ESTRO 38 Abstract book

S664 ESTRO 38

9.1% (2/22) in level IV, and 9% (1/11) in level V. The level V LN was clinically evident preoperatively. 24.2% elective neck dissection contained occult LNs, all of which were in level II and without extranodal extension. 18 (64.28%) patients underwent adjuvant radiation; 7 (25%) patients underwent adjuvant chemoradiation. With a mean follow- up of 25 months on multivariate analysis, >5 involved lymph nodes was significantly associated with worse progression free survival (PFS) (P = 0.001). Rates of 3-year locoregional recurrence (LRR) in patients with ≤5 vs > 5 were 8% and 23% (P = 0.1). Rates of 3-year distant metastases (DM) were 14% and 52% between ≤5 and >5 For p16+OPSCC, therapeutic NDs should include any levels bearing suspicious LNs and levels II, III, and IV, while elective NDs should be encompass at least levels II and III, followed by indicated adjuvant treatment, are associated with a low nodal recurrence rate . Patients with more than 5 involved lymph nodes appear to have worsened rates of disease recurrence. While these patients appear to be at high risk of both LRR and DM, the predominant mechanism of failure is distant (over 50%). More ambitious studies are necessary that allow to enlarge our knowledge in this entity. EP-1199 Metastatic lymph node features with and without extracapsular extension in head and neck Squamous Cell Carcinoma J. Zheng 1 , A. Flaman 2 , D. Yegendorf 3 , B. Purgina 2 , S. Chakraborty 4 , M. Gaudet 1 , H. Alain 1 1 The Ottawa Hospital Regional Cancer Centre, Radiation Oncology, Ottawa, Canada ; 2 The Ottawa General Hospital, Anatomical Pathology, Ottawa, Canada ; 3 University of Ottawa, Science, Ottawa, Canada ; 4 The Ottawa General Hospital, Radiology, Ottawa, Canada Purpose or Objective In Head and Neck Squamous Cell Carcinoma (HNSCC) patients with metastatic involved cervical lymph nodes, extra-capsular extension (ECE) carries with it increased risks of local recurrence and distant metastases. ECE is an indication for chemo-radiation after neck dissection which adds significant toxicity. This can potentially be avoided if patients could be risk stratified for ECE prior to surgery. We aim to analyze clinical, pathological and pre-operative radiological features of these patients to better risk stratify for presence of ECE. Material and Methods HNSCC patients aged 18 years and older diagnosed between October 2010 and May 2017 at a single institution. Patients had pathological positive metastatic lymph nodes and undergone neck dissection without neo-adjuvant treatment. The patients’ clinical information, pathological reports and pre- operative CT scans were obtained from the Ottawa Hospital electronic medical record and clinical imaging software. In cases of incomplete pathology data, slides were re-examined by a pathologist. Suspicious lymph nodes from each CT scan were matched to metastatic lymph nodes from dissection specimens based on corresponding level, size and morphological features. Statistical analyses were done through Chi-squared testing. Results Clinical and pathological data of 142 patients were analyzed in this study. ECE was present in 45 patients (32%) and absent in 90 (63%) while unknown for 7 (5%) patients. An association was found between the number of positive lymph nodes and the presence of ECE. For patients with 1-2 metastatic lymph nodes and 3 or more metastatic lymph nodes, the rate of ECE was 18% and 60% respectively (p<0.001). p16 status was only reported for 63 patients but was also associated with ECE. It was present in 19% of p16 positive and 46% of p16 negative (P < 0.001). Conclusion

recurrence in HPV positive oropharyngeal cancer to identify optimal post-treatment diagnostic tools. Material and Methods After institutional review board approval, records of patients with non-metastatic HPV-associated oropharyngeal cancer were retrospectively reviewed. We identified 406 patients with biopsy- proven, HPV positive oropharyngeal cancer treated with definitive radiotherapy. Patients were followed at intervals of 3 months and all patients underwent imaging 2-3 months post treatment in the form of PET/CT or neck CT. Locoregional control (LRC) and freedom from distant metastasis (FFDM) were estimated according to Kaplan- Meier method and comparisons were made by log rank test. Results The median follow up for all patients was 40 months. The 3 year LRC for the entire cohort was 91.3%. Patients with AJCC 8 th edition stage I, stage II, and stage III had a LRC of 94.9%, 85.7%, and 82.3% respectively. There were a total of 30 locoregional recurrences, 28 (93.3%) of which presented with either symptoms or persistent disease on 2-3 month post treatment imaging. Of these 28 patients, 13 patients were found to have recurrent disease on the initial 2-3 month post treatment imaging and 15 patients presented with symptoms that led to the identification of recurrent disease. 3-year FFDM for the entire cohort was 88.4% and when stratified by stage patients with stage I, II, and III had a FFDM of 91.6%, 87.6%, and 74.1%, respectively. There were a total of 31 distant recurrences, 25 (80.6%) of which were identified due to symptoms or on 2-3 month post treatment imaging. Of these 25 patients, 15 patients were found to have recurrent disease on the initial 2-3 month post treatment imaging and 10 patients presented with symptoms that led to the identification of recurrent disease. Conclusion Our results showed that identification of the majority of locoregional and distant recurrences was either symptom driven or identified in the initial post treatment imaging performed 2-3 months after completion of radiation. Based on these results it is reasonable to follow patients clinically with a history and physical examination with direct visualization and to forgo further imaging after the initial post treatment imaging shows a complete response. EP-1198 Knowing the oropharyngeal cancer associated with the human papillomavirus L. Gutierrez Bayard 1 , M.D.C. Salas Buzón 1 , S. Garduño Sánchez 1 , M.J. Macias Lozano 1 , R. Rodríguez Sánchez 1 , S. Sayago Gil 1 , V. Días Díaz 1 , E. Gonzalez Calvo 1 , I. Villanego Beltrán 1 , A. Ruiz Herrero 1 , M. Lorente Sánchez 1 , J. Jaén Olasolo 1 1 Hospital Universitario Puerta del Mar, Radiation Oncology, Cadiz, Spain Purpose or Objective For resected HPV-positive oropharynx squamous cell carcinoma (HPV+ OPSCC) high node number is a critical determinant of survival, based on the 8th edition American Joint Committee on Cancer staging system. In our study, we analyzed the relationship between the number of affected lymph nodes (LNs), neck level and the pattern of failure (local, regional, and distant recurrences). Material and Methods We retrospectively studied 28 HPV+ OPSCC patients treated with neck dissection (ND) and either resection or radiotherapy of the primary tumor between 2014 and 2017. External beam radiation (+/- chemotherapy) was given based on the pathologic findings. Results Metastatic pathologic LNs were the following : 0% (0/5) in level I, 78.5% (22/28) in level II, 25% (7/28) in level III,