Introduction

Traditional prognostic factors have been well established for prognosis in oral squamous cell carcinoma (OSCC) [1]. Beside advanced International Union Against Cancer (UICC) stages of the disease [2] one of the most important predictors of patient survival rates is lymph node metastasises [3–5]. Other prognostic factors have also been proposed in the literature. Hence, analyzing well-established clinicopathological parameters in association with other prognostic parameters concerning the inflammatory microenvironment has become an important issue in the management of OSCC [6–11].

The link between inflammation and the development of cancer is known since 1863 as Virchow hypothesized that the origin of cancer occurs at sites of chronic inflammation, in part based on his hypothesis that some classes of irritants, together with the tissue injury and ensuing inflammation they cause, enhance cell proliferation [7]. Following, the importance of molecular and cellular pathways linking cancer and inflammation has been clearly demonstrated [7–13]. Based on the assumption that the processes underlying such a response plays important roles in the progression of OSCC in the presence of a systemic inflammatory response has been thought to indicate poor prognosis in OSCC [10, 11].

In the pathogenesis of OSCC pretreatment measurement of elevated inflammatory serum C-reactive protein (CRP) levels [11, 14] and increased leucocytes/white blood cell count (WBC) [11, 15] are associated with adverse prognosis in OSCC. In addition, evidence for the use of other hematologic markers of inflammation as predictors of clinical outcome in OSCC is available. In this context pretreatment monocytes [16, 17], lymphocytes [17–19], neutrophils [17, 20, 21], basophils [22], eosinophils [22–25], platelets [26–28], neutrophil-to-lymphocyte ratio (NLR) [14, 29–40], derived NLR (dNLR) [36], lymphocyte-to-monocyte ratio (LMR) [19, 36, 41, 42], and platelet-to-lymphocyte ratio (PLR) [30, 32, 33, 35, 36, 43, 44] derived from the peripheral blood emerged as simple, easily accessible, and cost-effective tool screening cancer diseases. Moreover, the characterization of these novel markers is easily reproducible and they have the potential to identify patients at high risk for disease recurrence and tumor conditional death.

Concerning this survey the following questionnaires should be addressed: first, no data regarding basophils, eosinophils, dNLR, LMR, and PLR are disposable for OSCC to date. Second, cut-off values for all pretreatment single laboratory tests and calculated ratios have not been determined for OSCCs compared with healthy individuals as yet, which are strongly needed for a follow-up (monitoring [35, 45]) of cancer patients. Third, we analyzed CRP expression in tumor tissues as a potential source of serum CRP elevation that may identify a more aggressive biological phenotype of the disease [11]. Taken together, analyzing these data might be helpful for providing a practical guideline of further postoperative OSCC management.

Given this background the purpose of this study was to examine an extension of clinicopathological parameters for prognosis, follow-up (monitoring [35, 45]), and treatment of patients with OSCC compared with controls based on routine pretreatment single laboratory tests and calculated ratios that have been not determined as yet.

Materials and methods

Patients, blood samples, and tumor specimen

Written informed consent to participate was obtained from all patients (Ethics Committee Tuebingen, Germany, approval number: 562-2013BO2). We retrospectively reviewed the records of 146 patients with histopathological confirmed OSCC in our department between 2011 and 2015 and 93 healthy controls. Patients with preoperative antineoplastic therapies (chemoradiation) were excluded from the study. Further selective criteria for the patients were: complete clinical and laboratory data, no evidence of sepsis, no hematological disorders or treatment that may influence laboratory parameters, no autoimmune disease or treatment with steroids [36]. Tumor and patient characteristics are summarized in Table 1. The material was archival formalin-fixed, paraffin-embedded (FFPE) tissue from routine histopathologic work-up and had been performed under standardized conditions.

Table 1 Clinicopathological characteristics and prognostic factors in patients (n = 146) with OSCC
Full size table

Blood samples (2.7 ml) were collected in ethylenediaminetetraacetic acid (EDTA) treated vials prior to surgery or palliative treatment. The blood samples were blinded to the clinical data and analyzed by standard laboratory techniques to determine pretreatment CRP value, leucocyte/WBC count, monocyte count, lymphocyte count, neutrophil count, basophil count, eosinophil count, and platelet count. Moreover, pretreatment NLR (quotient of neutrophil count to lymphocyte count), dNLR (quotient of WBC count–neutrophil count to lymphocyte count), LMR (quotient of lymphocyte count to monocyte count), and PLR (quotient of platelet count to lymphocyte count) were calculated from peripheral blood cell count.

