Detection Level of Salivary Survivin in Patients with OSCC

Santarelli A1, Mascitti M1, Lo Russo L2, Colella G3, Giannatempo G2, Bambini F1, Emanuelli M1, Procaccini M1 and Lo Muzio L*2 1Department of Clinical Specialistic and Dental Sciences, Marche Polytechnic University, Ancona, Italy 2Department of Clinical and Experimental Medicine, Foggia University, Foggia, Italy 3Department of Medical, Surgical and Dental Specialities, SUN University, Naples, Italy


Introduction
Oral Squamous Cell Carcinoma (OSCC) is the most common malignant tumor of the oral cavity, representing more than 90% of all cancers of this district [1,2]. It is the eighth most common cancer worldwide, even if epidemiologic variations between different geographic regions have to be taken into account [3,4]. In spite of the advances accomplished in oncology and surgery, the mortality rates in the last 20 years remain almost unchanged, with the 5-year survival rate ranging from 35 to 50% [5]. The diagnostic delay represents one of the principal reasons for the OSCC poor prognosis. Indeed, making the diagnosis at an advanced stage (ie. Stage III or IV) drastically reduces the patient's life expectancy [6].
Thus, discover of sensitive and specific molecular tumor markers for early cancer detection may improve the survival rates of OSCC.
Survivin, a member of Inhibitors of Apoptosis protein family (IAP) [7], is a 16.5 kDa protein composed of 142-amino acid encoded by a gene located on the 17q25 chromosome [8]. This protein controls chromosome spindle checkpoint assembly, suggesting that the primary function of Survivin is in controlling cell division, rather than apoptosis inhibition [9]. Survivin exhibits low or undetectable expression levels in most non-proliferating adult tissues, but is broadly expressed in a wide variety of cancers [10], suggesting it has been implicated in both control of cell survival and regulation of mitosis in tumor cells [11]. The levels of Survivin in OSCC tissues result greatly higher than those in normal mucosa [12], and its degree of expression is directly related to aggressive phenotype and poor prognosis [13][14][15]. Due to these features, surviving detection could be employed for diagnostic purposes. Indeed, recent studies showed a possible role for Survivin as a sensitive marker for the noninvasive diagnosis of bladder cancer [16][17][18][19]. Urine Survivin resulted to be a strong, independent predictor of the presence of bladder cancer and higher tumor grade [19].
In recent years, a growing interest towards salivary markers led to the study of many molecules, looking for a potential candidate as a biomarker that could provide prognostic and/or diagnostic information in OSCC [20].
previously. Saliva samples were collected prior to any curative therapy (ie. surgery with or without radiotherapy or/and chemotherapy). No case in this study concerned patients with a history of prior malignancy, immunodeficiency, autoimmune disorders, hepatitis or HIV infection. Clinical data were reviewed to record sex and age of the patients, and site and size of the lesions. The group consisted of 38 males and 17 females with a mean range of 62 years (range 29-85). The characteristics of patients and their tumors are showed in Table 1.
In addition, 30 healthy individuals, 21 males and 9 females with a mean range of 52 years (range 31-69), were enrolled as control from patients who had undergone routine oral surgical procedures. All participants approved the informed consent to donate saliva for experiments.

Saliva collection and preparation
Patients were asked to refrain from eating, drinking, smoking and oral hygiene procedures for at least 1 hour prior to saliva collection [21]. 5 ml of whole saliva samples were collected from each subject. Firstly, they were asked to rinse their mouth with water and then expectorate saliva into 50 ml centrifuge tube. Collection tubes were immediately centrifuged at 5000 g for 5 minutes at +4°C to remove cell pellets and debris. Surnatants were then divided into 2 aliquots and stored at -80°C for subsequent analysis.

