ESTRO 37 Abstract book

S649

ESTRO 37

and 56.1%, respectively. Prognostic factors for OS, in multivariate analysis, were performance status 0 (p = <0.001), small T1, T2 tumors (p = 0.002), concurrent chemoradiotherapy (p = 0.001), current smoker (p = 0.026), or UFND (p < 0.001). The size (median: 7 cm) of the neck node (≤ 7 cm vs > 7 cm) or lymph node laterality (one side vs two sides of the neck) were not significant factors in multivariate analysis. UFND was strongly associated with better OS (45.7% vs 21.2%) with hazard ratio (HR) of 0.5 (95%CI 0.38-0.70), where surgery of primary tumor had no impact on OS (p = 0.64). UFND was also associated with better loco-regional control (p < 0.001), and PFS (p < 0.001). A significantly different outcome for patients with primary non oropharynx cancers (p = 0.001) was observed with UFND and only a trend of better outcome was observed in patients with oropharynx cancers with UFND (p=0.072). Regardless of the neck node size, UFND improved survival (p=0.001 for ≤ 7 cm, and p=0.004 for > 7 cm). Conclusion Upfront neck dissection could improve the treatment outcome in N3 HNSCC, regardless neck node size, especially for non-oropharyngeal cancer. Confounding biases might influence the results. A comparison of patients with operable N3 nodes treated either with UFND or salvage neck dissection in case of positive post treatment PET-CT would be interesting EP-1159 Management and treatment outcomes of patients with nasal cavity melanoma from a UK cancer centre J. Lester 1 , S. Rawther 1 , N. Foley 1 , B. Foran 1 1 Weston Park Hospital, Oncology, Sheffield, United Kingdom Purpose or Objective Malignant melanoma of the nasal cavity is a rare cancer with an annual incidence in the region of 0.3 patients per million population. The prognosis is poor with a high risk of local and distant recurrence. Due to its rarity there is often uncertainty about its optimum management, localised disease however is often treated with surgical resection followed by adjuvant radiotherapy (ART). Published data on the responses to radiotherapy (RT) and systemic therapy is limited to a few case reports. In this series we report the demographics and treatment outcomes of patients diagnosed between Jan 2008 and Aug 2017 in our centre with a population base of approximately 1.9 million. Material and Methods Patients diagnosed with nasal cavity melanoma were identified from our head and neck cancer MDT meeting and radiotherapy database. Data was retrospectively collected on demographics, tumour characteristics (Stage and Resection margin), radiotherapy and systemic therapy treatments, local and distant relapse and survival outcomes. Results 21 patients were identified with a median age at diagnosis of 70yrs (range 54-95). 11 patients were male and 10 were female. At the time of diagnosis 14 (66.7%) had T3 disease and 7 (33.3%) were T4. 1 (4.8%) patient had regional nodes and 2 (9.5%) patients had distant metastatic disease. 15 (78.9%) underwent surgical resection and 12 of those had ART using 60Gy in 30 fractions. Only 2 (16.7%) developed a local recurrence following ART, however 7 (58.3%) patients subsequently developed distant metastatic disease. The median time to the development of metastasis was 5.1 months (range 3.1-36.6). Of the 6 patients who received Palliative RT, 5 (83.3%) had a good symptomatic and clinical response. 12 (57.1%) patients either presented with metastatic disease or subsequently developed it. Of these, 8 (66.7%) patients received 9 courses of systemic therapy (Pembrolizumab=5, Dacarbazine=3 and Ipilimumab=1).

Conclusion This study showed that photon and proton association is a good option for the treatment of stage T4N0M0 nasopharyngeal carcinoma. After a long follow-up, we observed high locoregional and survival rates without significant toxicities. EP-1158 Prognostic factors and role of neck dissection in N3 head and neck cancers treated with radiotherapy A. Boros 1 , P. Blanchard 1 , P. Gorphe 2 , I. Breuskin 2 , C. Even 3 , F. Nguyen 1 , E. Deutsch 1 , F. Bidault 4 , F. Janot 2 , S. Temam 2 , H. Mirghani 2 , Y. Tao 1 1 Institut de Cancérologie Gustave Roussy, Radiation Oncology, Villejuif, France 2 Institut de Cancérologie Gustave Roussy, Surgery oncology, Villejuif, France 3 Institut de Cancérologie Gustave Roussy, Oncology, Villejuif, France 4 Institut de Cancérologie Gustave Roussy, Radiology, Villejuif, France Purpose or Objective The optimal management of N3 (neck node ≥ 6 cm) head and neck squamous cell carcinoma (HNSCC) is not known. The role of upfront neck dissection (UFND) is still controversial as very few were included in the PET-NECK study (Mehanna et al. NEJM 2016). The purpose of our study is to identify prognostic factor of N3 HNSCC in our institution and to evaluate the role of UFND before radiotherapy Material and Methods We retrospectively reviewed the charts of patients with histologically proven N3 HNSCC treated with curative intent in our institution, including: oral cavity, oropharynx, hypopharynx, and larynx. All patients were treated with definitive radiotherapy. The Kaplan-Meier method has been used to estimate overall survival (OS), disease-free survival (DFS), locoregional control (LRC), and distant control (DC) and Cox regression model for multivariate analysis Results 323 patients whit N3 HNSCC treated between 1995 and 2011 were included for final analysis. The mean age was 58 ± 10.3 years. A total of 43.9% (142/323) had a neck dissection. Among those, UFND was performed in 88% (125/142). Median follow up was 3.9 years (0-14.8 years) for surviving patients. Overall survival (OS) at 5 years was 31.2% and progression free survival was 26%. Loco- regional control and distant control at 5 years were 47.4%