For immunohistochemistry of CRP staining, the records of healthy individuals (normal oral mucosal tissues, n = 14), patients with oral precursor lesions (simple hyperplasia, n = 21; squamous intraepithelial neoplasia SIN I, n = 5; SIN II, n = 9; SIN III, severe dysplasia, n = 10; SIN III, carcinoma in situ, n = 11), and patients with invasive OSCC (n = 46/146) were retrospectively assessed if available and the patient agreed to tissue investigation [46]. The diagnosis of normal oral mucosal tissues, precursor lesions, and invasive squamous cell carcinoma was confirmed by the Department of Pathology, University Hospital Tuebingen. Tumor blocks of FFPE tissue were selected by experienced pathologists, based on routine H&E stained sections. Oral precursor lesions were classified according to WHO criteria [47]. Tumor staging was performed according to the 7th edition of the TNM staging system by the UICC/AJCC of 2010 [48]. Grading of OSCC was defined according to WHO criteria [49].

Staining procedure and quantification of immunohistochemistry

We stained for CRP (LSBio, Eching, Germany, mouse mAb, LS-B7859, dilution 1:500, 5 µg/ml) in tissue sections. Staining was performed on serial sections of 2 µm thickness, which were deparaffinized in xylene and ethanol and rehydrated in water. Heat induced epitope retrieval (HIER) was performed with citrate buffer pH 6.0 (Dako, Hamburg, Germany). Endogenous peroxidase activity was quenched with 0.3 % hydrogen peroxide. Endogenous biotin activity was blocked using the streptavidin/biotin blocking kit (Vector Laboratories, Burlingame, CA, USA). After incubation with the primary or mouse control antibody (BD Pharmingen, Heidelberg, Germany [50]) the Dako LSAB2 peroxidase System (Dako, Hamburg) was used. Slides were subsequently incubated for 3–5 min in DAB (3,3′-diaminobenzidine, Biogenex) counterstained with haemalaun and mounted with Glycergel (Dako).

Five representative high power fields (1 HPF = 0.237 mm2, original magnification: 200×-fold) were analyzed for CRP expression in normal tissue, oral precursor lesions, tumor tissue and averaged, respectively. The extent of the staining, defined as the percentage of positive staining areas of tumor cells in relation to the whole tissue area, was semi-quantitatively scored on a scale of 0–3 as the following: 0, <10 %; 1, 10–30 %; 2, 30–60 %; 3, >60 %. The intensities of the signals were scored as 1+, 2+, and 3+. Then, a combined score (0–9) for each specimen was calculated by multiplying the values of these two categories [51]. Cases were classified as negative, 0 points, or positive, 1–9 points. Liver tissues were used as a representative positive controls. Two observers blinded to the diagnosis performed scoring on identical sections marked by circling with a water-resistant pencil and finally with diamond-tipped pencil on the opposite side of the microscopic slide. Pictures were analyzed using a Canon camera (Krefeld, Germany). The photographed images were imported into the Microsoft Office Picture Manager.

Statistical analysis

Statistical analysis was performed with MedCalc Statistical Software version 15.11.0 (MedCalc Software bvba, Ostend, Belgium; https://www.medcalc.org; 2015).

Receiver operating characteristic curves (ROC) [11] were plotted to determine the best cut-off ranges with relevant sensitivities and specificities for OSCC group compared with controls screening for each single value or ratio [52]. Area under the curve (AUC) analysis was determined for quality measurement of the classifier. The cut-off point was determined as the value corresponding with the highest diagnostic average of sensitivity and specificity (highest diagnostic accuracy/Youden’s index). These values were graphically displayed in an interactive dot diagram to study the accuracy of each diagnostic test. Based on resulting sensitivity and specificity the likelihood ratios (LRs) were calculated +LR = sensitivity/(1 − specificity) and −LR = (1 − sensitivity)/specificity. LRs were used to assess how good the single values and ratios are to determine OSCC disease. LR is the ratio between the probability of a positive test (positive likelihood ratio, +LR) or a negative test (negative likelihood ratio, −LR) result given the presence of OSCC disease and the probability of a positive or negative test result given the absence of OSCC disease.

