IJCRR

IJCRR - 12(6), March, 2020

Pages: 09-17

Date of Publication: 25-Mar-2020

Role of Immunohistochemical Markers in Surgical Margins of Patients with Head and Neck Carcinoma \? A Systematic Review

Author: Rajul Ranka, Minal Chaudhary, Preethi Sharma, Madhuri Gawande, Alka Hande, Swati Patil

Category: Healthcare

Abstract:Objective(s): The aim and objective of this study was to systematically evaluate the prognostic role of immunohistochemical markers in surgical margins of patients with head and neck squamous cell carcinoma (HNSCC).
Materials and Methods: MEDLINE/PubMed, Scopus and Cochrane library were searched for relative studies until December 2018. Retrospective and prospective original research studies published in English language assessing the prognostic value of immunohistochemical markers and disease-free survival in HNSCC and oral squamous cell carcinoma (OSCC) were included.
Results: A total of eight studies were included comprising of 269 cases. The studies included here used eukaryotic initiation transcription factor 4E (eIF4E) in HNSCC patients; Matrix metalloproteinase (MMP), MMP2, MMP3, MMP9, Dentin Sialo-Phosphoprotein (DSPP), Bone Sialoprotein (BSP), Osteopontin (OPN), Beta-2-adrenergic receptors and E-cadherin in OSCC patients and p53 in HNSCC and OSCC. Among all the markers studied MMP9 had the highest accuracy at 80% followed by p53 (75%), DSPP (70%) and OPN (70%) while eIF4E (33.3%) had least accuracy. A study suggested that E-cadherin is the preferred marker over MMP9. Almost all the studies used Fisher's exact and Fisher-Freeman-Halton significance test. Only one study was at low risk of bias, three studies were at moderate risk of bias, three studies had serious risk of bias and in one study bias could not be calculated due to inadequate information.
Conclusions: The study shows that immunohistochemical markers can significantly contribute to the field of head and neck carcinomas. Future efforts should concentrate on improving the antibody selection and its performance in the patients.

Keywords: E-cadherin, eIF4E, Matrix Metalloproteinase, Osteopontin, p53, Squamous cell carcinoma

Full Text:

Introduction

Head and neck cancer are the eighth most common cancer worldwide with majority being head and neck squamous cell carcinoma (HNSCC). HNSCC arise in the epithelial linings of the oral cavity, sino-nasal tract, pharynx, larynx and paranasal sinuses [1]. HNSCC is increasing rapidly since past decades and oral squamous cell carcinoma (OSCC) is the commonest tumor of head and neck region [2]. Despite various new advances in diagnosis and management, the long-term survival in HNSCC patients has not improved considerably.

The primary mode of treatment till date is surgery which may be followed by chemotherapy and radiotherapy depending upon the individual case [3]. To appropriately demarcate the exact margins is a dilemma for any surgeon. The presence of carcinoma in or close to the margin is an imperative prognosticator which could influence the local relapse of the patients [4]. The histopathologically tumor positive surgical margins are prognostic indicator for tumor relapse or distant metastasis. But relapse often occurs in cases with clear margins as well. This is in support with the hypothesis by Slaughter et al. [5] that residual or altered field in propinquity to the primary tumor might be the principal cause of local recurrence and treatment failure.

Immunohistochemical assessment of clear surgical margins could represent a more sensitive approach to detect the minimal residual cancer in them as genetic alteration precedes the phenotypic changes of the epithelium. Few studies have been done focusing on identification of immunohistochemical markers which may provide prognostic information and persuade the clinical outcome of HNSCC patients. The assessment of margins using immunohistochemistry (IHC) can be utilized by surgeons to provide better treatment to the HNSCC patients.

To the best of our knowledge, no study has systematically evaluated the prognostic role of immunohistochemical markers in histopathologically negative surgical margins of patients with HNSCC and specifically OSCC. This research followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines [6].

Methods

Eligibility Criteria

The studies were chosen based upon the following criteria:

Study design:   Retrospective and prospective original research studies published in English language were included from the date of inception to December 2018. Case series, animal studies, review papers, conference papers, abstracts and unpublished data, article published by same author with duplication of data were excluded.

