ESTRO 37 Abstract book

S790

ESTRO 37

Conclusion In our experience, SBRT for PCA is effective and feasible, able to achieve a high local control rate with a favorable toxicity profile, even if a larger case series with a longer follow-up period is necessary. EP-1452 Safety & tolerability of radiotherapy dose escalation in cervical-upper thoracic oesophageal cancer C. Jones 1 , P. Murray 1 , R. Goody 1 , G. Ward 1 , N. Casanova 1 , P. Hatfield 1 , G. Radhakrishna 1 1 St James's Hospital, Leeds Cancer Centre, LEEDS, United Kingdom Purpose or Objective For patients with carcinoma of the upper oesophagus, an organ-preserving treatment approach using definitive chemoradiotherapy (dCRT) avoids much of the morbidity associated with surgical resection. dCRT to the upper oesophagus is nevertheless associated with higher rates of locoregional failure than at lower oesophageal sites. Dose escalation of radiotherapy (RT) using intensity modulated RT (IMRT) has been proposed as a means to enhancing locoregional control, yet there remains little consensus regarding dose constraints in treatment of upper oesophageal tumours. We therefore sought to determine the safety and tolerability of RT dose escalation using IMRT as part of dCRT for tumours of the cervical and upper thoracic oesophagus. Material and Methods Retrospective cohort study of 26 (25 SCC) patients with upper oesophageal cancer managed within Leeds Cancer Centre, UK, between November 2011 and February 2014. Patients were eligible for inclusion if they had a confirmed tumour of the cervical oesophagus or a tumour within 2cm of cricopharyngeus, extending above the thoracic inlet. Treatment was delivered at 60-66Gy in 30- 33 fractions administered five days per week for six- seven weeks. Target volumes included primary and nodal disease, bilateral neck levels III, IV and Vc, and upper mediastinal nodal regions for prophylactic nodal irradiation. Concurrent chemotherapy consisted of two cycles of cisplatin or carboplatin with either 5FU or Capecitabine. Thirteen patients received neoadjuvant chemotherapy consisting of the same regimen. Patients were followed up three-monthly following completion of treatment, with dysphagia assessed at baseline and at the first post-treatment review. Results Median age at diagnosis was 67.9 (IQR 60.6-74.1) years with six (23.1%) patients of performance status (PS) 0, 17 (65.4%) PS1 and three (11.5%) PS2. All patients completed planned treatment and there were no treatment-related deaths. Three (12%) experienced grade 3 odynophagia. Twenty-five (96%) patients reported a subjective improvement in their swallow during treatment. There were no instances of grade 4 toxicity or of trache- oesophageal fistula formation. Median OS was 154 weeks with respective 1 and 2 year survival of 79.9% and 52.4%. Median recurrence-free survival was not reached at the time of sensor but at 1 and 2 years was respectively 70% and 50%. Conclusion In this cohort of patients with cancer of the cervical or upper thoracic oesophagus, dose escalation of up to 66Gy using IMRT as part of dCRT was safe and well tolerated by patients. Future studies investigating both the toxicity and the relative efficacy of dose-escalated dCRT are required.

Results Median follow up time was 6.1 months (range 2.9-11.8 months). 19 pts (95%) completed the planned treatment. 1 patient (5%) developed G4 gastro-intestinal toxicity (abdominal pain) that required interruption of SBRT after 3 fractions. No other acute or late grade ≥3 toxicities were observed during follow up. Most pts (17/20, 85%) had a radiographic response or stable disease upon restaging. 11 pts (55%) underwent exploratory laparotomy and 7 (35%, 6 BRPC, 1 LAPC) surgical tumor resection (3 pts achieved R0 resection, 42%). 2 pts (10%) underwent radiofrequency ablation during laparotomy, while 2 (10%) underwent exploration, but not resection, due to intraoperatively found liver metastases. All pts were alive with a median OS of 14.5 months from the time of diagnosis and a median PFS of 12.0 months. Tumor relapse occurred in 7 pts (35%), distant treatment failure without local failure in 4 and, a combination of local progression and distant failures in 3. Overall, a LCR of 85% was obtained.