Should an elective neck dissection be performed for maxillary sinus squamous cell carcinoma?

Abstract

Paranasal sinus malignancies constitute 3% to 5% of head and neck cancers, with squamous cell carcinoma (SCC) representing 70% to 80% of all paranasal sinus malignancies and with maxillary sinus SCC (MS-SCC) the most common location.1 MS-SCC commonly presents as high T-stage tumors, as patients generally present only after the tumor has caused outflow obstruction or invasion into adjacent tissues.2 The majority of relapses in MS-SCC are secondary to local recurrence, though regional metastases are not uncommon. In the setting of clinically positive lymph nodes in the neck with any head and neck malignancy, treatment of the neck is recommended. However, there is a relative paucity of evidence in the literature indicating when an elective neck dissection (END) should be performed in MS-SCC. Current National Comprehensive Cancer Network (NCCN) guidelines do not recommend neck dissections in an N0 neck, even in high T-stage MS-SCC. Here, we summarize the available evidence to best answer the question when should an END be performed in MS-SCC. Dooley and Shah reviewed the role of elective neck treatment in MS-SCC patients with an N0 neck at presentation and found the overall recurrence rate was 44% to 68%; however, the rate of isolated neck failure was low (4–17%).2 Rather, most cases of neck recurrence were associated with local recurrence or distant metastases, both of which carry poor prognosis and unlikely amenable to salvage therapy. Additionally, the rate of neck involvement was higher in more advanced primary tumors (T3 or T4). The authors argue that elective treatment of the neck is not necessary in N0 tumors of the maxillary sinus, but may be justifiable in T3 or T4 tumors. The authors further suggest that elective neck dissection in T3 or T4 tumors should only be performed if the neck is to be entered regardless, such as if microvascular anastomosis is to be attempted; otherwise, elective neck irradiation should be used. Park et al.3 conducted a retrospective analysis of 67 patients with N0 maxillary SCC; though in this cohort (Table 1); the primary site was classified as either the maxillary sinus (N = 35) or the maxillary gingiva (N = 32). Within the entire cohort, 9/67 (13.4%) underwent END at the time of primary surgical resection and 58/67 (86.6%) were treated without END. The risk of occult cervical metastasis in this study was 14.9%, and the risk was higher in maxillary gingiva primaries (17.1%) versus maxillary sinus primaries (12.5%). There were seven patients who had isolated neck recurrences, all of which were in the group who did not undergo END. Five of these seven patients (71.4%) had successful treatment for isolated neck recurrence. Though the data is heterogenous between maxillary gingiva and maxillary sinus primary tumors, the authors argue that END is not necessary in maxillary SCC, particularly if patients can be observed for early neck failure. Maxillary sinus: n = 35 Maxillary gingiva: n = 32 Maxillary: n = 31 isolated LN recurrences (all histologies) Cantu et al. described 704 patients with malignant tumors of the paranasal sinuses, including 156 MS-SCC tumors.4 In 16/156 (10.3%) of MS-SCC patients (Table 1), N+ disease was noted at presentation, and 11/16 were T2 tumors. There were 51 cases of lymph node recurrence in the MS-SCC group, with recurrences in 18.0% of T2 tumors. Of the 51 described lymph node recurrences, 31 were isolated to the lymph nodes, and 30 of the 31 patients with isolated lymph node recurrence underwent neck dissection. One of those 30 was found to have a second recurrence in the dissected field. The authors argued that T2 tumors were more commonly associated with lymph node recurrence because in MS-SCC, T2 tumors invade into the oral cavity (hard palate), whereas T3 and T4 tumors do not. The hypothesis is that T2 MS-SCC may behave closer to oral cavity SCC with associated higher rate of lymph node involvement. Therefore, prophylactic lymph node dissection can be considered for T2 disease, but may not be necessary for T3 or T4 MS-SCC. Sangal et al. found conflicting results in a retrospective analysis of 927 patients with MS-SCC with clinical N0M0 disease (Table 1), 146 of whom underwent END.1 There was a statistically significant survival benefit in patients who received END with maxillectomy and adjuvant radiation therapy; however, in patients who did not receive adjuvant radiation therapy, END did not confer a survival benefit. There was a statistically significant increase in overall survival and disease-specific survival in those patients who underwent END with moderately differentiated tumors, poorly differentiated tumors, and tumor size >3 cm. Further stratification to control for covariables indicated that END independently provided a statistically significant reduction in the 5-year hazard of death in T3 tumors, but not in T1, T2, or T4 tumors. These results suggest that END does not provide benefit in early stage tumors as these tumors are unlikely to metastasize. Similarly, elective neck dissection does not appear to confer benefit in advanced tumors (T4), as the course of the disease may not be meaningfully impacted by neck dissection. The ultimate conclusion is that in MS-SCC, END should be performed in patients with T3 tumors and in tumors >4 cm in size. In a national database study, Ahn et al. identified 550 patients with MS-SCC 20.7% of whom had N+ disease at presentation, and 733 patients with nasal cavity SCC (NC-SCC), 9.3% of which had N+ disease at presentation.5 In this cohort, T stage corresponded to higher rates of nodal involvement. Within the MS-SCC cohort, 18.6% of T2 tumors and 22.3% of T3-T4 had nodal involvement at baseline, though size of primary tumor did not contribute to rate of nodal involvement. This study is limited by its descriptive nature and lack of long-term follow-up, but possible treatment modalities are proposed. For T2 MS-SCC that would not require adjuvant radiation, END is considered. Similarly, in T3 and T4 MS-SCC tumors that would require adjuvant radiation, the authors propose including treatment of the neck with radiation. In N0 MS-SCC, strong recommendations cannot be made for upfront elective neck dissection, as evidence and conclusions are heterogenous and there is a lack of prospective trials. However, evidence also suggests that the rate of occult nodal metastases remains low, and it is more appropriate to consider elective neck dissection or treatment of the neck for higher T stage tumors. Additionally, if the neck is to be entered regardless in the event of planned microvascular anastomosis, then elective neck dissection should be strongly considered, particularly in T3 or T4 tumors. Recommendations were made using four retrospective cohort studies1, 3-5 and one review2 (Level 3 evidence).