Intervention: Only immunohistochemistry (IHC) studies were included. Studies that used other methods like RT-PCR, DNA methylation, gene analysis were excluded.

Participants: Patients diagnosed with SCC and treated with surgery having histopathologically negative (without dysplasia) surgical margins were included. Sites other than Head and neck region were excluded.

Outcome: Prognostic value of immunohistochemical markers and disease-free survival in HNSCC and OSCC.

Search Strategy

We conducted a systematic literature search online in MEDLINE/PubMed, Scopus and Cochrane library to identify pertinent articles. Following search terms and combinations were used as follows: (HNSCC) or (OSCC) and (Surgical margins). Citation lists from all retrieved studies were used to identify other relevant publications. Review articles were also scanned to recognize other related studies. Title and abstract of each recognized study were scanned to rule out any unrelated publications.

Data Extraction

Based on the title and/or abstract, two authors independently excluded irrelevant articles. After that, the remaining full text publications was assessed by two authors to review information according to the inclusion and exclusion criteria, and disagreement was resolved by third author. Two independent authors extracted all data of eligible publications. The following data were extracted from the received papers: author, year, country, recruitment period, number of patients, age range, tumor site, specimen analyzed, detection method, antibody source, cutoff value, follow-up time. Subsequently, the following information was extracted to clarify the association between expression of molecular markers and clinicopathological parameters including the gender, histopathological grade, size of tumor, lymph node metastasis, TNM stage based on the American Joint Committee on Cancer (AJCC), treatment, recurrence and disease free survival.

Quality Assessment

The Cochrane ROBINS-I (Risk of Bias in Non-randomized Studies-of Interventions) [7] tool was used for the quality assessment of the included studies. The quality assessment was carried out by two authors and disagreements were resolved by a third author. ROBINS-I tool include the use of signaling questions to guide the risk of bias judgments within seven bias domains. The studies were assessed based on the following domains: confounding; selection of participants; classification of interventions; deviations from intended intervention; missing data; measurement of outcomes; selection of the reported results. An overall estimation risk of bias was calculated at the end. (Supplementary table)

Results

Search Results and Outcome

Figure 1 shows the study search process. A total of ninety-seven articles were yielded in the initial search strategy. After screening of the titles and abstracts, sixty-six articles were excluded as they were found irrelevant. Among the remaining, twenty full text articles were screened and thirteen were excluded for the following reasons- i)Two were review articles, ii) One article was from same author and had data duplication, iii) Ten articles had interventions other than IHC. One relevant full text article was found from the screened citation lists. Eventually, eight full text publications were included in the present systematic review which met the criteria for qualitative synthesis.

General Characteristics of eligible studies

The main characteristics of the eligible studies are summarized in table 1. The total number of patients was 269, three articles were from Asian [1,11,12], and five were from non-Asian ethnicity [8,9,10,13,14]. Four studies composed of OSCC [10,12-14], one study compose of OSCC with Oropharynx SCC [8] and three composed of HNSCC [1,9,11] patients. All analyzed histopathologically negative surgical margin specimens and immunohistochemistry was used as detection method. One study used western blot as primary detection method and IHC was used for confirmation [9]. Follow-up period was varied in all the studies.

