Abstract Book

S297

ESTRO 37

6. Rehammar et al. Radiother Oncol 2017;123:299 7. Whetal et al. Radiother Oncol 2014;110:448 8. Greenland et al. Circulation 2007; 115:402 9. Oudkerk et al. Eur Radiol 18:2785–2807 10. Roos et al. Radiother Oncol 2017 [Epub ahead of print] 11. Mast et al. Radiother Oncol 108:248 12. Mast et al. Strahlenther Onkol. 2016;192:696 SP-0563 Prone whole-breast irradiation: are the benefits overshadowed by the challenges? L. Veldeman 1 1 University Hospital Ghent, Department of Radiation Oncology, Gent, Belgium Abstract text From an anatomical point of view, the prone position has clear benefits over the supine position for whole-breast irradiation. First, the breast falls away from the heart and lung by gravity, reducing the dose to these organs. Second, the typical breast shape in prone position improves dose homogeneity compared to supine position, especially in large breasted patients. Despite these advantages, only few radiotherapy departments worldwide use the prone position in clinical practice. The reasons for this are variable. One of the most important drawbacks of the prone position is the complex setup procedure which is more time-consuming and less reproducible than for the supine position. Another drawback is the more challenging target volume delineation because the anatomy is clearly different. Long-term randomized data on the efficacy of prone whole-breast irradiation are not available and sceptics fear that local control or overall survival data might be compromised. In this session, the benefits and challenges of prone breast irradiation are discussed including setup reproducibility, patient comfort, heart sparing and lymph node irradiation. PV-0564 Predicting Genitourinary Toxicity by Machine Learning on Genome-Wide Single Nucleotide Polymorphisms S. Lee 1 , J. Oh 1 , S. Kerns 2 , B. Rosenstein 3 , H. Ostrer 4 , J. Deasy 1 1 Memorial Sloan Kettering Cancer Center, Department of Medical Physics, New York, USA 2 University of Rochester Medical Center, Department of Radiation Oncology, Rochester, USA 3 Icahn School of Medicine at Mount Sinai, Department of Radiation Oncology and Genetics and Genomic Sciences, New York, USA 4 Albert Einstein College of Medicine, Departments of Pathology and Pediatrics, New York, USA Purpose or Objective Genitourinary (GU) toxicity after radiotherapy (RT) compromises the quality of life of prostate cancer survivors. Predicting RT-induced GU toxicity using clinical or dosimetric information remains challenging. We hypothesized that single nucleotide polymorphisms (SNPs) modify patients’ congenital risk of GU toxicity and good prediction can be achieved by combining contributions of genome-wide SNPs to the risk using machine learning methodology. Material and Methods We studied 324 prostate cancer patients who received brachytherapy with or without external beam radiotherapy, were genotyped for 606895 SNPs, and followed up for minimum 3 years according to patient- Poster Viewing : Poster viewing 11: Emerging technologies: radiobiology and physics hand in hand

reported International Prostate Symptom Score (IPSS) guidelines. A late toxicity event for each symptom was defined as an increase in IPSS grade ≥ 3 from good (grade 0/1) baseline. We studied four (frequency, nocturia, urgency and weak stream) GU symptoms with an event rate higher than 10%. We considered 14 previously reported clinical GU risk factors including RT dose to a tumor. The pre-conditioned random forest regression (PRFR) model (Oh et al., 2017, Scientific Reports) was trained using the SNPs as predictors with association p- value < 0.001 and clinical variables with p-value < 0.05. Model training and validation was respectively performed in randomly split training (2/3) and validation (1/3) datasets. Using a database of biological knowledge, we searched for enriched gene ontology biological processes and a group of connected proteins from the SNPs identified by PRFR as high importance. Results No clinical variable was significantly associated with any of the endpoints after Bonferroni correction. Performance of PRFR varied across symptoms: the areas under the curve (AUC) on the hold-out validation set were: weak stream: 0.7, frequency: 0.64, nocturia: 0.55, and urgency: 0.53. The AUC was significant only at weak stream (p = 0.01) where the odds ratio between the 1/3 lower and 1/3 higher risk groups was 3.6 (fig1). At this endpoint, the proposed model outperformed alternative multivariate methods including conventional random forest and logistic regression. Out of the 617 SNPs with high importance to weak stream, we discovered groups of enriched biological processes characterized by neurogenesis and ion transport (fig2), both of which had been shown to be involved in urinary tract functions. We also discovered a network of 15 proteins with interactions, among which 7 proteins (PKC, PKG, EGFR, Schwannomin, Annexin I, ASIC2, and Neurexin) were shown to be relevant to GU functions using systematic literature survey.

Fig1:Risk stratification for weak stream. Error bars=1 standard error.

Fig2: Biological processes for weak stream.