ESTRO 2020 Abstract book

S155 ESTRO 2020

Median follow-up was 13 months. SBRT-PATHY showed improved treatment outcomes in terms of survival, tumor and symptom control, and toxicity compared to standard of care. Table 1 and Figure 1, respectively summarize the main clinical results. Multi-variate analysis for cancer specific survival was significant for treatment effect with SBRT-PATHY (p < 0.001) independent of age, sex, performance status, histology, stage, treated bulky site and tumor diameter.

Conclusion We present a novel concept of SBRT-PATHY to treat unresectable bulky NSCLC patients which showed improved RT therapeutic ratio compared to standard of care in this phase 2 study with 60 patients. A larger prospective trial is ongoing to provide additional hope and promise for bulky, unresectable NSCLC patients, especially in the context of emerging IT combinations.

Poster Highlights: Poster highlights 10 PH: Toxicity modelling

PH-0283 Machine learning methods to predict rectal bleeding after prostate cancer radiotherapy M. Ibrahim 1 , E. Mylona 2 , N. Boussion 1 , O. Acosta 2 , R. De Crevoisier 2 , M. Hatt 1 1 University of Western Brittany, LaTIM- INSERM- UMR 1101, Brest, France ; 2 University of Rennes, LtSI-INSERM- UMR 1099, Rennes, France Purpose or Objective The goal of this work was to predict rectal bleeding (RB) following prostate cancer (PC) radiotherapy (RT) exploiting dose volume histograms (DVH) and clinical variables in a multicentric setting using 4 ML machine learning algorithms and 3 deep learning (ML, DL) techniques as well majority voting. A specific issue associated with multicentric data was the covariate shift issue, i.e., variables from each center could have strongly different distributions, which hampers the ability to efficiently train and validate multiparametric models using ML. An additional challenge was the high imbalance in the The records of 591 patients with more than 3 years follow up (including DVH, clinical data and rectal bleeding events) who underwent RT for localized PC were collected prospectively. The target volume was defined as the prostate and seminal vesicles. The mean dose delivered to data (i.e., few events). Material and Methods