Enzyme-linked immunosorbent assay
Enzyme-linked immunosorbent assay (Total Survivin Titerzyme® Enzyme Immunometric Assay, Assay Designs, Ann Arbor, MI USA) was performed to determine survivin levels in the saliva samples of OSCC (n=55) and control patients (n=30). Undiluted samples were loaded, in duplicate, onto a 96-microtiter plate coated with antihuman Survivin monoclonal antibody. In order to quantify Survivin levels, in each experiment a standard curve was created with increasing quantities of purified recombinant Survivin. After incubation at room temperature on a plate shaker for 1 hour at 500 rpm, the excess samples or standards were washed out and a rabbit polyclonal antibody to Survivin was added.
After another 1-hour incubation, the excess antibody was washed out and goat anti-rabbit IgG conjugated to Horseradish peroxidase was added. After 30 minutes incubation, excess conjugate was washed out and 3,3' ,5,5'-tetramethylbenzidine substrate solution was added. After another 30 minutes incubation, the enzyme reaction was stopped and the colour generated was immediately read at 450 nm. The measured optical density was directly proportional to the concentration of Survivin in either standards or samples.

Statistical analysis
Data were analysed using GraphPad Prism software version 6.00 for Windows (GraphPad Software, San Diego California USA, www. graphpad.com). Significant differences (p<0.05) between groups were determined using one-way analysis of variance and the Student-Newman-Keuls-test. Contingency table and Fisher's exact test were used to assess the sensitivity, specificity, positive and negative predictive value of salivary Survivin test. All data were presented as mean ± standard deviation.

Results
Salivary Survivin levels in OSCC patients and in controls were measured using a specific ELISA test. The mean level of Survivin for OSCC patients (n=55) was 8.69 pg/ml ± 10.15, whereas it was 2.44 pg/ ml ± 4.22 for the control subjects (n=30). This difference in Survivin levels between OSCC cases and controls was statistically significant (p<0.05), as shown in Figure 1.
Using the value obtained from the average of Survivin levels in controls as cutoff, the sensitivity and specificity values of Survivin salivary levels in OSCC patients were 63.64% and 73.33% respectively, whereas positive and negative predictive value were 81.4% and 52.38%.
The association of Survivin levels in OSCC patients with age, gender, Grading, Staging and lymph node metastasis is shown in Table  2. Results showed that salivary levels of Survivin in OSCC patients were increased in advanced stages (III + IV) (8.13 pg/ml ± 10.94) compared to the early stages of disease (I + II) (7.09 pg/ml ± 8.59), even if this difference was not statistically significant.
The presence of lymph node metastases was associated with increased levels of Survivin (9.38 pg/ml ± 11.26 vs 5.77 pg/ml ± 8.21),

