Article Text
PDF
Abstract
Aims Head and neck cancer, including oral squamous cell carcinoma (OSCC), is the sixth most common cancer worldwide and has a high potential for locoregional invasion and nodal metastasis. Therefore, discovery of a useful molecular biomarker capable of predicting tumour progression and metastasis of OSCC is crucial. We have previously reported zinc finger AN1-type containing 4 (ZFAND4) as one of the most upregulated genes in recurrent OSCC using a cDNA microarray analysis. Although ZFAND4 has been shown to promote cell proliferation of gastric cancer, its expression and clinicopathological roles in OSCC remain unclear.
Methods In this study, we examined ZFAND4 expression by immunohistochemistry in 214 cases of OSCC.
Results High cytoplasmic expression of ZFAND4 was observed in 45 out of 214 (21%) patients with OSCC. Expression levels of ZFAND4 were strongly associated with metastasis to the lymph nodes (p=0.0429) and distant organs (p=0.0068). Cases with high expression of ZFAND4 had a significantly unfavourable prognosis compared with patients with low expression of ZFAND4 (p<0.0001). Furthermore, ZFAND4 overexpression was an independent poor prognostic factor for OSCC as determined by multivariate analysis using the Cox proportional hazards model (p<0.0001).
Conclusions These results suggest that ZFAND4 is a useful marker for predicting metastasis and poor prognosis in patients with OSCC.
- ZFAND4
- metastasis
- prognosis
- oral cancer
Statistics from Altmetric.com
Introduction
Head and neck cancer, including oral squamous cell carcinoma (OSCC), ranks as the sixth most common cancer worldwide, with an estimated 300 400 patients being diagnosed with the disease and an estimated 145 400 patients succumbing to the disease annually.1 It is predicted that 49 670 new cases will be diagnosed and 9700 deaths will occur due to OSCC each year in the United States.2 Moreover, it is one of the leading causes of morbidity and mortality in Melanesia, South-Central Asia and Central and Eastern Europe because of smoking, betel nut chewing and alcohol abuse.1 Despite remarkable progress in cancer detection and treatment, the 5-year survival rate of OSCC has remained below 50% over the past three decades, with more than 70% of cases with advanced stage disease not responding to therapy.3 For this reason, elucidating the detailed molecular mechanisms underlying tumour progression and metastasis of OSCC is essential.
Generally, the following 10 key alterations are fundamental to the development of cancer cells: sustained proliferative signals, evasion of growth suppressors, resistance to cell death, limitless replicative potential, induction of angiogenesis, activation of invasion and metastasis, avoidance of immune destruction, deregulation of cellular energetics, genome instability and mutation and tumour-promoting inflammation.4 These hallmarks are applicable to OSCC and we have defined several molecular markers related to it. Some typical examples of molecular markers related to OSCC include the following: members of the melanoma inhibitory activity gene family that play an important role in progression of OSCC3 5–8; and the storkhead box protein 2 (STOX2) and pancreatic adenocarcinoma upregulated factor, which promotes chemoresistance to anticancer drugs in human OSCC cells.9 10 Nevertheless, useful and specific tumour markers of OSCC have not yet been clarified.
Investigation of differential global gene expression analysis using microarray and Serial Analysis of Gene Expression methods will be useful for discovering novel markers for diagnosis and treatment of OSCC.11 We have previously compared the transcriptional profiles of primary and recurrent OSCC using a cDNA microarray. One of the most upregulated genes identified in recurrent OSCC was zinc finger AN1-type containing 4 (ZFAND4),12 also known as AN1, ubiquitin-like, homolog (ANUBL1). It has been shown that higher expression of ZFAND4, as determined by immunohistochemistry, is strongly associated with clinical stage progression in gastric cancer.13 However, the functional roles of ZFAND4 in malignancies are still controversial. The aim of the present study is to demonstrate the utility of ZFAND4 as a prognostic predictive marker in human OSCC.
Materials and methods
Surgical specimens
A total of 214 formalin-fixed, paraffin-embedded cases of primary OSCC (131 men and 83 women, age range40–89 years; mean age 66.5 years) were collected from Nara Medical University Hospital, Kashihara, Japan. The study was performed according to the ethical standards presented in the Declaration of Helsinki, and was approved by the Medical Ethics Committee of Nara Medical University (approval number 719). None of the patients had had radiotherapy and/or chemotherapy before surgical resection. Tumour staging and the histological grade of OSCCs were classified according to the UICC TNM classification system, seventh edition and the WHO criteria, respectively. Medical records and prognostic follow-up data were obtained from the patient database maintained by the hospital. The follow-up period was 102–1981 days (mean 1372 days, median 1469 days).
