A retrospective study of maxillary squamous cell carcinoma (SCC) patients attending a department of oral and maxillofacial surgery was performed for the period 2000–2010. The clinical information of 100 cases treated during this period was acquired and analyzed. Patient survival was calculated using the Kaplan–Meier method. For these 100 cases, the total metastatic rate was 34.0% and occult metastatic rate was 27.5%. Positive lymph nodes were mostly detected at levels I–III. There was no significant difference in metastatic rate between the primary sites of maxillary gingiva and hard palate. Tumours involving the gingival–buccal sulcus presented a significantly higher risk of metastasis. Advanced stage (T3/4) was significantly correlated with a higher metastasis rate. The pathological grade also showed a significant relationship with metastasis. Twenty-four patients presented regional recurrence. Elective neck dissection could significantly reduce the recurrence rate. The overall 3-year and 5-year survival rates were 66.3% and 56.7%, respectively. Both the T and the N stages had a significant impact on survival rates. Selective neck dissection from level I to III is recommended for T3/4 stage cN0 patients, especially those with gingival–buccal sulcus involvement.
Squamous cell carcinoma (SCC) is the most common malignant tumour of the oral cavity, accounting for more than 80% of all oral cancers. Cervical metastasis is one of the well-known behaviours of oral SCC, and it may have a distinct influence on the prognosis and clinical outcome for the patient. It is well documented that oral SCCs of the tongue, floor of the mouth, and mandibular gingiva have a strong tendency for cervical metastasis. Elective neck dissection (END) is already well accepted in these patients. However, the management of the cN0 neck patient with hard palate, maxillary alveolar, or gingival SCC remains on a ‘watch and wait’ basis due to the low metastatic rate. Only recently have studies focused on the cervical metastasis of maxillary SCC. However, prospective and evidence-based studies are still lacking and the treatment of the clinical negative neck remains controversial.
The aim of this retrospective study was to determine the incidence of cervical metastasis of SCC of the maxilla and to define the risk factors and outcome of cervical metastasis. We also sought to propose recommendations in relation to the treatment strategy for the clinically negative neck.
Patients and methods
A series of cases of SCC originating from the hard palate and maxillary alveolus or gingiva, treated in the department of oral and maxillofacial surgery of the university hospital between 2000 and 2010, were reviewed. Clinical information including the primary site of the tumours, TNM staging, pathological staging, and type of neck dissections, as well as other demographic and clinical data, were retrieved from the electronic medical record system (EMRS) of the hospital. A total of 137 patients fulfilled the inclusion criteria, which were the following: (1) pathologically confirmed primary SCC of the hard palate and maxillary alveolus or gingiva; (2) primary treatment comprising surgery only. Exclusion criteria were the following: (1) SCC originating from the nasal cavity or paranasal sinus; (2) primary tumour invading the soft palate, oropharynx, or retromolar area; (3) adjunctive radiotherapy given after surgery.
All patients were staged according to the Union for International Cancer Control (UICC) TNM classification based on a complete clinical examination of the head and neck as well as computed tomography (CT) or magnetic resonance imaging (MRI) scan. In all cases, the primary tumour sites were treated with radical resection aimed at 1.5-cm margins. The margins were confirmed intraoperatively by frozen section. For patients with clinically positive lymph nodes (cN+), a radical or modified radical neck dissection was performed at the same time. However, there was no standard protocol for the negative neck (cN0). In most of the early cases, a ‘watch and wait’ approach was applied. END was carried out in some of the more recent cases. Patients were followed up every 3 months in the first 2 years, then every 6 months until the fifth year, and then annually after 5 years. Local recurrence and regional failure were determined by clinical as well as radiographic examinations, and histopathology if necessary. Salvage surgery was performed if delayed metastasis or regional failure was confirmed. Radiotherapy was recommended to patients with a pN+ neck after neck dissection.
The data collection and statistical analysis were performed using SPSS version 17.0 software (SPSS Inc., Chicago, IL, USA). The χ 2 test or Fisher’s exact test was used to determine the incidence of metastasis and correlated factors. Multivariate analysis by logistic regression was also performed. A Kaplan–Meier plot was used to determine the overall survival rate, and a log-rank test was performed to evaluate any statistical significance ( P < 0.05).
One hundred and thirty-seven patients were included in this study; 59 were male and 78 were female. The median age of patients at the time of diagnosis was 70.1 years (range 44–99 years). Detailed clinical information was available for only 100 of these 137 patients. The follow-up rate was 73.0%. We were unable to make contact with the remaining 37 patients after the primary surgery and were therefore not able to define the exact N stage or the survival rate of these cases; thus they were excluded from the study. The follow-up period ranged from 2 to 140 months and the mean was 45.8 ± 34.2 months.
