The aim of this study was to evaluate the association of preoperative plasma fibrinogen levels with clinico-pathological parameters and disease-free survival in patients with oral tongue squamous cell carcinoma (OTSCC). We retrospectively studied 76 patients with OTSCC who underwent a partial glossectomy only, at a single centre, between 1996 and 2007. Among the 76 patients, 30 eventually developed cervical metastasis. Preoperative plasma fibrinogen levels were determined and correlated with clinico-pathological findings by t -test or analysis of variance methods. Univariate and multivariate analyses were used to determine the association of preoperative plasma fibrinogen levels and disease-free survival. Elevated levels of plasma fibrinogen were positively related with growth type ( P < 0.001), differentiation ( P < 0.001), thickness ( P < 0.001), and the infiltrative growth ratio ( P = 0.032). Univariate analysis showed that growth type ( P < 0.001), differentiation ( P < 0.001), thickness ( P < 0.001), and preoperative plasma fibrinogen levels ( P < 0.001) were significantly correlated with disease-free survival. Multivariate analysis showed that the plasma fibrinogen level remained an independent factor for disease-free survival after partial glossectomy for OTSCC ( P = 0.029). A high preoperative plasma fibrinogen level is an independent predictor of cervical metastasis after partial glossectomy for OTSCC. A conservative supraomohyoid neck dissection is appropriate in patients with stage I/II carcinoma of the tongue whose preoperative plasma fibrinogen is >300 mg/dl.
The objective of treatment in patients with stage I/II carcinoma of the tongue is to control the primary tumour and regional neck metastasis. Several types of treatment are used to control metastasis: resection of the primary tumour with or without elective neck treatment (dissection or irradiation) and radiotherapy to the primary tumour with or without elective neck treatment. However, cervical micrometastasis can be found in approximately 40% of early T1–2 oral tongue squamous cell carcinomas (OTSCC). Micrometastasis is not detectable using the best contemporary diagnostic technology. Loco-regional recurrences are the main cause of treatment failures in oral tongue carcinoma.
What features can be used to distinguish tumours that metastasize to the neck after surgery from those that do not? A better prognostic evaluation system is necessary to guide clinical management, particularly in the prediction of subclinical nodal metastasis. Several studies have shown that tumour thickness and vascular invasion are risk factors for cervical metastasis in patients with squamous cell carcinoma of the head and neck. Other predictive models have been reported for head and neck cancers over the years, including tumour shape (reductive or expansive), tumour growth pattern (exophytic or endophytic), various pathological features, and various invasive front malignancy grading scores. There have been very few reports of biomarkers in serum related to OTSCC. Increased serum prolactin and a marker for fibrinogen degradation products in the serum, AMDL DR-70, have been associated with a poor prognosis in tongue cancer. Many of these studies have used pooled data from heterogeneous sites, stages, and treatments. However, there is no consensus on the best prognostic model for early stages I and II oral tongue carcinoma.
Recent studies in humans have shown that elevated fibrinogen plasma levels are a useful prognostic parameter for several human malignancies, such as gastric cancer, cervical cancer, ovarian cancer, and kidney cancer. However, whether preoperative plasma fibrinogen could be used to predict cervical metastasis in patients with stage I/II carcinoma of the tongue has not been investigated. The purpose of this study was to evaluate the association of preoperative plasma fibrinogen levels and clinico-pathological parameters with cervical metastasis in patients after resection of the primary tumour without elective neck dissection for stage I/II OTSCC.
Patients and methods
We reviewed the medical records of 327 patients who were treated for OTSCC in a department of oral and maxillofacial surgery and a department of head and neck surgery between 1997 and 2007. Computed tomography (CT) or magnetic resonance imaging (MRI) was used to supplement the clinical evaluation and for staging the primary tumour and regional lymph nodes. Only patients with clinical T1N0M0 or T2N0M0 who had primary surgical treatment without elective neck dissection were recruited. Moreover, no patients received either preoperative or postoperative treatment. There were no recurrences at the primary site. Finally, 76 patients were selected for the study: 30 who developed cervical metastases postoperatively and 46 who had no evidence of disease for 5 years after surgery. The preoperative clinical American Joint Cancer Committee/Union Internationale Contre le Cancer (AJCC/UICC) TNM stages (1997) were 32 cT1N0M0 and 44 cT2N0M0. The median patient age was 61 years (range 29–85 years), and the male to female ratio was 2.3:1 (53 men and 23 women).
A peripheral blood sample was collected 48–72 h before the operation, and the plasma fibrinogen was measured by Clauss clotting method using DADE Thrombin ReagentP and a Sysmex CA-7000 automated coagulometer (Sysmex, Kobe, Japan). A plasma fibrinogen level between 200 mg/dl and 300 mg/dl was considered to be normal.
All pathology slides were reviewed at the maximum cross-section by two pathologists who had no knowledge of the clinical data. The grading scheme of Crissman et al. was used for nuclear polymorphism. Tumour cells were graded in the deep portion. The growth type and infiltrative growth ratio were examined to assess the mode of invasion. The infiltrative growth ratio was calculated as the percentage of the tumour area comprised of small nests of 10 or fewer cells at the periphery of the tumour to the total area of the tumour. Regardless of whether the growth type of the carcinoma was exophytic or invasive, tumour thickness was measured from the surface of the normal mucosa to the deepest portion of the tumour. Vascular invasion and lymphatic vessel invasion were evaluated to assess the cellular response of the host. Perineural invasion was also recorded.
