ESTRO 36 Abstract Book

S89 ESTRO 36 2017 _______________________________________________________________________________________________

salvage HDRB was based on the Phoenix definition. All non-failure patients were followed for a minimum of 5 years. Eighteen different clinical risk features were collected from each patient. Machine Learning was used to identify subpopulations that would most likely to remain biochemically disease free after the treatment. Decision tree algorithms were constructed using Matlab R 2011a. The complexity of the decision tree was fine-tuned by selecting the optimum number of observations per terminal node that minimized the “Leave One Out Cross-Validation” estimation of the deviance. Results were compared to those obtained using Ensemble Methods. Random permutation experiments were also performed to estimate the probability that the tree found was the result of random variations. Results A subpopulation of patients with a high risk of biochemical failure after salvage HDRB was identified. Those patients with a fraction of positive nodes from those sample that was greater than 0.354 and disease free interval less than 4.12 years had a failure rate after salvage HDRB of 0.75 vs 0.38 for the remainder of the population, Figure 1 . The probability that the conclusions reached in this paper are not due to random fluctuations is 0.7, Figure 2 .

Proffered Papers: Breast

OC-0177 Brachytherapy for the Palliation of Dysphagia Owing to Esophageal Cancer: A Systematic Review. L. Fuccio 1 , D. Mandolesi 1 , A. Farioli 1 , C. Hassan 2 , L. Frazzoni 1 , A. Guido 3 , F. Violante 1 , C. Pierantoni 1 , A. Galuppi 3 , F. Bazzoli 1 , A. Repici 4 , A. Morganti 3 1 S.Orsola-Malpighi University Hospital, Department of Medical and Surgical Sciences, Bologna, Italy 2 Nuovo Regina Margherita Hospital, Gastroenterology Unit, Rome, Italy 3 S.Orsola-Malpighi University Hospital, Radiation Oncology Unit, Bologna, Italy 4 Humanitas Research University Hospital, Endoscopy Unit, Rozzano MI, Italy Purpose or Objective The management of dysphagia owing to eso phageal cancer is challenging. Brachytherapy has been proposed as an alternative option to stent placement. We performed a systematic review to examine its efficacy and safety in the Prospective studies recruiting at least 20 patients with malignant dysphagia and published up to April 2016 were eligible. The dysphagia-free survival (DFS) and adverse event rates were pooled by means of a random effect Six studies for a total of 9 treatment arms (623 patients) were eligible for inclusion. After 1 month since treatment, the DFS rate was 86.9% [95%CI: 76.0%–93.3%]; after 3 months, it was 67.2% [95%CI: 56.1%–76.7%]; after 6 months, it was 47.4% [95%CI: 38.5%–56.5%]; after 9 months, it was 37.6% [95%CI:30.0%–45.9%]; and, finally, after 12 months, it was 29.4% [95%CI: 21.6%–38.7%]. The heterogeneity between studies was high at 1-, 3- and 6- month assessment; the values of I 2 were 86.3%, 80.0% and 57.8%, respectively. The meta-regression analysis showed total radiation dose and number of fractions as the only positively influencing factors. Severe adverse event rate was 22.6% (95%CI 19.4–26.3). The main reported adverse events were brachytherapy-related stenosis (12.2%) and fistula development (8.3%). Two cases (0.3%) of deaths were reported due to esophageal perforation. Conclusion Brachytherapy is a highly effective and relatively safe treatment option therefore its underuse is no longer justified. Further studies should investigate the optimal radiation dose and number of fractions able to achieve the highest DFS rates. OC-0178 Demonstration of Catheter Insertion Using Electromagnetic Guidance in Breast Brachytherapy H. Brastianos 1 , T. Vaughan 2 , A. Lasso 2 , M. Westerland 1 , J. Gooding 1 , T. Ungi 2 , G. Fichtinger 2 , C. Falkson 1 1 Queen's University Cancer Research Institute, Radiation Oncology, Kingston- Ontario, Canada 2 Queen's University, School of Computing, Kingston- Ontario, Canada Purpose or Objective Accelerated partial breast irradiation using multi-catheter interstitial brachytherapy may be used for early stage breast cancers. To ensure ideal dosimetry over the tumor bed, the catheters need to be placed in parallel with equal spacing. The breast is a deformable organ; thus, placing catheters in the correct position is challenging. To ensure adequate spacing and position, we will apply real-time electromagnetic guidance (EM) in combination with ultrasound (US) to optimize the catheter insertions. This study will discuss the use of electromagnetic tracking catheter with ultrasound to insert catheters in phantoms. resolution of dysphagia. Material and Methods model. Results

Figure1. The Optimal Decision Tree obtained for predicting failure after Salvage HDRB.

Figure 2. Random permutation of output labels. 1000 iterations were created in each case were the outcome of was randomly permuted. Not correlation between features and outcome should be present in this case. The probability of obtaining a cross-validated leave one out error smaller than the one obtained by the tree in Figure 1 was calculated. Conclusion Patients with a fraction of positive nodes higher than 0.35 and a disease free interval bigger than 4.12 years are at higher risk of biochemical failure after salvage HDBR. Machine Learning is effective in identify subtle variables that can affect the treatment outcome.