ESTRO 2021 Abstract Book

S807

ESTRO 2021

This retrospective monocentric study included patients with synchronous mHNSCC and undergoing chemotherapy. Locoregional treatment was defined as either exclusive radiotherapy or surgery with or without adjuvant radiotherapy. Multivariate analyses were carried out using two different Cox models according to the treatment: one was performed with a 2-category variable (systemic therapy with or without locoregional treatment) and the other with a 3-category variable (chemotherapy alone or associated with locoregional radiotherapy delivering a dose inferior to 70 Gy or equal to 70 Gy). To account for a selection bias favoring the administration of radiotherapy in patients with better prognosis, we performed a subgroup analysis in patients with complete, partial or stable response for metastatic disease after first-line chemotherapy. Results 148 patients were included from 1998 to 2018. 80 patients were treated with chemotherapy alone and 68 patients with chemotherapy combined with locoregional treatment. Locoregional treatment consisted in exclusive radiotherapy in 62 patients, associated with concurrent chemotherapy in 31 patients, and surgery in 6 patients, followed by adjuvant radiotherapy in 4 of them. Median overall survival (OS) was 13 months [10.7- 15] and median progression free survival (PFS) was 7.7 month [6.5-8.9]. Median OS was significantly improved in patients treated with locoregional treatment in comparison to patients treated with chemotherapy alone (16.8 vs. 7.5 months, p<0.001). Median PFS was also significantly improved (10 vs. 5.2 months, p<0.001). On multivariate analysis, the presence of locoregional treatment in the first model, or a radiation dose equal to 70 Gy in the second model, were significantly associated with longer OS. Other significant good prognostic factors for OS were good performance status and early N stage. Among the 75 patients displaying stability, partial or complete response after initial systemic therapy (51% of the whole cohort), locoregional treatment was associated with prolonged median OS (21.9 vs. 14.7 months, p=0.047) and a trend toward improved locoregional median PFS (13.8 vs. 8.1 months, p=0.07). Again, on multivariate analysis, the presence of locoregional treatment in the first model, or a radiation dose equal to 70 Gy in the second model, and early N stage, remained associated to prolonged OS. Conclusion The addition of a locoregional treatment to systemic therapy is associated with prolonged OS in patients with synchronous mHNSCC. However, we must take account of selection bias, potentially leading patients with better prognosis to be more likely to receive locoregional treatment. PO-0974 Comparing delineation techniques in radiotherapy of node positive head and neck squamous cell cancer C. Brooks 1 , M. Powell 1 , A. Sibtain 1 1 St Bartholomew's Hospital London, Clinical Oncology, London, United Kingdom Purpose or Objective At St Bartholomew’s Hospital, two oncologists manage the head and neck cancer practice. Each uses a different delineation technique in irradiating the node positive neck. One uses an anatomical technique, delivering high dose to any nodal level including a positive node and an elective dose to remaining nodal levels with concurrent 3-weekly cisplatin 100mg/m 2 . The other uses a geometric technique delivering high dose to confirmed nodal disease with a margin and an elective dose to remaining nodal areas with concurrent weekly cisplatin 40mg/m 2 (see figure 1). This study aimed to evaluate if there is a difference in outcome between the two techniques with regard to patterns of recurrence and toxicity.

Materials and Methods Retrospective analysis of the clinical records of 71 patients treated with definitive radiotherapy for node positive head and neck SCC with or without concurrent systemic treatment between December 2014 and December 2017 was made. Comparisons were made between the two groups with regard to volume of tissue treated, chemotherapy delivered, toxicity and recurrence patterns. Results The sample showed a median age of 61 and a 4:1 male predominance. The most common primary tumour site was oropharynx (62%), with equal numbers of laryngeal (11%), hypopharyngeal (11%) and nasopharyngeal cases