Immunohistochemical markers related parameters of included studies

There are varied IHC markers which could be used for detection of residual cancer in surgical margins of HNSCC patients. The studies included here used eIF4E in HNSCC patients; MMP2, MMP3, MMP9, DSPP, BSP, OPN, Beta-2-adrenergic receptors and E-cadherin in OSCC patients while p53 in both HNSCC and OSCC patients. p53 antibody was used in three studies, in one study the dilution is not mentioned and was sourced from oncogene science [8] and in other two studies it was used at 1:200 dilution sourced from dako and pre-diluted from ventana laboratory and it stained nuclei brown [10,1]. Three studies used eIF4E antibody at dilution of 1:500, the antibody expressed as reddish-brown peri-nuclear staining and the positive control used were Zenker’s diverticulum and Breast carcinoma tissue [1,9,11]. MMP2, MMP3, DSPP and BSP were used in one study at 1:100 dilutions and expressed as reddish-brown stain [10]. MMP9 was used in two studies at 1:100 and 1:20 dilution sourced from SantaCruz biotechnology and novocastra respectively, it stained cytoplasm brown [10,12]. Osteopontin (OPN) was used in two studies at 1:100 and 1:250 dilutions sourced from SantaCruz biotechnology and Abcam respectively, it gave dark brown membranous and cytoplasmic staining [10,12]. Human colon carcinoma tissue was used as positive control for OPN in one study [13]. E-cadherin was used in one study at a dilution of 1:50 and it stained nuclei and cell membrane brown [13]. One study used Beta-2 adrenergic receptor at 1:50 dilution which stained plasma membrane and cytoplasm brown [14]. Vascular smooth muscle was taken as internal control for the antibody.

Immunohistochemical studies: Potential markers

A list of studies on immunohistochemical markers with potential to identify tissue at risk of local relapse is depicted in Table 3. All the studies that used formalin-fixed paraffin embedded margins taken from adjacent to tumor were included. p53 was used to examine twenty-four tumor free surgical margins in two different studies out of which fourteen and thirteen margins expressed positively for antibody and ten and three patients had recurrence respectively [9,11]. Twenty-four surgical margins were evaluated using eIF4E antibody out of which twenty-one margins exhibited positive expression and six patients showed recurrence [11]. Twenty resection margins were evaluated using MMP2, MMP3 and MMP9 out of which twelve demonstrated positive expressions for MMP2, thirteen expressed for MMP3 and six margins expressed for MMP9 antibody and nine patients had recurrence [10]. MMP9 and E-cadherin were used to evaluate fifty-eight tumor free surgical margins, but the recurrence data had not been described in the study [12]. OPN was used to examine twenty negative resection margins out of which eleven expressed positively for the antibody and nine patients suffered recurrence [10]. Twenty negative surgical margins were evaluated for BSP amongst which thirteen were positive for antibody and nine patients had recurrence [10]. Beta-2-adrenergic receptor was used to evaluate sixty-two tumor free surgical margins amongst that fifty confirmed positive expression for the antibody and twenty-six patients had recurrence [14].

Immunohistochemical studies: case-control marker studies

Table 4 shows four studies with ten comparisons that have addressed the possible clinical efficacy of IHC markers for predicting local recurrence. Only studies that used a case-control approach, measuring the marker performance in a group with (cases) and without (controls) the development of local recurrence as end point have been included. Margin samples, deep or mucosal formalin fixed paraffin embedded were the tissue source of these studies and were always reported to be tumor free on routine histopathological examination. The studies show that this kind of research is in a ‘learning’ phase at current stage, since most markers have not been validated in an independent set of patients. The number of patients was twenty in two studies and was twenty-four in two studies [8,10,11,13].  We recalculated the data of each study and expressed the marker performance as sensitivity and specificity. BSP, DSPP, OPN, MMP2, MMP3 and MMP9 revealed more than 50% of sensitivity, DSPP (89%) illustrating the highest sensitivity [10]. p53 (80%) in one study exhibited high sensitivity and low in another [8,11] eIF4Eand OPN exhibited less sensitivity [11,13].  p53, DSPP, OPN, MMP3, MMP9 and eIF4E demonstrated more than 50% of specificity with OPN and MMP9 showing 100% specificity [8,10,13]. BSP, MMP2, MMP3 and p53 in another study lacked specificity [10,11]. Among all the markers studied MMP9 had the highest accuracy at 80% followed by p53 (75%), DSPP (70%) and OPN (70%) [8,10] while eIF4E [11] had least accuracy. Almost all the studies used Fisher’s exact and Fisher-Freeman-Halton significance test.