Immunohistochemistry
Consecutive 3 µm sections were cut from each block, and immunohistochemistry was performed. An immunoperoxidase technique was done following antigen retrieval with microwave treatment (95ºC) in citrate buffer (pH 6.0) for 45 min. After endogenous peroxidase was blocked with 3% H2O2–methanol, specimens were incubated in a 10% skim milk solution (Morinaga Milk, Tokyo, Japan) for 20 min to block non-specific antibody reactions. Slides were rinsed with phosphate-buffered saline three times. Anti-ZFAND4 antibody (Atlas Antibodies, Stockholm, Sweden) diluted to 0.5 µg/mL was used for the primary antibody. After 2 hours of incubation at room temperature, specimens were treated for an hour at room temperature with the peroxidase-conjugated anti-rabbit secondary antibody (Medical & Biological Laboratories, Nagoya, Japan) diluted to 0.5%. The specimens were colour developed with diaminobenzidine (DAB) solution (Dako, Carpinteria, California, USA). After washing, specimens were counterstained with Meyer’s-hematoxylin (Sigma–Aldrich, St Louis, Missouri, USA). Immunostaining of all samples was performed using the same conditions of antibody reaction and DAB exposure.
Evaluation of immunohistochemistry
Immunoreactivity was evaluated by proportional score and intensity score using the Allred score.14 Proportional scores were classified by the ratio of immunopositive cells as follows: 0, no cell was stained; 1, 1 ⁄ 100 cells were stained; 2, 1 ⁄ 10 cells were stained; 3, 1 ⁄ 3 cells were stained; 4, 2 ⁄ 3 cells were stained; 5, all cells were stained. Intensity scores were categorised by the intensity of immunopositivity as follows: 0, negative; 1, weak; 2, intermediate; 3, strong. The total score was calculated by the sum of the proportional and intensity scores (its grades ranged from 0 to 8). The purpose of this study is to elucidate the usefulness of ZFAND4 as a prognostic factor; therefore, the best cut-off score for ZFAND4 expression was selected based on receiver operating characteristic (ROC) curve analysis according to the survival status.15 16 The score closest to the point with both maximum sensitivity and specificity (0.0, 1.0) on the curve was selected as the optimal cut-off score.
Statistical analysis
The ROC curve analysis was used to determine a cut-off point for ZFAND4 expression. Relationships between ZFAND4 expression and clinicopathological parameters were evaluated using Fisher’s exact test. Disease-free survival was analysed according to the Kaplan–Meier method, and differences between groups were calculated using a log-rank test. Univariate and multivariate analyses were carried out using the Cox proportional hazards model (described as a hazard ratio (HR) with 95% confidence intervals (95% CI), together with the p value). Statistical analysis was carried out with JMP13 (SAS Institute, Cary, North Carolina, USA) and a p value less than 0.05 was regarded as statistically significant.
Results
Immunostaining of ZFAND4 in human OSCC
We examined the immunostaining of ZFAND4 in 214 OSCC cases. Although immunoreactivity for ZFAND4 was negative or very weak in non-cancerous oral mucosa, including the minor salivary gland, (figure 1A, B), cytoplasmic detection of ZFAND4 was found in OSCC cells (figure 1C-E). To determine the optimal cut-off score of ZFAND4 immunoreactivity, we performed the ROC curve analysis. The cut-off score by ROC analysis was four with an area under the ROC curve (AUC) of 0.7051 (95% CI 0.6095 to 0.8007), a sensitivity of 88.6% and a specificity of 57.45% (figure 2). Therefore, cases with an Allred score above four were defined as ZFAND4 high expression, and those with a score of 4 or less were defined as ZFAND4 low expression.
Expression of zinc finger AN1-type containing 4 (ZFAND4) by immunohistochemistry. Weak and/or no detection of ZFAND4 was observed in normal oral mucosa (A) and the minor salivary gland (B). In oral squamous cell carcinoma (OSCC) cases, cytoplasmic immunostaining of ZFAND4 was found at the primary site (C), the nodal metastatic lesion (D) and the pulmonary metastatic focus (E). Original magnification was 400×.