For all data, the Kolmogorov–Smirnov test was applied to test for a normal distribution. Chi square test (χ 2) and Fisher’s exact test were used to investigate the relation between two categorical variables. Data were analyzed using the Student’s t test when means of two groups were compared. Correlation between two variables was judged by the Pearson’s correlation (Rr) coefficient test. The trend lines of the correlation analysis were displayed in a scatter diagram and plotted by local regression smoothing (LOESS). Regression analysis was used to describe the relationship between two variables and to predict one variable from another. The coefficient of determination R 2 is the proportion of the variation in the dependent variable explained by the regression model. It can range from 0 to 1 and is a measure of the goodness of fit of the model.

Mean values, median values, and ROC analysis results were given with 95 % confidence intervals (CI). All p values presented were two-sided and p < 0.05 was considered statistically significant.

Results

Evaluation of pretreatment single laboratory tests and calculated ratios for screening OSCC disease compared with controls

ROC analysis determined cut-off values for single laboratory tests and calculated ratios in OSCC compared with controls. A significant cut-off value has been determined for CRP value, WBC count, monocyte count, lymphocyte count, neutrophil count, NLR, dNLR, LMR, and PLR. The cut-off points that gave the best sensitivity and specificity for the diagnosis of OSCC were evaluated using AUC analysis (Fig. 1). Additionally, values were graphically displayed in an Interactive dot diagram to show and to control the highest diagnostic accuracy of each diagnostic test (Fig. 2). A comparison of variables, calculated sensitivities and specificities between the parameters is presented in Table 2.

Fig. 1
figure 1

Receiver operating characteristic curves (ROC) of single laboratory tests (CRP, WBC count, platelet count a; neutrophil count, basophil count, eosinophil count b; monocyte count and lymphocyte count c) and calculated ratios (NLR, dNLR, LMR, PLR, d). The true positive rates (sensitivity) are plotted in function of the false positive rate (100-specificity) analyzing different cut-off points with highest diagnostic accuracy to distinguish controls (n = 93) from OSCC patients (n = 146). The corresponding test characteristics sensitivity and specificity are given in Table 2

Full size image
Fig. 2
figure 2

Interactive dot diagrams (part of ROC curve analysis, Fig. 1) of calculated ratios (NLR a, LMR b, PLR, c) of controls (n = 93) and OSCC group (n = 146) are displayed as dots on two vertical axes. The red horizontal lines indicate the cut-off points with the best separation/highest accuracy (minimal false negative and false positive results) between controls and OSCC group. The corresponding test characteristics sensitivity and specificity are given in Table 2

Full size image
Table 2 Comparison of calculated sensitivities, specificities, positive likelihood ratios, and negative likelihood ratios of single laboratory tests and calculated ratios in OSCC (n = 146) compared with controls (n = 93)
Full size table

Table 3 shows mean values of pretreatment single laboratory tests and calculated ratios for screening of OSCC disease compared with controls.

Table 3 Comparison of mean values of single laboratory tests and calculated ratios in OSCC (n = 146) and controls (n = 93)
Full size table

Association of clinicopathological characteristics with single laboratory tests and calculated ratios in patients with OSCC

Older patients (≥64 years) were significantly associated with decreased lymphocyte count (≤1.41 × 103/µl; p = 0.0029) and increased NLR (>2.68; p = 0.0063).

Males had a significantly increased monocyte count (>0.42 × 103/µl; p = 0.0026), NLR (>2.68; p = 0.0049), dNLR (>1.99; p = 0.0251), and a decreased LMR (≤4.1; p = 0.0009).

Patients with alveolar ridge carcinoma and section overlapping tumors had a significantly decreased lymphocyte count (≤1.41 × 103/µl; p = 0.0085) compared with patients with carcinoma of the palate and floor of the mouth.