Quality of the included studies

Only one study [10] was at low risk of bias, three studies [8,11,14] were at moderate risk of bias and three studies [9,12,13] had serious risk of bias. The studies with serious risk mostly had missing data which questions the quality of these studies. The remaining one study [1] did not have enough information to calculate the risk of bias. (Supplementary table)

Significance of Potential Markers

Mutation in p53 gene is a well-known genetic abnormality in variety of cancers including HNSCCs [15]. Studies have revealed that p53 gene mutation leads to the pathogenesis of HNSCC cases with 50 to 60 percent of tumor cells expressing the p53 protein [16,17].

Eukaryotic translation initiation factor 4E (eIF4E) has role in initiation of protein synthesis, overexpression of which induces both the transformation and tumorigenesis as well as initiates metastasis [18]. Studies have reported 100% expression of this marker in HNSCC [9,19] which makes it a potential marker to predict recurrence.

Matrix Metalloproteinases (MMPs) have a key role in breakdown of basement membrane (BM), extracellular matrix (ECM) and contribute in release of growth-promoting signals, modulation of immune responses, apoptosis, and neo-angiogenesis, all of which is essential for tumor progression and growth [20]. The MMP-2 and MMP-9 degrades type IV collagen and its upregulation is associated with the degradation of the BM and the ECM, and with an increase in tumor aggressiveness [21]. Studies have revealed that MMP-9 is a prognostic indicator for malignant potential of OSCC [22,23].

Dentin sialophosphoprotein (DSPP), Osteopontin (OPN) and Bone sialoprotein (BSP) are reported to be upregulated in variety of cancers including OSCCs [24]. DSPP has been associated with aggressiveness of OSCCs and it has been expressed in oral premalignant lesions with dysplasia indicating subsequent invasive OSCC [25]. OPN has been coupled with proliferation, cancer cell growth, invasion and metastasis [26].

Beta-2-adrenergic receptor play role in regulation of tumor cell mechanisms including apoptosis, angiogenesis, proliferation, migration and metastasis through catecholamine induced activation under chronic psychological stress [27]. Studies have reported that ????2-AR has role in metastasis of OSCC and it is involved in proliferation and invasion of tumor cells [27,28].

E-cadherin is an inter-cellular adhesion molecule and regulates epithelial cell to cell adhesion [29]. The reduced immuno-expression of E-cadherin is correlated with malignancy, tumor invasiveness and carcinogenesis [30].

Discussion

Currently, the value of the use of immunohistochemical markers in routine examination of surgical margins of HNSCC has not yet established. More should be known about the analytical validity of the marker used, regarding accuracy and reliability.

This systematic review was an attempt to synthesize the existing data on prognostic role of immunohistochemical markers in surgical margins of patients with HNSCC. The total of eight studies met our inclusion criteria. Since included studies had heterogeneity in intervention (duration, nature, and content), inclusion and exclusion criteria, setting, duration of follow?up and methods of outcome assessment, we could not undertake meta?analysis. However, we have unambiguously offered the details and findings of the included studies.

Very few studies and diversity of evidence restrict the deductions that we can draw in this review and are deficient for definitive evidence. Limited data from very few included studies in this systematic review confirms the need for further large randomized control trials with long?term follow?up focusing on the evaluation of more markers and combination of markers. We look forward to such trials especially for oral squamous cell carcinoma patients.

There are varied IHC markers which could be used for detection of residual cancer in surgical margins of HNSCC patients. The studies included here used eIF4E in HNSCC patients; MMP2, MMP3, MMP9, DSPP, BSP, OPN, Beta-2-adrenergic receptors and E-cadherin in OSCC patients while p53 in both HNSCC and OSCC patients. The results were inconsistent in different studies due to methodological flaws, different sample size and follow-up time.

One study analyzed eIF4E in histologically negative surgical margins by western blot method; they used eIF4E antibody by immunohistochemistry in one case who had recurrence and elevated levels of eIF4E in western blot analysis[9]. Another study that analyzed eIF4E and p53 kept a 5% cut off value on the basis of expression of marker in basal cell layer of surgical margin and compared the expressions of the antibody concluding that the eIF4E is a better predictor of residual cancer compared to p53[11]. A prospective, observational and bilateral study in Australia and India was conducted using eIF4E and p53 concluding eIF4E to be a better prognosticator[1]. While, a study predicted p53 a useful biomarker as there is 5.333 times greater chance of recurrence in a p53 positive surgical margin[8].