Preoperative plasma fibrinogen values were recorded as the mean ± standard deviation (SD). The correlation between preoperative plasma fibrinogen and clinico-pathological factors was evaluated by unpaired t -test and one-way analysis of variance (ANOVA) with the least significant difference (LSD) multiple comparison, where appropriate. Testing of homogeneity variance was performed between fibrinogen levels ≤300 mg/dl and >300 mg/dl ( P = 0.23).
The disease-free survival rate was calculated by Kaplan–Meier method and differences between groups were tested with the log-rank test. Disease-free survival times of patients were censored at the last follow-up date. Univariate analysis comprised sex (male vs. female), age (≤70 vs. >70 years), growth type (exophytic vs. invasive), T-stage (T1 vs. T2), differentiation (well vs. moderately vs. poorly differentiated), thickness (≤4 mm vs. >4 mm), nuclear polymorphism of the deep portion (few, moderate vs. markedly enlarged), infiltrative growth ratio (≤20% vs. >20%), vascular invasion (absent vs. present), lymphatic vessel invasion (absent vs. present), perineural invasion (absent vs. present), and preoperative plasma fibrinogen level (≤300 mg/dl vs. >300 mg/dl). All the variables observed to be statistically significant in the univariate analysis were included in the multivariate analysis (Cox proportional hazards model) by the ‘enter’ method for variable selection. P -values of <0.05 were considered statistically significant. Statistical analyses were performed and graphics drawn using the SPSS 13.0 statistical software package (SPSS Inc., Chicago, IL, USA).
The mean preoperative plasma fibrinogen level was 327.2 ± 93.3 mg/dl. The clinico-pathological parameters assessed included sex, age, growth type, T-stage, differentiation, thickness, nuclear polymorphism of the deep portion, infiltrative growth ratio, vascular invasion, lymphatic vessel invasion, perineural invasion, and preoperative plasma fibrinogen levels. Associations between preoperative plasma fibrinogen levels and clinico-pathological parameters are shown in Table 1 . Elevated plasma fibrinogen levels were associated with growth type ( P < 0.001), differentiation ( P < 0.001), thickness ( P < 0.001), and the infiltrative growth ratio ( P = 0.032), but were not significantly associated with T-stage ( P = 0.142), nuclear polymorphism of the deep portion ( P = 0.581), vascular invasion ( P = 0.133), lymphatic vessel invasion ( P = 0.508), or perineural invasion ( P = 0.581). Patients who had cervical metastasis during follow-up had higher preoperative plasma fibrinogen levels.
|Variable||Number of patients||Mean preoperative plasma fibrinogen level (mg/dl)||P -value|
|Female||23||332.6 ± 86.4|
|Male||53||324.9 ± 109.6|
|≤70||53||327.1 ± 94.2|
|>70||23||327.4 ± 93.4|
|Exophytic||33||281.9 ± 81.3|
|Invasive||43||361.9 ± 87.5|
|Well differentiated||39||286.9 ± 80.5|
|Moderately differentiated||34||363.5 ± 88.8|
|Poorly differentiated||3||439.0 ± 25.8|
|≤4||31||269.8 ± 72.9|
|>4||45||366.7 ± 85.5|
|T1||39||311.8 ± 101.7|
|T2||37||343.4 ± 81.9|
|Nuclear polymorphism of the deep portion||0.581|
|Few, moderate||50||322.9 ± 90.7|
|Markedly enlarged||26||335.5 ± 99.5|
|Infiltrative growth ratio||0.032|
|≤20%||54||312.6 ± 91.2|
|>20%||22||363.1 ± 90.7|
|Absent||70||272.1 ± 64.7|
|Present||6||331.9 ± 94.2|
|Lymphatic vessel invasion||0.508|
|Absent||72||296.9 ± 94.8|
|Present||4||328.9 ± 94.8|
|Absent||55||317.5 ± 87.3|
|Present||21||330.9 ± 96.0|
|Absent||46||292.5 ± 84.9|
|Present||30||380.4 ± 80.5|
The 5-year disease-free survival rate was 39.5% for all 76 patients. In the univariate analysis, significant prognostic factors included growth type ( P < 0.001), differentiation ( P < 0.001), thickness ( P < 0.001), preoperative plasma fibrinogen level ( P < 0.001), Nuclear polymorphism of the deep portion ( P < 0.001) and infiltrative growth ratio ( P < 0.001) ( Table 2 ). Kaplan–Meier curves for disease-free survival according to the different preoperative plasma fibrinogen levels (≤300 mg/dl and >300 mg/dl) are shown in Fig. 1 .
|Variable||Number of patients||5-year disease-free survival rate (%)||P -value|
|Nuclear polymorphism of the deep portion||<0.001|
|Infiltrative growth ratio||<0.001|
|Lymphatic vessel invasion||0.571|