Advanced tumor stages UICC III/IV were significantly associated with increased WBC count (>8.5 × 103/µl; p = 0.0370) and neutrophil count (>5.36 × 103/µl; p = 0.0008).

Nicotine abuse was significantly associated with males (p = 0.0001), increased WBC count (>8.5 × 103/µl; p < 0.0001), neutrophil count (>5.36 × 103/µl; p = 0.0008), alcohol intake (p = 0.0001), and weakly significant associated with increased monocyte count (>0.42 × 103/µl; p = 0.0710), NLR (>2.68; p = 0.0795), and decreased LMR (≤4.1; p = 0.0970) but not with elevated CRP values (>1.5 mg/l; p = 0.1943).

Alcohol abuse was significantly associated with cervical lymph node metastasis pN+ (p = 0.0206), advanced tumor stages UICC III/IV (p = 0.0388), and weakly significant associated with advanced tumor size pT3/4 (p = 0.0826) but not with elevated CRP values (>1.5 mg/l; p = 0.3359).

Synchronous nicotine and alcohol abuse were significantly associated with males (p = 0.0298), advanced tumor size pT3/4 (p = 0.0258), advanced tumor stages UICC III/IV (p = 0.0214), weakly significant associated with advanced grading (G3/4 = 0.0997), and cervical lymph node metastasis pN+ (p = 0.0732).

To assess a trend of clinicopathological characteristics with single laboratory tests and calculated ratios weak associations (p = 0.05–0.1) were further described as the following: advanced grading (G3/4) was weakly associated with increased WBC count (>8.5 × 103/µl; p = 0.0997) and advanced tumor stages UICC III/IV have been weakly associated with increased monocyte count (>0.42 × 103/µl; p = 0.0915). Cervical lymph node metastasis (cN+) was weakly associated with increased neutrophil count (>5.36 × 103/µl; p = 0.0836) and PLR (>143; p = 0.0623). An increased neutrophil count (>5.36 × 103/µl) has been weakly associated with males (p = 0.0888) and advanced tumor size (pT3/4; p = 0.0565).

Association of pretreatment single laboratory tests and calculated ratios in OSCC

Significantly associated parameters of pretreatment laboratory tests and calculated ratios (Tables 4, 5, 6) were correlated and graphical displayed for a trend (Fig. 3).

Table 4 Association of pretreatment single laboratory tests in OSCC (n = 146)
Full size table
Table 5 Association of pretreatment laboratory tests with calculated ratios in OSCC (n = 146)
Full size table
Table 6 Association of pretreatment calculated ratios in OSCC (n = 146)
Full size table
Fig. 3
figure 3

Correlation analysis of calculated ratios (a–c) in OSCC patients (n = 146) with trend lines. NLR is significant directly correlated with PLR (a). LMR is significant inversely correlated with NLR (b) and PLR (c). The trend lines are plotted by local regression smoothing (LOESS). The degree of smoothing is controlled by the span (80 %) which is the proportion (expressed as a percentage) of the total number of points that contribute to each local fitted value

Full size image

NLR was significant directly correlated with WBC count (Rr = 0.3764, p < 0.0001; R 2 = 0.1417; p < 0.0001), neutrophil count (Rr = 0.5740, p < 0.0001; R 2 = 0.3295; p < 0.0001), and monocyte count (Rr = 0.2085, p = 0.0121; R 2 = 0.0434; p = 0.0121), but significant inversely correlated with lymphocyte count (Rr = −0.5345, p < 0.0001; R 2 = 0.2857; p < 0.0001). LMR was significant inversely correlated with neutrophil count (Rr = −0.2522, p = 0.0023; R 2 = 0.0636; p = 0.0023) and monocyte count (Rr = −0.5668, p < 0.0001; R 2 = 0.3213; p < 0.0001).

LMR was significant directly correlated with lymphocyte count (Rr = 0.5470, p < 0.0001; R 2 = 0.2992; p < 0.0001). PLR was significant inversely correlated with lymphocyte count (Rr = −0.5205, p < 0.0001; R 2 = 0.2709; p < 0.0001) (Fig. 2).

