complexities without universal good prognosis may suggest that mutations affecting chromatin remodeling genes or expression altering chromatin remodeling proteins do not directly determine the aggressiveness or prognosis of malignancies [16]. Besides, this discrepancy may be resulted from patient selection and treatment modality. In the present study, we firstly reported the incidence of ATRX deficiency in NPC patients, and found that ATRX deficiency predicts good prognosis in NPC patients. However, the underlying mechanism is not fully understood, and it may be related with the following aspects. Firstly, since the ATRX protein plays significant epigenetic roles in maintaining telomere length via depositing histones at heterochromatin and telomeric DNA, tumors with ATRX mutations or loss exhibit altered telomeres and telomere dysfunction [25,26]. Moreover, ATRX mutations increase tumor mutational burden (TMB), PD-L1 expression, as well as manifested interferon gamma (IFN-γ) signaling [12,19]. Given that radiotherapy or chemoradiotherapy is the primary treatment for NPC, and the potential role of ATRX mutation in glioma patients treated with RT has been reported. Collectively, the mechanism of ATRX mutation conferring survival advantages in NPC is worth further investigation.
Mutations in ATRX are thought to result in loss of ATRX protein which mediates loss of function (LOF), and therefore in the present study, immunohistochemistry (IHC) is used as a surrogate to detect mutations. In previous research in glioma, pancreatic neuroendocrine tumors, and neuroblastoma have demonstrated that ATRX mutations are associated with absence of ATRX protein. However, in another research, Chami et al. [27] reported that using IHC as a surrogate may miss ATRX mutations in neuroblastoma, which indicated detecting ATRX mutation by IHC protein expression as a surrogate still needs further validation. Furthermore, the consistency between ATRX protein absence and ATRX mutation in NPC is worth further investigation.
Although this study identified the prognostic significance of ATRX loss in NPCs, there are several limitations. First, confirmation of ATRX mutations by other techniques, such as sequencing from fresh tissue was not included. Additionally, this study is a retrospective study, and due to a limited sample size, there is a need for validation in larger cohorts with clinical, molecular alterations, conjointly with protein expression/activation.
In conclusion, we evaluated the clinicopathological significance of ATRX expression in 227 patients with NPC. ATRX loss was significantly associated with good prognosis in patients.