Receiver operating characteristic (ROC) curve analysis of zinc finger AN1-type containing 4 (ZFAND4) expression as a predictor for disease-free survival of patients with oral squamous cell carcinoma (OSCC). The arrow shows the cut-off score of 4 (a sensitivity of 88.6%, and a specificity of 57.45%). Patients with an Allred score above four were defined as ZFAND4 high expression, and those with a score of 4 or less were defined ZFAND4 low expression.
Relationship between ZFAND4 expression and clinicopathological variables in OSCC specimens
Summaries of ZFAND4 immunostaining and the clinicopathological characteristics of OSCC are represented in table 1. High expression of ZFAND4 was found in 21% (45/214) of OSCC cases. A significant relationship was found between ZFAND4 expression level and lymph node metastasis (p=0.0429). Detection of ZFAND4 high expression was observed in 30.2% (19/63) of cases with nodal metastasis and only 17.2% (26/151) of cases without nodal metastasis. Distant metastases were found in the pulmonary tissue (six cases), bone (three cases) and liver (one case) in a total of 10 patients (4.7%). These distant metastases were strongly associated with ZFAND4 expression levels (p=0.0068). High expression of ZFAND4 was detected in 41 of 209 patients (19.1%) without distant metastases, whereas 60% (6/10) of the cases with distant metastases overexpressed ZFAND4. Moreover, expression levels of ZFAND4 were positively correlated with vascular (p=0.0234) and lymphovascular infiltration (p=0.0377). There was no significant difference between immunoreactivity for ZFAND4 and other clinicopathological factors in OSCC.
Relationship between ZFAND4 expression and clinicopathological parameters
Relationship between ZFAND4 expression and OSCC prognosis
During the follow-up period, 47 of the 214 patients presented with a local or metastatic recurrence of cancer. The disease-free survival curves calculated using the Kaplan–Meier method showed that patients with high ZFAND4 expression had a shorter disease-free survival compared with those patients with low ZFAND4 expression (p<0.0001, figure 3). We performed univariate and multivariate Cox proportional hazards analyses to confirm factors involved in disease-free survival. In the univariate analysis, T factor (p=0.0450), clinical stage (p=0.0337), nodal metastasis (p=0.0021), distant metastasis (p=0.0119) and ZFAND4 expression levels (p<0.0001) were associated with a poor outcome in patients with OSCC and these comparisons were statistically significant. Additionally, nodal metastasis (p=0.0457) and ZFAND4 overexpression (p<0.0001) were considered as independent predictors of unfavourable prognosis in patients with OSCC as determined by multivariate analysis (table 2). These results may indicate that ZFAND4 is a useful marker for predicting metastasis and poor prognosis of patients with OSCC.
Disease-free survival curve in patients with oral squamous cell carcinoma (OSCC) based on zinc finger AN1-type containing 4 (ZFAND4) expression levels. ZFAND4 high expression cases had significantly worse disease-free survival compared with those with ZFAND4 low expression (p<0.0001).
Univariate and multivariate analysis of disease-free survival
Discussion
We previously identified ZFAND4 as one of the most overexpressed genes in recurrent OSCC using a cDNA microarray analysis.12 In gastric cancer, expression levels of ZFAND4 are higher in advanced cases than in early cases.13 ZFAND4 promotes cell proliferation by activation of cyclin-dependent kinase and downregulation of p21 and p53 in SGC-7901 cells, which are derived from human gastric cancer.13 In addition, transcriptional upregulation of ZFAND4 is also observed in phosgene-induced acute lung injury in mice.17 However, there is still a lot of uncertainty surrounding the functional roles of ZFAND4 in human malignancies because the literature concerning ZFAND4 is limited. Lastly, it remains unresolved whether ZFAND4 is beneficial as a prognostic predictive marker for OSCC.
