Effect of Surgical Modality and Hypofractionated Split-course Radiotherapy on Local Control and Survival from Sinonasal Mucosal Melanoma

Abstract Aims To assess the efficacy of surgery and high-dose split-course radiotherapy in sinonasal head and neck mucosal melanoma (SHNMM). Material and methods Between 1991 and 2006, 23 patients (median age 73 years, male:female ratio 0.4) with non-metastatic SHNMM underwent surgery and postoperative radiotherapy, two had exclusive radiotherapy. Radiotherapy consisted of three series of 18 Gy (3 × 6 Gy every other day for 1 week) with 3 week planned treatment breaks. Chi-squared tests, Kaplan–Meyer method and Log-rank test were used to assess prognostic factors for survival and local control. Results There were 20 nasal cavity tumours; 12 of these involved more than one sinonasal site. One patient (4%) had lymphadenopathies at diagnosis. Six SHNMMs (24%) were amelanotic. The median follow-up was 39 months. Fourteen patients had en bloc surgery, 16 underwent radiation (14 postoperative, two exclusive). Eleven patients had local relapse, three had regional relapse and three had bone or liver metastases. Five year local control was 49 ± 12%. Five year overall and SHNMM-specific survival was 38 ± 12% and 62 ± 12%, respectively. Five patients were alive without disease after 5 years and three after 10 years. En bloc excision (tumour removed in one piece) was prognostic for survival. Conclusions En bloc surgery was a prognostic factor on outcomes for local control and survival in this series. Data from the literature have shown that postoperative radiation therapy improves local control. Most series were carried out with conventional fractionation. The effect of planned breaks (split-course radiotherapy) may be deleterious, as suggested in this series. Therefore, split-course radiotherapy cannot be recommended for SHNMM.