ESTRO 36 Abstract Book

S577 ESTRO 36 2017 _______________________________________________________________________________________________

Purpose or Objective Postoperative concurrent chemoradiotherapy (CCRT) is supprior to RT alone for squamous cell carcinoma of the head and neck (SCCHN) with high-risk factors (such as extracapsular invasion, positive resection margin) by both RTOG 9501 and EORTC 22931 trials. We developed a new protocol of intensive chemotherapy followed by IMRT (IntCT+RT) and compared the toxicity and efficacy with CCRT for patients with SCCHN and poor prognostic factors after surgery. Material and Methods Ninety-two SCCHN patients who received curative resection first and with at least one of the following characteristics, resection margin involvement or close to the tumor, extracapsular invasion, perineural invasion, angiolymphatic invasion, pathological stage T4, and multiple neck nodes metastasis were eligible for this study. Postoperative multi-drugs combination chemotherapy (Methotrexate 30 mg/m2 d1, Epirubicin 30 mg/m2 d1, alternating with Mitomycin-C 4 mg/m2 d8, Oncovin 1 mg/m2 d8, Cisplatin 25 mg/m2 d8, Leucovorin 120 mg/m2 d8, 5-fluoroUracil 1000 mg/m2 d8, and Bleomycin 10 mg/m2 d8) for 10-12 weeks were administered followed by IMRT in the IntCT+RT groups. Patient in the CCRT group received similar RT and concurrent chemotherapy of either tri-weekly cisplatin 100mg/m2 alone or tri-weekly cisplatin 50mg/m2 plus oral tegafur-uracil 2# bid for 7 weeks. Results The IntCT+RT group (n=37) had less regional (8.1% vs. 12.7%) and distant (2.7% vs. 7.3%) failures, compared with the CCRT group (n=55). The Kaplan-Meier survival analyses revealed that patients treated by IntCT+RT had better regional failure-free survival (3-year rate, 94.5% vs. 72.4%, P=0.0294) and overall survival (75.9% vs. 61.8%, P=0.1343) than those who received CCRT. Grade 3/4 acute toxicity of IntCT phase included leucopenia (51.4%), anemia (27.0%), vomiting (5.4%), alopecia (5.4%) and mucositis (2.7%) but patients could tolerate it well. During RT period, patients in the CCRT group had significantly higher grade 3/4 mucositis (76.4% vs. 57.7%, P=0.0356) and vomiting (27.3% vs. 0%, P<0.0001) than those in the IntCT- RT group. Conclusion IntCT+RT in postoperative setting for high-risk SCCHN patients is feasible. We observe a significant better regional control, favorable overall survival and less grade 3 or 4 mucositis and vomiting in the IntCT-RT group compard with the CCRT. This novel approach deserves to be studied in a phase III randomized trial. EP-1056 Radiation and concurrent superselective intra-arterial cisplatin for maxillary sinus cancer T. Ebara 1 , K. Ando 1 , M. Kawahara 1 , M. Suzuki 2 , H. Horikoshi 3 , Y. Tamaki 4 1 Gunma Prefectural Cancer Center, Division of Radiation Oncology, Ota, Japan 2 Gunma Prefectural Cancer Center, Division of Head and Neck Surgery, Ota, Japan 3 Gunma Prefectural Cancer Center, Division of Radiology, Ota, Japan 4 Tsukuba University Hospital, Department of Radiation Oncology, Tsukuba, Japan Purpose or Objective This study aimed to evaluate the efficacy of radiation and concomitant superselective high-dose intra-arterial cisplatin (RADPLAT) for maxillary sinus squamous We conducted a retrospective chart review of MS-SCC patients treated with RADPLAT between 2008 and June 2016. Results carcinoma (MS-SCC). Material and Methods

Thirty-four MS-SCC patients were received RADPLAT. There were 9 patients (26%) diagnosed with T3, 14 (41%) with T4a, and 11 (32%) with T4b disease. Lymph-node involvement was present in 6 patients. Cisplatin with median 150 mg was administered using the superselective intra-arterial infusion method bi-weekly. Of them, 29, 3, 1 and 1 patients were received 4, 3, 2 and 1 times cisplatin infusions, respectively. Radiation, ranged with 50–74 Gy with median 60 Gy was administered by 2 Gy fraction in 5 times a week. The median follow-up was 24.6 months, ranged with 4.1-92.4 months. Complete responses in the primary site were obtained in 11 (32%) patients and partial responses in 19 (56%) patients. The 2 and 5-year overall survival rates (OS) were 66 and 45%, respectively. The 2 and 5-year primary-site recurrence free survival rates (PRFS) were 49 and 40%, respectively. The 2 and 5-year disease-free survival rates (DFS) were 45 and 37%, respectively. The patients with T3 showed significantly better OS (p=0.03). The patients with 4 time infusions showed significantly better OS, PRFS, and DFS (p<0.05). There was no life-threatening toxicity. Conclusion In RADPLAT for MS-SCC, 4 times cisplatin infusions could improve the prognosis. EP-1057 Predictive and prognostic value of pretreatment [18F] FDG-PET parameters in head-and- neck cancer L. Deantonio 1 , M. Paolini 1 , E. Puta 2 , L. Vigna 3 , R. Matheoud 3 , L. Masini 1 , G. Sacchetti 2 , M. Brambilla 3 , M. Krengli 1 1 University Hospital Maggiore della Carità, Radiotherapy, Novara, Italy 2 University Hospital Maggiore della Carità, Nuclear Medicine, Novara, Italy 3 University Hospital Maggiore della Carità, Medical Physics, Novara, Italy Purpose or Objective To evaluate the predictive and prognostic value of [F18] FDG-PET parameters performed prior to radiotherapy in head-and-neck cancer patients. Material and Methods Thirty-eight patients with newly diagnosed head-and-neck cancer treated with concomitant chemoradiotherapy underwent [F18] FDG-PET before the treatment course. The maximum and the mean standardized uptake value (SUVmax, SUVmean), the metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were analysed. Multiple threshold levels were tested in order to define the most suitable threshold value for the metabolic activity of each patient's tumour: the fixed threshold of the 41% and 50% of the maximum uptake value (SUV41%, SUV50%) and an adaptive threshold algorithm (ATA) implemented on the iTaRT workstation (Tecnologie Avanzate, Italy). We evaluated the relationship of mean values of SUVmax, SUVmean, MTV, and TLG with tumour characteristics, treatment response, local recurrence, distant metastasis and disease-related death. Receiver-operating characteristic (ROC) curve analysis was done to obtain optimal predictive cut-off values for PET parameters. Disease-free (DFS) and overall survival (OS) disease- related were examined according to these cut-offs. Results The mean value and range of each parameters were tumour staging (p= 0.04). Thirty-two/38 patients (84.2%) had a complete response, 4/38 (10.5%) a partial response, and 2/38 (5.2%) a no response 8 weeks after the completion of treatment. SUV parameters resulted not predictive of tumour response. After a median follow-up of 22 months, 6/38 (15.8%) patients developed local recurrence and 6/38 (15.8%) calculated (Table1). Higher SUVmean ATA was associated to higher primary