NLR was significant directly correlated with PLR (Rr = 0.5552, p < 0.0001; R 2 = 0.3082; p < 0.0001). LMR was significant inversely correlated with NLR (Rr = −0.5996, p < 0.0001; R 2 = 0.3595; p < 0.0001) and PLR (Rr = −0.3903, p < 0.0001; R 2 = 0.1523; p < 0.0001) (Fig. 2).

A combination of the ratios or a combination with CRP using a combinatory index was not beneficial for improved association with any clinicopathological characteristics (data not shown).

Immunohistochemistry of CRP expression in the carcinogenesis of OSCC

Liver tissues stained positive for CRP. However, CRP expression was not found in normal oral mucosa, oral precursor lesions, or OSCC specimen.

Discussion

This is the first study focusing on the development for cut-off values based on single laboratory results and calculated ratios that might serve as an extension to clinicopathological parameters for prognosis, treatment, and monitoring [35] of patients with OSCC. Moreover, the calculated baselines can be used as a goal for successful immunotherapies in the future.

Evidence for the use of inflammatory hematologic markers as predictors of clinical outcome in OSCC and head and neck squamous-cell carcinoma (HNSCC) are available. Increased CRP values [11, 14], leucocytosis [11, 15], monocytosis [16], lymphopenia [29], neutrophilia [20, 21], and elevated NLR [14, 29] have been associated with reduced tumor-specific survival in OSCC. However, no data regarding OSCC survival are available for decreased LMR and increased PLR but have been described in the pathogenesis of HNSCC [19, 30, 35, 41]. Lymphopenia [29] was demonstrated by our previous data [18] in OSCC patients and was confirmed in this study. Our results showed that older patients (≥64 years) were more frequently associated with lymphopenia, which is in accordance with previous published data in tumor patients [53], including HNSCC [17]. Therefore, the issue of immunosuppression seems to be important for the carcinogenesis of OSCC. As a consequence of an inadequate immunologic reaction with subsequent weakened defense against cancer interleukin-2 (IL-2) cytokine-based immunotherapies designed to increase the number and biological activity of circulating lymphocytes were introduced in renal cancer [54], metastatic melanoma [55], and have been suggested for OSCC [56]. Lymphocytes are very important cellular components of the immune system that are crucial for activation of an effective antitumor response [57], the destruction of residual tumor cells, as well as related micrometastasis [58, 59]. Due to sustained activation of T cells in cancer patients [60] tumor-infiltrating T lymphocytes could help to drive tumor cells towards apoptosis and by presenting tumor-associated antigens to immune cells lymphocytes lead to death of cancer cells in response to chemoradiation (CRT) [12, 61]. Hence, lymphocytes are very crucial for improvement of adjuvant therapies and preventing tumor recurrence.

For the first time, similar to the survival data we defined cut-off values for all single laboratory results and calculated ratios compared with controls. These data are strongly needed for a follow-up of cancer patients, which were measured for each parameter by ROC analysis and have not been determined for all cancer diseases at all. Indeed, significant cut-off values for single laboratory parameters (CRP, WBCs, monocytes, lymphocytes, neutrophils) and calculated ratios (NLR, dNLR, LMR, PLR) were found in OSCC patients compared with controls. In the clinical course of carcinogenesis the relation of calculated ratios could be useful for predicting tumor recurrence and probably tumor disease. Compared with single laboratory tests calculated ratios were superior in measuring sensitivity and specificity of OSCC disease. In our study, we found that NLR was significant directly associated and correlated with PLR. Additionally, LMR was significant inversely associated and correlated with NLR and PLR. This link might be helpful to evaluate the clinical course (follow-up/monitoring [35, 45]) of OSCC patients. However, compared with others [14, 35] a combination of the ratios or a combination with CRP using a combinatory index was not beneficial for improved association with any clinicopathological characteristics.

In patients with breast cancer NLR was shown to be better than dNLR in terms of predicting prognosis [62], which confirms our results of NLR to be a more sensitive and specific marker for OSCC than dNLR.

