This was a retrospective study of 33 patients treated for sinonasal squamous cell carcinoma between 1995 and 2008. Epidemiological, clinical, histological, and therapeutic aspects of this series of patients were analysed, and their impacts on overall survival and disease-free survival established using the Kaplan–Meier method. A search for prognostic factors was made using a log-rank test. There were 27 men. The average age at diagnosis was 64.7 years. Tobacco-smoking was found to be a risk factor in 24 patients (72.7%). The median follow-up was 66 months (range 0–99 months). Tumours were classified as T1 in 18.3%, T2 in 27.3%, T3 in 6%, and T4 in 48.5% of cases. Disease-free survival rates at 1 and 5 years were 58.5% and 46.1%, respectively, and overall survival rates were 70.3% and 40%, respectively. Overall survival was correlated to tumour status (TNM, American Joint Committee on Cancer) ( P = 0.010) and involvement of key structures (skull base, dura mater, brain, orbit, cavernous sinus, infratemporal fossa, skin) ( P = 0.049). Surgery followed by radiotherapy improved overall survival ( P = 0.005) and disease-free survival ( P = 0.028) when compared to other treatment modalities. When compared to surgery alone, it improved disease-free survival ( P = 0.049) regardless of tumour stage.
Sinonasal squamous cell carcinomas (sinonasal SCCs) are aggressive tumours associated with frequent recurrence and a high mortality. They are rare tumours even though they represent 35–75% of all sinonasal cancers. This is why many authors include tumours with different histologies to present larger series. If this strategy makes for improved significance of the results, it runs the risk of erasing certain characteristics related to a specific tumour type.
Because they arise in air-filled cavities, most patients have advanced stage disease with involvement of adjacent sites (skull base, orbit, brain, etc.) at the time of diagnosis. These lesions are challenging to diagnose at an early stage and difficult for the surgeon to treat.
The aim of this study was to statistically analyse the epidemiological, clinical, histological, and therapeutic aspects of a series of patients with sinonasal SCC and the impact of these on overall survival and disease-free survival.
Materials and methods
This was a retrospective study of 33 consecutive patients treated for sinonasal SCC between July 1995 and July 2008. Epidemiological (age, sex), clinical (tumour side, tumour staging), radiological (tumour staging, involvement of key structures), histological, treatment modality, and patient outcome data were studied. To ensure the homogeneity of the series, the 2009 American Joint Committee on Cancer TNM classification was applied retrospectively to all patients. A database was created using FileMaker Pro (version 5). Only patients treated at our institution were included in this study.
Overall survival and disease-free survival were established using the Kaplan–Meier method. A search for prognostic factors was performed using a log-rank test. Multivariate analysis of all significant factors was performed by Cox proportional hazards model. All statistical tests were conducted using SPSS version 17.0 software (SPSS Inc., Chicago, IL, USA).
Six women (18.2%) and 27 men (81.8%) were included in the study. Their average age at diagnosis was 64.7 years (range 39–84 years). The tumour was on the right side in 12 cases (36.4%), on the left in 14 cases (42.4%), and on the midline in seven cases (21.2%). Tobacco-smoking was found to be a risk factor in 24 patients (72.7%); average intake was 37 packet years (range 4–180 packet years). The median follow-up was 66 months (range 0–99 months).
Clinical data were available for all 33 patients (100%). Initial symptoms were unilateral in 84.8% of cases, the most frequent being nasal obstruction (67.9%), epistaxis (60.7%), and visible nasal polyp (53.6%). Other symptoms included pain (39.3%), rhinorrhoea (35.7%), eye signs (17.9%), skin damage (17.9%), neurological signs (10.7%), and homolateral cervical lymphadenopathy (3.6%).
Location, extension, and TNM staging
Radiological investigations in all cases included a sinus computed tomography (CT) image for local staging and a neck and thorax CT for regional and remote staging. Sinus magnetic resonance image (MRI) completed this assessment in 28 patients (84.8%). Image analysis was used to study primary tumour sites and their modes of propagation and to define TNM staging. Tumour locations were: maxillary sinus in 16 cases (48.5%), nasal cavity in nine cases (27.3%), nasal septum in six cases (18.2%), ethmoid in one case (3.0%), and sphenoid in one case (3.0%).
There was a breach of key structures such as the orbit in eight cases (24.2%), infratemporal fossa in six cases (18.2%), skull base in three cases (9.1%), skin in three cases (9.1%), hard palate in three cases (9.1%), the cavernous sinus in one case (3.0%), and the sphenoid bone in two cases (6.1%). The involvement of key structures was more frequent among larger tumours (T3–T4 89%, T1–T2 7%; P < 0.001).
The tumour was classified as T1 in six cases (18.2%), T2 in nine cases (27.3%), T3 in two cases (6.1%), and T4 in 16 cases (48.5%). Of the 16 tumours classified as T4, 56% were T4a and 44% were T4b. Two patients were N+ at the time of initial diagnosis. There was no metastatic patient at the time of initial diagnosis. One patient classified as T1N0M0 had synchronous primitive lung cancer (adenocarcinoma).
Of the 33 cases of histologically proven sinonasal SCC, the exact subtype was noted in 22 cases (66.7%). It was a keratinizing sinonasal SCC in 16 cases (72.2%): well-differentiated in six cases (37.5%), moderately differentiated in four cases (25%), and poorly differentiated in six cases (37.5%). Two particular subtypes were found: a verrucous squamous cell carcinoma in four cases (18.2%) and a basaloid squamous cell carcinoma in two cases (9.1%).
Six types of initial treatment were used: surgery alone in seven cases (21.2%), surgery followed by radiotherapy in eight cases (24.2%), surgery followed by concurrent chemoradiotherapy in three case (9.1%), concurrent chemoradiotherapy in 13 cases (39.4%), chemotherapy in one case (3.0%), and radiotherapy alone in one case (3.0%; elderly patient classified as T4aN0M0).
Eighteen patients (54.5%) underwent surgery. Surgery was performed by a cranio-facial approach in one case, a trans-facial approach in nine cases, a degloving approach in six cases, and an endoscopic approach in two cases. The quality of excision margins was identified in 18 cases (100%). All patients showed clear margins on final histological analysis. No patient underwent neck dissection in this series. The two patients classified with lymph node metastasis at the time of the diagnosis were treated with concomitant chemoradiotherapy (one T4aN2cM0 and one T4aN1M0).
Chemotherapy was performed using platinum salts with fluorouracil when the patient had no contraindication. For radiotherapy, using an unfractionated protocol, doses ranging from 45 to 70 Gy were delivered to the tumour site and cervical lymph node area.
Recurrence and survival
Disease-free survival rates at 1, 5, and 10 years were 58.5%, 46.1%, and 46.1%, respectively ( Fig. 1 ). The overall survival rates at 1, 5, and 10 years were 70.3%, 40%, and 40%, respectively ( Fig. 2 ).