Sinonasal squamous cell carcinoma without clinical lymph node involvement: Which neck management is best?

Objectives: The purpose of this work was to report outcomes of patients with nonmetastatic sinonasal squamous cell carcinoma (SNSCC) and to discuss the impact of elective neck irradiation (ENI) and selective neck dissection (SND) in clinically negative lymph node (N0) patients. Methods: Data from 104 nonmetastatic SNSCC patients treated with curative intent were retrospectively analysed. Uni- and multivariate analyses were used to assess prognostic factors of overall survival (OS) and locoregional control (LRC). Results: Median follow-up was 4.5 years. Eighty-five percent of tumours were stage III–IV. Treatments included induction chemotherapy (52.9 %), surgery (72 %) and radiotherapy (RT; 87 %). The 5‑year OS, progression-free survival, and LRC rates were 48, 44 and 57 %, respectively. Absence of surgery predicted a decrease of OS (hazard ratio [HR] 2.6; 95 % confidence interval [CI] 1.4–4.7), and LRC (HR 3.5; 95 % CI 1.8–6.8). Regional relapse was observed in 13/104 (13 %) patients and most common sites were level II (n = 12; 70.6 %), level III (n = 5; 29.4 %) and level Ib (n = 4; 23.5 %). Management of the neck in N0 patients (n = 87) included 11 % SND alone, 32 % ENI alone, 20 % SND + ENI and 37 % no neck treatment. In this population, a better LRC was found according to the management of the neck in favour of SND (94 % vs. 47 %; p = 0.002) but not ENI. Conclusion: SND may detect occult cervical positive nodes, allowing selective postoperative RT. ENI (ipsilateral level II, ±Ib and III or bilateral) needs to be proposed in selected patients, especially when SND has not been performed.