Various outcome thresholds for these ratios were identified analyzing the relative risk of OSCC and HNSCC tumor recurrence: NLR >1.99 [29] (OSCC), >2.44 [14] (OSCC); LMR <5.07 [19] (HNSCC), <5.2 [41] (HNSCC); PLR >167 [30] (HNSCC), >170 [35] (HNSCC). In our case control study, we found NLR (>2.68), LMR (≤4.1), and PLR (>143) values comparable with previous published results of survival data in OSCC and HNSCC. However, this is the first study describing LMR and PLR cut-off values in OSCC compared with controls. To the best of our knowledge, neither data regarding survival nor data comparing tumor patients with controls are available for OSCC to date. Drawing a conclusion for this question of different cut-off values in the follow-up period (monitoring [35, 45]) the clinical course of ratios (relation of NLR, LMR, PLR) indicated by the scatter plots will be crucial, whereby clinicians are principally involved. For example, in cases of tumor recurrence an increase of NLR with PLR in parallel to a decrease of LMR would be expected. Using NLR and PLR the data published by Rassouli et al. [35] were at least as good as TNM staging system in predicting survival of HNSCC patients. Moreover, predicting response to CRT is an important issue in the treatment of cancer. The results showed by Lin et al. [19] demonstrated that increased LMR is associated with better prognosis in patients with newly diagnosed metastatic nasopharyngeal carcinoma receiving chemotherapy. He et al. [39] reported that pretreatment NLR, percentages of lymphocytes and neutrophils are independent prognostic factors for survival for HNSCC patients undergoing radiation or chemoradiation. Similar to this, An et al. [40] demonstrated NLR as a significant predictor of both survival and response to CRT in nasopharyngeal carcinoma. In patients with gastric cancer the normalization of NLR and PLR after chemotherapy was found to be associated with significant improvement in progression free survival [45]. The pretreatment NLR and PLR represented significant prognostic indicators of survival in patients treated for early-stage non-small-cell lung carcinoma with stereotactic radiation [63]. Determined by NLR dominant pro-tumor activities of neutrophils or reduced anti-tumor immune response by lymphocytes may have an impact on poor tumor response to preoperative CRT and unfavorable prognosis in rectal cancer patients [64]. The results published by Huang et al. demonstrated that neutrophils and monocytes appear to have a strong impact on radiation outcome in HNSCC [17]. Thus, as a supplement using systemic inflammatory markers as single laboratory parameters or as calculated ratios provide a clear rationale for predicting surgical outcome and/or CRT response. Our results strengthen the evidence suggesting that neutrophils, monocytes, or lymphocytes of the systemic inflammatory response and anti-tumor response represent targets for novel therapeutic strategies. The utility of NLR, LMR, and PLR in the posttreatment management (monitoring [35, 45]) of OSCC patients including surgical outcome and/or CRT response have currently not been determined.

Several explanations for the relation of increased NLR, PLR and decreased LMR with poor prognosis in cancer are available. Neutrophils contain and secrete various cytokines including IL-8, vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), fibroblast growth factor (FGF), matrix metalloproteinases (MMPs), and elastases [17, 20, 21, 65–68]. These cytokines create a microenvironment for tumor cell dissociation, endothelial cell migration, and extracellular matrix remodelling [17, 20, 21, 65–68]. Subsequently, the cytolytic activity of activated lymphocytes (T cells) and natural killer cells (NK) is suppressed and the degree of suppression is closely associated with the number of neutrophils [69, 70]. Therefore, neutrophilia may cause immunosuppression and can sustain tumor growth. On the other hand, increased NLR could result from lymphopenia, which impair the role of cell-mediated immunity and its function of host cancer cell destruction [71]. Finally, our data analyzed in OSCC patients confirmed the findings of other tumor entities demonstrating neutrophilia, lymphopenia, and increased NLR with strong prognostic impact on survival or treatment outcome results.

In OSCC conflicting data are available concerning thrombocytosis [27, 28], which has been associated with higher levels of IL-6 as one major cytokine of inflammation [72]. This cytokine was linked with various tumor-promoting activities in the tumor microenvironment, such as angiogenesis and tumor cell proliferation [73]. Hence, in our study, no indication of thrombocytosis has been found. Conversely, increased PLR could result from lymphopenia. Therefore, the significant increased PLR is rather the consequence of lymphopenia than the result of thrombocytosis in OSCC patients.

