Abstract Book

S55

ESTRO 37

toxicity, we performed a preselection based on univariable analyses (p<0.2) and multi-collinearity (R- square <0.8) and included these variables in a multivariable Cox regression analysis. Results Of the 137 patients, 3 patients were excluded because of early withdrawal from neoCRT due to intercurrent causes. At time of analysis, 86 (64%) patients were alive and the median follow up was 12 months (range, 2 to 35 months). Cardiac toxicity was detected in 32 (24%) patients. The most frequently detected toxicity was atrial fibrillation (23 patients), for which medical intervention was indicated in 18/23 (78%) patients. Several clinical characteristics and DVH parameters were significantly associated with OS and cardiac toxicity at univariable analysis and were assessed for multi- collinearity. At multivariable Cox regression analysis (forward selection), relative weight loss, cardiac event history, PTV volume and RT technique (IMRT vs VMAT) were significant prognostic factors for OS. For cardiac toxicity, age, cardiac event history and the volume of the RV receiving 5 Gy (RV_V5) remained significant (p<0.05) in multivariable analysis. Results of multivariable analysis are summarized in table 1.

whether we should expect an substantial increase of cardiopulmonary toxicity in patients who received neo- adjuvant chemoradiotherapy followed by surgical resection (neoCRT). The aim of this study was to assess cardiopulmonary toxicity in esophageal cancer (EC) patients treated with neoCRT as compared to matched EC patients treated with surgery alone. Material and Methods Between 2006 and 2012, 96 EC patients were t reated with neoCRT, which consisted of 41.4 Gy, in 23 fractions, combined with weekly carboplatin and paclitaxel. These patients were compared with 96 EC patients treated with surgery alone between 2000 and 2012 by using propensity score matching 1 . Patients were matched for age, sex, tumor characteristics, comorbidity, ASA score and side of thoracotomy. All patients underwent an open transthoracic esophagectomy with 2- field lymphadenectomy. We retrospectively evaluated the cardiopulmonary toxicity and treatment outcome in these patients. Follow up continued until October 2016. Differences in cardiopulmonary toxicity between the 2 groups were compared using an independent t-test. Prognostics factors for overall survival (OS), calculated from start of treatment, pneumonia and atrial fibrillation, calculated from resection date, were identified using a multivariable Cox regression analysis. Results At time of analysis, 69 (36%) patients were alive and the median follow up time was 67 months (95%CI: 56 to 78 months). Heart failure (HF) and atrial fibrillation (AF) were the most frequently observed cardiac toxicities, and were more frequently observed after neoCRT. HF occurred in 33% of the cases after neoCRT and in 26% after surgery alone (p=0.27). Significantly more AF was observed after neoCRT than after surgery alone (38% vs 25%; p=0.06). Finally, the rate of pneumonia was higher after neoCRT (56% vs 35%; p=<0.01). For atrial fibrillation, age was the only significant prognostic factor at multivariate analysis (p=<0.01). For pneumonia, side of surgical resection seemed related (p=<0.01) . For both toxicities, type of treatment (neoCRT vs surgery alone) was borderline significant (p=0.06) at multivariate cox regression analysis. We could not reveal a significant impact of HF, AF and pneumonia on OS in this relatively small population. Furthermore, OS was not significantly different between neoCRT and surgery alone (p=0.33). Conclusion Atrial fibrillation and pneumonia were more frequently seen in EC patients treated with neoCRT followed by surgery compared to surgery alone. However, both toxicities did not seem to influence OS. Reference: 1 Bosch DJ et al. Ann Surg Oncol. 2014 Feb; 21(2):605-11. PV-0103 Can we safely reduce the radiation dose to the heart in esophageal cancer patients? J.C. Beukema 1 , Y. Kawaguchi 2 , J.A. Langendijk 1 , P.V. Luijk 1 , N.M. Sijtsema 1 , A.V.D. Schaaf 1 , T. Teshima 2 , C.T. Muijs 1 1 University Medical Centre Groningen, department of radiation oncology, Groningen, The Netherlands 2 Osaka International Cancer Institute, department of radiation oncology, Osaka, Japan Purpose or Objective Traditionally, radiotherapy planning for intrathoracic tumours focused on reducing the dose to the lungs in order to prevent pulmonary toxicity. Recently, several papers suggested worse overall survival (OS) with higher cardiac dose, raising awareness of cardiac toxicity. However, reducing dose to the heart may increase lung dose. The main objective of this abstract was to

Conclusion This prospective cohort study demonstrated PTV volume and RT technique (in favour of IMRT) to be prognostic factors for OS. Moreover, cardiac radiation dose was a significant factor for the development of cardiac toxicity after neo-CRT and esophagectomy for EC. These results suggest that the RT dose distribution plays an important role in the outcome of EC patients. Our PDRP allows for an accurate assessment of cardiac toxicity and evaluation of changes in RT techniques (heart sparing VMAT, breath- hold and protons) on these endpoints. PV-0102 Cardiopulmonary toxicity after tri-modality treatment versus surgery alone for esophageal cancer M.J. Van der Heiden 1 , E. Oldehinkel 1 , J.C. Beukema 1 , D.J. Bosch 2 , J.T. Plukker 3 , G.A.P. Hospers 4 , J.A. Langendijk 1 , C.T. Muijs 1 1 University Medical Center Groningen- The Netherlands, Department of Radiation oncology, Groningen, The Netherlands 2 University Medical Center Groningen- The Netherlands, Department of Anesthesiology, Groningen, The Netherlands 3 University Medical Center Groningen- The Netherlands, Department of Surgical Oncology, Groningen, The Netherlands 4 University Medical Center Groningen- The Netherlands, Department of Medical Oncology, Groningen, The Netherlands Purpose or Objective The benefit of tri-modality treatment is clearly established for esophageal cancer, but the impact on toxicity is not well defined and will become increasingly important with improved survival. Moreover, there is growing evidence for cardiopulmonary toxicity after thoracic radiotherapy. Therefore, the question arises