DSPP, BSP and OPN are three members of the Small Integrin-Binding Ligand N-linked Glycoprotein (SIBLING) family of proteins and a study [10] reported that 45% of histologically negative surgical margins of OSCCs express at least one of the SIBLINGs and one of the MMPs (MMP2, MMP 3, MMP 9).  However, only DSPP, OPN and MMP9 exhibited significant association with recurrence or recurrence free survival. When sensitivity, specificity and predictive value were considered, MMP9 had the greatest overall accuracy. While another study that evaluated E-cadherin and MMP9 expression at negative surgical margins of OSCC, suggested that E-cadherin is the preferred marker over MMP9 [12]. A study which evaluated only OPN documented that the 55% of tumor free surgical margins showed positive expression of OPN in the epithelium, inflammatory cells and stromal cells which can be correlated to local recurrence and a poor prognosis[31].

A single study was found that evaluated β2-AR expression in the tumor free surgical margins of OSCC patients, but no statistically significant correlation was found with clinical variables and elevated levels of β2-AR [14].

The studies mainly conducted qualitative analysis of the IHC staining. Only one study did quantitative analysis using Intensity Reactivity Score (IRS) which is calculated by multiplying Staining Intensity (SI) with percentage of positive cells (PP) [12]. One study did semi-quantitative analysis depending upon the percentage of positive cells [10]. A study used MATLAB computing language-based software and the representative images were segmented using the software and analysis was done. The median was established as cut off value to classify the intensity of staining [14].

The studies used log rank test and a Kaplan-Meier curve to analyze the probability of local recurrence of cancer [9,10,11]. Univariate analysis and Multivariate Cox regression analysis was used when multiple markers were analyzed for their significance in recurrence free survival of the patients [10,11]. Contingency tables and ?²-test were used to evaluate the association of molecular markers with clinical characteristics [10]. Wilcoxon, Mann-Whitney, Kruskal-Wallis, Spearman’s rank correlation coefficient test and Fisher’s exact tests were used to assess the difference between expression of markers and clinico-pathological parameters in the studied groups [12,14].

There are certain limitations that should be taken into consideration in this systematic review. The studies had different inclusion and exclusion criteria, a study [10] included margins of patients with metastatic lymph node while in another study [11] such patients were excluded. A study included surgical margins with presence of dysplasia as well which should have been excluded [8]. The main limitation being the confounding factors and selection of participants as reflected by ROBINS-I tool of assessment. The results varied from low to critical bias in confounding factors and low to serious risk of bias in selection of participants (Supplementary table). The wide heterogeneity of these studies regarding varied group of immunohistochemical markers in different population for diverse follow up time is an important limitation. Such variable factor does not allow creating a standard prognostic marker for cancer patients. Therefore, further studies with enough and elaborative data regarding immunohistochemical markers with large follow up time and large number of samples is recommended.

Conclusion

In summary, the study shows that immunostaining of resected margins for tumor markers can significantly contribute to the field of head and neck carcinomas. Future efforts should concentrate on improving the antibody selection and its performance in the patients with detailed information. Long term studies with different panel of tumor markers, their qualitative analysis with survival data can help in establishing a potential prognostic marker which can be used in surgical margins of HNSCC and OSCC.

Acknowledgement: I thank Dr. Mahalaque Quazi, Professor, Dept of Physiology, Datta Meghe Institute of Medical Sciences (DU) for her kind guidance in writing this review. Authors acknowledge the immense help received from the scholars whose articles are cited and included in references of this manuscript. The authors are also grateful to authors / editors / publishers of all those articles, journals and books from where the literature for this article has been reviewed and discussed.