Beside lymphopenia the decreased LMR can be attributed to a significant monocytosis found in OSCC patients. The study conducted by Tsai et al. [16] showed that leucocytosis, monocytosis, and neutrophilia were associated with advanced OSCC tumor stages and poor tumor differentiation. The monocyte count was also increased in those patients with lymph node metastasis. Moreover, the pretreatment circulating monocyte count was an independent prognostic factor for worse OSCC specific survival [16]. This is in accordance with our findings showing leucocytosis, neutrophilia, and monocytosis to be associated with advanced tumor stages. Additionally, neutrophilia was associated with advanced tumor size and cervical lymph node metastasis, and both, neutrophilia and monocytosis with male gender. In the context of carcinogenesis monocytes (and tumor-associated macrophages, M2 [74]) secrete several proinflammatory cytokines such as IL-1, IL-6, tumor necrosis factor-alpha (TNF-α) [75], and immunosuppressive anti-inflammatory IL-10 [74, 76], which have been associated with poor clinical outcome in malignancies [75–78]. Therefore, neutrophils and monocytes acting the opposite role as lymphocytes and seem to promote carcinogenesis.

The increased NLR and decreased LMR in male gender is attributed to the detected neutrophilia and monocytosis, which have been both associated with nicotine abuse. Therefore, smoking may contribute to a systemic inflammatory reaction supporting tumor progression. In general, social habits like nicotine and alcohol abuse were associated with advanced clinicopathological characteristics, e.g. advanced tumor stages and confirm previous published data [79]. Smoking is associated with a broad range of alterations in systemic immune and inflammation marker levels including accelerated erythropoiesis, thrombocytosis, leukocytosis [80] specifically with chronic neutrophilia [81], and elevation of CRP levels [82]. Beside platelets (secreting IL-6) and neutrophils (secreting IL-8), which are attracted by the inflammatory microenvironment to the tumor, cancer cells including OSCC have been shown to secrete IL-6 [83] and IL-8 [84] and in turn induce the production of CRP [85, 86]. These mechanisms imply that increased CRP levels are a response to the neoplastic process and/or the systemic inflammatory response of leucocytes (e.g. neutrophils) to the tumor. Measurement of both, IL-6 and IL-8 have been associated with reduced tumor-specific survival in OSCC [87] and are worth for standardized analysis in further surveys. Although some tumors have been shown to express CRP [88, 89] our data didn´t reveal tumor cells or precancerous lesions as a potential source of elevated CRP expression in OSCC patients. Our data support the suggestion of CRP as an important marker for monitoring of the disease although the cut-off value has been determined at a low level but confirm our previous published results [11]. Increased serum CRP levels were significantly associated and correlated with elevated blood leucocytes/neutrophils but they were not associated with calculated ratios (NLR, dNLR, LMR, PLR). Therefore, increasing numbers of leucocytes and more specific neutrophils together with serum CRP reflect a systemic inflammatory response, which could be useful predicting tumor disease and probably tumor recurrence after excluding other criteria (e.g. bacterial infection) in the after-care. Elevated CRP values were not associated with social habits (nicotine and alcohol abuse) and therefore they could be attributed as a consequence to the inflammatory cancer microenvironment.

Basophils [22] and eosinophils [22–25] in the peripheral blood have not been determined for OSCC as yet. The clinical relationship between eosinophilia and basophilia with patients’ outcome is not really clear. In patients of myelodysplastic syndromes eosinophilia and basophilia have been significantly associated with reduced survival [22]. In our study, basophils and eosinophils were not associated with any clinicopathological characteristics, single laboratory results, or calculated ratios. Therefore, these parameters seem to be not relevant for monitoring of OSCC patients.

Conclusions

This study highlights the first analysis for cut-off values of pretreatment single laboratory tests and calculated ratios, which are strongly needed for a follow-up of cancer patients. Additionally, the calculated baselines can be used as a goal for successful immunotherapies in the future. The links between NLR, LMR, and PLR might be helpful for the clinical course (monitoring [35, 45]) of cancer patients and have been first described for OSCC in this study. Taken together, analyzing these data provides an additional practical guideline of further postoperative OSCC management.