Discussion
OSCC is the leading malignant neoplasm in the oral cavity, with the 5-years survival rate of less than 50% [5]. Despite advances in oncology and surgery treatments, this trend remained unchanged over the last decades mainly due to diagnostic delay, which leads to make the diagnosis at an advanced stage, drastically reducing effectiveness of therapies and life expectancy [6]. Thus, the identification of possible markers for early cancer detection is still a mandatory issue.
Biomarkers for early cancer detection must fulfill several criteria: 1) it can be objectively measured; 2) must be measurable in small specimens; 3) must be altered in high-risk tissues, but not in normal tissues; 4) must be altered in early stages during cancer development [20]. Moreover, markers detectable also in body fluids (as blood, urine or saliva) would be preferable over tissue markers. Indeed, body fluids have several advantages that make them good choice for disease screening and diagnosis, like ease of collection, low-invasive procedure, low cost, and multiple sampling for monitoring over time [22]. Furthermore, saliva, over the other body fluids, has the following advantages: non-invasive easy collection and a low content in cells, DNA, proteins and inhibitory substances [23].
For the above stated reasons, in recent years, a growing interest towards salivary markers led to the study of many molecules, looking for a potential candidate as a biomarker that could provide prognostic and/or diagnostic information in OSCC [20].
Patients with premalignant oral lesions show levels of TNF-α, IL-1-α, IL-6 and IL-8 lower than the values measured in patients with OSCC, while the levels of these pro-inflammatory cytokines are higher compared to healthy control samples [25,29,31]. There are contrasting opinions about the differences in the expression of Survivin between normal mucosa and potentially malignant lesions. Some authors have found that Survivin expression is presented at a low rate both in normal oral mucosa and in leukoplakias [32]. Other surveys report significant difference between potentially malignant and normal mucosa [33]. Several authors agree in maintaining that Survivin expression rates can be considered a putative predictive value that appears among lesions evolving into OSCC compared with lesions not evolving in malignancies [34,35].
Strong evidence supports the general consensus that Survivin overexpression is a constant findings in OSCC, thus making Survivin a potential candidate as a useful biomarker in OSCC.
Hence, in the present study the possibility of using Survivin as salivary biomarker was evaluated. Indeed, Survivin resulted widely expressed in many types of cancers [10], including OSCC [13,32], while showed low or undetectable expression levels in most non-proliferating adult tissues [10].
In this work the salivary levels of Survivin in 55 patients with OSCC at different stages were compared with the levels of this protein in 30 healthy subjects. The results of the present study indicate that the Survivin levels in patients with OSCC are significantly higher than those of healthy subjects.
Furthermore, the differences in Survivin levels between early stages of OSCC and control subjects were statistically significant, confirming the potential usefulness of this protein as an early diagnostic tool.
Given that Survivin is a non-secreted protein, these findings suggest that the measured Survivin in saliva derives from the cancer cells released into saliva.
To the best of our knowledge, this is the first study dealing with salivary Survivin in OSCC. However, previous works on urine Survivin confirmed that cancer patients might have higher levels of detectable Survivin in body fluids respect to healthy control subjects [19]. Furthermore, it has recently been shown that concentrations of salivary MMP-1 and MMP-3 are higher in patients with OSCC than in samples from healthy controls and that these concentrations increase proportionally with increasing stage of the disease [36].
The salivary levels of Survivin results to be higher in advanced stages (III + IV) and in N+ cases, compared to the early stages (I + II) and N-cases, although these differences were not statistically significant. These results are in agreement with other studies that showed lack of correlation between tissue Survivin expression and staging or grading in OSCC [13]. It means that Survivin may be considered an independent predictor of the presence of oral cancer regardless of cancer stage and grade.
Considering the mean value of the Survivin levels of control subjects as the cut-off, the sensitivity and the specificity were respectively 63.64% and 73.33%. The same sensitivity value was reached in other works that have measured the levels of urinary Survivin in bladder cancer [19], even if specificity was lower. Differences in sample size, body fluids employed and type of cancer must be taken into account for explanation. Moreover, among other body fluids, saliva has the less standardized method of collection, thus may leading to variations in biomarkers quantification.
These considerations, combined with the limited number of patients enrolled in this study, may explain the high standard deviation found in OSCC patients. The presence of a high standard deviation can limit the diagnostic relevance of the measurement of salivary Survivin as single biomarker.
However, in the identification of potential salivary biomarkers, it was possible to reach higher values only by the use of genomic and proteomic technologies and the combination of a panel of different markers. Indeed, Li et al. reported that 7 transcripts, of the genes DUSP1, H3F3A, IL-1β, IL-8, OAZ1, SAT and S100P, were elevated in patients with OSCC and lower in healthy controls [37]. The combination of these biomarkers showed 91% of Sensitivity and specificity [37].
However, one of the major limits for the development of a salivary test there is no biologically justified cutoff value for levels of Survivin in patients with cancer, including oral cancer. Further studies in larger cohorts of patients would be necessary to establish biological and clinically relevant cut points.
Even considering the above cited limitations of the present study, Survivin seems to be an interesting salivary biomarker for noninvasive OSCC diagnosis. However, present data should be considered preliminary, and thus the use of Survivin as a single marker may be not sufficient for the early diagnosis of OSCC. Further investigations with larger samples and multiple biomarkers evaluation are required to clarify this issue.