References:

1. Joseph S, Janakiraman R, Chacko G, etal. Predictability of Recurrence using Immunohistochemistry to delineate Surgical Margins in mucosal Head and Neck Squamous Cell Carcinoma (PRISM-HNSCC): study protocol for a prospective, observational and bilateral study in Australia and India. BMJ Open 2017;7:e014824. doi:10.1136/ bmjopen-2016-014824.

2. Travassos DC, Fernandes D, Massucato EMS, Navarro CM, Bufalino A. Squamous cell carcinoma antigen as a prognostic marker and its correlation with clinicopathological features in head and neck squamous cell carcinoma: Systematic review and meta-analysis. J Oral Pathol Med. 2018;47:3–10.doi.org/10.1111/jop.12600.

3. Braakhuis BJ, Bloemena E, Leemans CR, Brakenhoff RH. Molecular analysis of surgical margins in head and neck cancer: more than a marginal issue. Oral Oncol. 2010 Jul;46(7):485-91. doi: 10.1016/j.oraloncology.2010.01.019.

4. Eckardt A, Barth EL, Kokemueller H, Wegener G. Recurrent carcinoma of the head and neck: treatment strategies and survival analysis in a 20-year period. Oral Oncol 2004;40(4):427–32.

5. Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer 1953;6:963-8.

6. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151:264-269, w64.

7. Sterne JAC, Hernán MA, Reeves BC, Savovi? J, Berkman ND, Viswanathan M, et al. ROBINS-I: A tool for assessing risk of bias in non-randomized studies of interventions. BMJ 2016; 355; i4919; doi: 10.1136/bmj.i4919.

8. Ball VA, Righi PD, Tejada E, Radpour S, Pavelic ZP, Gluckman JL. p53 immunostaining of surgical margins as a predictor of local recurrence in squamous cell carcinoma of the oral cavity and oropharynx. Ear Nose Throat J. 1997 Nov;76(11):818-23

9. Nathan CO, Liu L, Li BD, Abreo FW, Nandy I, De Benedetti A. Detection of the proto-oncogene eIF4E in surgical margins may predict recurrence in head and neck cancer. Oncogene. 1997 Jul 31;15(5):579-84. DOI:10.1038/sj.onc.1201216

10. Ogbureke KU, Weinberger PM, Looney SW, Li L, Fisher LW. Expressions of matrix metalloproteinase-9 (MMP-9), dentin sialophosphoprotein (DSPP), and osteopontin (OPN) at histologically negative surgical margins may predict recurrence of oral squamous cell carcinoma. Oncotarget.2012;3(3):286-98. 

doi: 10.18632/oncotarget.373; PMCID: PMC3359885; PMID: 22410369

11. Singh J, Jayaraj R, Baxi S, Mileva M, Skinner J et al. Immunohistochemical expression levels of p53 and eIF4E markers in histologically negative surgical margins, and their association with the clinical outcome of patients with head and neck squamous cell carcinoma. Mol Clin Oncol. 2016;4:166-172. DOI: 10.3892/mco.2015.689

12. Mohtasham N, Anvari K, Memar B, Saghravanian N, Ghazi N, Bagherpour A, Ramtin M. Expression of E-cadherin and matrix metalloproteinase-9 in oral squamous cell carcinoma and histologically negative surgical margins and association with clinicopathological parameters. Rom J Morphol Embryol. 2014;55(1):117-21. PMID:24715175

13. Subramani VN, Narasimhan M, Thiyagarajan M, Munuswamy BD, Jayamani L. Expression of Osteopontin in Oral Squamous Cell Carcinoma and its Surgical Margins- An Immunohistochemical Study. JCDR. 2015;9(11):ZC669. doi: 10.7860/JCDR/2015/12777.6836; PMID:26675878; PMCID:PMC4668527

14. Oliveira DT, Bravo-Calderón DM, Lauand GA, et al. Correlation of Beta-2 Adrenergic Receptor Expression in Tumor-Free Surgical Margin and at the Invasive Front of Oral Squamous Cell Carcinoma. J Oncol. 2016; 2016:3531274. doi: 10.1155/2016/3531274PMID: 27042179PMCID: PMC4793135

15. Jalali MM, Heidarzadeh A, Zavarei MJ, et al. p53 overexpression impacts on the prognosis of laryngeal squamous cell carcinomas. Asian Pac J Cancer Prev. 2011;12:1731-4.