Of the 100 patients included, nine were diagnosed as cN+ and this was confirmed by histopathological examination after radical neck dissection. The other 91 patients were considered as cN0 cases based on the clinical or radiographic examinations. END was not performed routinely for these cN0 cases and only 34 underwent a selective neck dissection from level I to III. Positive lymph nodes were detected in seven cases. The other 57 patients underwent routine observation and 18 presented with delayed metastasis. There were palpable lymph nodes in the necks of these 18 patients, and lymph nodes larger than 10 mm with suspected liquefaction were examined with CT/MRI scans. Metastasis was confirmed histopathologically, and salvage neck dissections were performed in all of these cases. Hence the overall rate of metastasis was 34.0% (34/100), while the rate of occult metastasis was 27.5% (25/91) ( Fig. 1 ).
A summary of the details of these 100 cases is presented in Table 1 . There was no significant difference in the rate of metastasis according to the primary site ( P = 0.454), although SCC of the maxillary gingiva showed a higher risk (34.3%) when compared to that of the hard palate (26.7%). Of note, when the tumour involved the gingival–buccal sulcus, the cervical metastasis rate increased to 70.0%, which was significantly different from those without sulcus involvement ( P < 0.01). Histopathologically, positive lymph nodes were mostly detected at level I (70.5%), followed by levels II (56.8%) and III (13.6%); only 4.5% were found at level IV and none at level V. The rate of metastasis of advanced stage tumours (T3/4) was 40.8%, which was higher than that of early stage tumours (T1/2) (13.7%). This result showed the T stage to be significantly correlated with cervical metastasis in maxillary SCC patients ( P = 0.002). Concerning the histological grading, the prevalence of nodal disease differed significantly in relation to the differentiation of the primary tumours. Poorly differentiated tumours tended to have a much higher risk than the moderately or well-differentiated tumours ( P < 0.01).
|Total||N stage||N+%||P -value|
The results of the multivariate analysis are shown in Table 2 . The risk of cervical metastasis of maxillary SCC was highly correlated with the pathological grade and the involvement of the gingival–buccal sulcus. The primary site and clinical stage were not significant in this context.
|Factors||B||Wald||Sig. ( P -value)||OR||95% CI of OR|
|Primary site (gingiva/palate)||0.298||0.248||0.619||1.347||0.417||4.358|
|Sulcus involvement (involved/not involved)||1.667||5.615||0.018||5.298||1.334||21.043|
|T stage (T3–4/T1–2)||0.306||1.295||0.255||1.357||0.802||2.298|
The mean follow-up time for these 100 patients was 45.8 ± 34.2 months, ranging from 2 to 140 months. During follow-up, 15 patients presented with local recurrence and 24 presented with regional recurrence; two of them had both local and regional recurrence. Ninety-one of the 100 patients were diagnosed with a cN0 neck and 34 of them underwent END together with primary tumour resection. The remaining 57 patients were managed by observation only. Four patients in the END group presented with regional recurrence and only one of them survived after the salvage surgery within the follow-up period. In the observation group, 18 patients presented with regional recurrence and seven had a salvage neck dissection and survived. The results indicated that END can significantly reduce the risk of regional recurrence ( P = 0.033). In the nine cN+ patients, a modified radical neck dissection was performed and recurrence occurred in two patients. Both of them died within 6 months ( Table 3 ).
|Present status||Treatment of primary neck disease||Total||Regional recurrence||Treatment of regional failure||Salvage rate, %||P -value for recurrence|
|cN0||END||34||4 (11.8%)||Surgery||25% (1/4)||0.033|
|Observation||57||18 (31.6%)||Surgery||38.9% (7/18)|
|cN+||MRND + RT||9||2 (22.2%)||RT + chemotherapy||0% (0/2)|
The overall 3-year and 5-year survival rates were 66.3% and 56.7%, respectively ( Fig. 2 ). The 3-year and 5-year survival rates were 80.7% and 72.3%, respectively, in the N0 group, while the survival rates were 30.8% and 20.5%, respectively, in the pN+ group ( P < 0.01). With regard to the T stage, advanced T stage tumours had a higher cervical metastatic rate than early T stage tumours. The 5-year survival rates were 93.3%, 73.2%, 32.7%, and 25.4% for T1 to T4 stage tumours, respectively ( P < 0.01). The survival rates according to the T and N stages are shown in Table 4 . Kaplan–Meier survival curves for the various T and N stages are shown in Figs 3 and 4 .
|T/N stage||Mean survival time||3-year overall survival rate||5-year overall survival rate||P -value|
|T1||118.4 ± 6.4||100.0%||93.3%||<0.01|
|T2||95.3 ± 9.5||82.3%||73.2%|
|T3||56.1 ± 14.7||32.7%||32.7%|
|T4||53.8 ± 9.9||46.2%||25.4%|
|N0||107.7 ± 7.3||80.7%||72.3%||<0.01|
|N+||41.6 ± 8.3||30.8%||20.5%|