The following steps are obligatory for cancer invasion and metastasis: ablation of the basement membrane, reduced adhesion and segregation of cancer cells, gain of cancer cell movement and stromal infiltration, intravascular influx and migration and cancer cell growth in the metastatic organ.9 18 Recent advances in cancer biology have elucidated the molecular mechanisms underlying the metastasis of OSCC. For example, the epithelial-mesenchymal transition, characterised by the loss of epithelial features and the acquisition of a mesenchymal phenotype, is necessary for cancer metastasis because it leads to an increase in the number of circulating tumour cells in the blood.19 20 However, no available molecular biomarker for metastatic OSCC has been discovered. Our data strongly indicate that ZFAND4 overexpression is correlated to nodal and distant metastasis and a poor prognosis in patients with OSCC. ZFAND4 expression was also associated with lymphatic involvement and hematogenous spread of tumour cells, further implicating it in metastasis. It is well known that angiogenesis and lymphangiogenesis strongly promote cancer metastasis.21 We have previously revealed that angiogenesis and lymphangiogenesis play a key role in the local progression, nodal metastasis and poor prognosis observed in patients with OSCC.6 ZFAND4 may facilitate metastasis to the lymph nodes and distant organs, possibly by evoking angiogenesis and/or lymphangiogenesis in OSCC. Further investigation into the relation between ZFAND4 and the molecular biological aspects including angiogenesis/lymphangiogenesis is required and would benefit from further examinations using appropriate in vivo and in vitro models.
In the current study, it is noteworthy that upregulation of ZFAND4 was correlated with distant metastasis of OSCC. While distant metastases are uncommon in OSCC, when they do occur, the disease becomes highly lethal.22 Considering there is no effective therapy except for palliative care, the average survival for patients with distant metastasis from OSCC is 2.4 months in Taiwan.23 The incidence of distant metastasis accounts for 3.8–12.6% of OSCC cases. The sites that OSCCs most frequently metastasise to are the lung, bone and liver.22 24–26 In our study, 10 patients (4.7%) with OSCC had distant metastasis with the most common metastatic site being the lung (6/10, 60%), followed by the bone (3/10, 30%) and then the liver (1/10, 10%). Recent reports have shown that the clinical stage, T grade, N grade, histological grade, depth of invasion, pattern of invasion and the locoregional control are pivotal in distant metastasis of OSCC.22–26 However, tumour-related molecules responsible for distant metastases of OSCC are largely unknown. Our findings indicate that ZFAND4 is essential for distant metastasis of OSCC cells and further large-scale studies using clinical samples are required to clarify the underlying mechanism.
In summary, ZFAND4 has proved to be a useful tool for predicting metastasis and a worse outcome for patients with OSCC. To the best of our knowledge, this is the first study describing the ability of ZFAND4 to predict the poor prognosis of patients with cancer. The present results might be helpful to understand the molecular mechanism of tumour progression and metastasis of OSCC. If ZFAND4 is a secretory marker, it might be detectable in serum, saliva, urine and other samples that can be collected easily from patients with cancer. There are several limitations to this study and we must reveal the precise molecular biological mechanism of why and how ZFAND4 acts as an oncogene in OSCC. There is also some question as to whether ZFAND4 is a useful therapeutic target that may inhibit the growth and metastasis of OSCC. Although additional biological and clinicopathological examinations are fundamental to elucidating the roles of ZFAND4, our results indicate that ZFAND4 can serve as an effective and novel target for diagnosis and treatment of human OSCC.
Take home messages
We investigated the expression of zinc finger AN1-type containing 4 (ZFAND4) in cases of oral squamous cell carcinoma (OSCC).
Overexpression of ZFAND4 is correlated with metastasis and poor prognosis of OSCC.
ZFAND4 expression is an independent predictor of disease-free survival in patients with OSCC.
Our results indicate that ZFAND4 is a useful molecular target for the diagnosis and treatment of OSCC.
Abstract translation
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.Acknowledgments
This work was supported in part by Grant-in-Aid for Scientific Research from Japan Society for the Promotion of Science, Japan.
References
Footnotes
MK-S and TS contributed equally.
Handling editor Runjan Chetty.
Contributors Conception and design: TS, HK. Acquisition of data: TS, MKS, HN, CN. Analysis and interpretation of data: TS, MKS, CN, TK. Writing and review of the manuscript: TS, MKS, HK, TK.
Disclaimer This abstract has been translated and adapted from the original English-language content. Translated content is provided on an ‘as is’ basis. Translation accuracy or reliability is not guaranteed or implied. BMJ is not responsible for any errors and omissions arising from translation to the fullest extent permitted by law, BMJ shall not incur any liability, including without limitation, liability for damages, arising from the translated text.
Competing interests We declare that there is not any financial support or relationships which may pose a conflict of interest in the contents of the submitted manuscript.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.