16. Nathan CAO, Sanders K, Abreo FW, et al. Correlation of p53 and the proto-oncogene eIF4E in larynx cancers: prognostic implications. Cancer Res. 2000;60:3599.

17. Bradford CR, Kumar B, Bellile E, et al (2013). Biomarkers in advanced larynx cancer: triological society thesis. Laryngoscope. 2014;124(1):179–187.doi: 10.1002/lary.24245; PMCID: PMC4123871; NIHMSID: NIHMS585804; PMID: 23775802

18. Chen CN, Hsieh FJ, Cheng YM, et al. Expression of eukaryotic initiation factor 4E in gastric adenocarcinoma and its association with clinical outcome. J Surg Oncol. 2004;86:22-7.

19. Cardesa A, Nadal A. Carcinoma of the head and neck in the HPV era. Acta Dermatoven APA. 2011;20:161-73.

20. Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2002;2:161-74.

21. Mishev G,  Deliverska E,  Hlushchuk R,  Velinov N,  Aebersold D, et al. Prognostic value of matrix metalloproteinases in oral squamous cell carcinoma. Biotechnol Equip. 2014;28(6):1138–1149. doi: 10.1080/13102818.2014.967510; PMCID: PMC4433935; PMID: 26019601.

22. Gu ZD, Chen KN, Li M, Gu J, Li JY, Clinical significance of matrix metalloproteinase-9 expression in esophageal squamous cell carcinoma, World J Gasteroenterol. 2005;11(6):871–874.

23. Andisheh-Tadbir A, Mardani M, Pourshahidi S, Nezarati K, Bahadori P. Prognostic value of matrix metalloproteinase-9 expression in oral squamous cell carcinoma and its association with angiogenesis. J Clin Exp Dent. 2016;8:e130-5.

24. Bellahcène A, Castronovo V, Ogbureke KU, et al: Small integrin-binding ligand N-linked glycoproteins (SIBLINGs): multifunctional proteins in cancer. Nat Rev Cancer. 2008;8:212-226.

25. Ogbureke KU, Abdelsayed RA, Kushner H, et al: Two members of the SIBLING family of proteins, DSPP and BSP, may predict the transition of oral epithelial dysplasia to oral squamous cell carcinoma. Cancer. 2010;116:1709-1717.

26. Wu IC, Yang SF, Wu CC, Hsu WH, et al. Expression of osteopontin protein in esophageal squamous cell carcinoma. J Intern Med Taiwan. 2010;21:419-26.

27. Z. J.Shang, K.Liu, and D. F. Liang. Expression of ????2-adrenergic receptor in oral squamous cell carcinoma, JOPM. 2009;38(4):371–376.

28. D. G. Bernab´e, A. C. Tamae,´E. R. Biasoli, and S. H. P. Oliveira. Stress hormones increase cell proliferation and regulates interleukin-6 secretion in human oral squamous cell carcinoma cells. Brain Behav Immun. 2011;25(3):574–583.

29. Sun GY, Wu JX, Wu JS, Pan YT, Jin R, Caveolin-1, E-cadherin and β-catenin in gastric carcinoma, precancerous tissues and chronic non-atrophic gastritis. Clin J Cancer Res. 2012; 24(1):23–28.

30. Vered M, Allon I, Buchner A, Dayan D, E-cadherin in oral SCC: an analysis of the confusing literature and new insights related to its immunohistochemical expression, Histol Histopathol. 2012;27(2):141–150.

31. Huber GF, Züllig L, Soltermann A, Roessle M, Graf N, Haerle SK, Studer G, Jochum W, Moch H, Stoeckli SJ, Downregulation of E-cadherin (ECAD) – a predictor for occultmetastatic disease in sentinel node biopsy of early squamouscell carcinomas of the oral cavity and oropharynx, BMC Cancer. 2011;11:217–218.