ESTRO 2023 - Abstract Book

S1896

Digital Posters

ESTRO 2023

on 18fluorodeoxyglucose (FDG) positron emission tomography (PET) and magnetic resonance imaging (MRI) in patients with locally advanced OPC treated with chemo-radiotherapy (CRT) or bio-radiotherapy (BRT). Materials and Methods Data of 82 patients with advanced OPC who underwent FDG PET and T2-weighted MRI before definitive CRT or BRT 08/2011 to 12/2019 were retrospectively analyzed. Patients had to meet the following eligibility criteria: histologically proven OPC; treated with definitive radiotherapy (RT) combined with concurrent chemotherapy or cetuximab; curative intent; use of IMRT or VMAT (photons) to a total dose of 70 Gy, respectively. Quantitative 428 PET and MR radiomics imaging characteristics were extracted from the primary and nodal tumor volumes, and multiple clinical variables included sex, age, tumor p16 status, tobacco were assessed. Pearson’s correlation was used to explore possible correlations among them. The radiomics score was constructed according to the least absolute shrinkage and selection operator (LASSO) method. Survival analysis (loco-regional control; LRC, disease-free survival; DFS, distant metastasis free survival; DMFS, and overall survival; OS) was performed using the log-rank test and Cox's proportional hazards regression model. We split the dataset into training and internal validation sets. Receiver operating characteristics (ROC) curve with 5-fold cross validation were used to evaluate their prognostic performance. Results The median follow-up period of the training cohort was 41 months (range, 13-72 months). The multivariate survival analysis showed that age (<67) (p = 0.032), HPV positivity (p = 0.022), and the low radiomics score (p = 0.005) were correlated with LRC. In DFS and DMFS, radiomics feature scores were independent prognostic factors. (p < 0.05) In addition, age (p = 0.032) and radiomics feature score (p = 0.004) were predictive of survival in OS. The combined radiomics model based on the multivariate analysis showed the best predictive ability for DFS, with an AUC of 0.828 [95% confidence interval (CI): 0.774- 0.873] in the training cohort and 0.821 (95%CI: 0.689-0.897) in the validation cohort. As shown in Figure, when stratified by the cutoff value of the radiomics feature score, CRT was superior to BRT for locoregional control in the poor prognosis group. (p = 0.045)

Conclusion The combined radiomic models based on FDG PET and MR images were independent prognostic factor for DFS and DMFS in patients with OPC. The radiomics score-based nomogram could improve prognostic stratification ability and thus contribute to individualized therapy for HPV-associated OPC patients.

PO-2112 Radiomics of diffusion-MRI for predicting Gleason Score in Prostate Cancer treated with radiotherapy

L. Morelli 1 , C. Paganelli 2 , G. Marvaso 3,4 , S. Annunziata 1 , G. Parrella 1 , M. Pepa 3 , M. Zaffaroni 3 , M.G. Vicini 3 , L.J. Isaksson 3 , G. Corrao 3,4 , P. Pricolo 5 , S. Alessi 5 , P.E. Summers 5 , F. Cattani 6 , O. De Cobelli 7,8 , R. Orecchia 9 , G. Petralia 5,10 , B.A. Jereczek- Fossa 3,10 , G. Baroni 11 1 Politecnico di Milano, Department of Electronics, Information and Bioengineering, Milan, Italy; 2 Politecnico di Milano, Department of Electronics, Information and Bioengineering, MIlan, Italy; 3 European Institute of Oncology IRCCS, Department of Radiotherapy, Milan, Italy; 4 University of Milan, Department of Oncology and Hemato-oncology, Milan, Italy; 5 European Institute of Oncology IRCCS, Division of Radiology, Milan, Italy; 6 European Institute of Oncology IRCCS, Medical Physics Unit, Milan, Italy; 7 University of Milan, Department ofOncology and Hemato-Oncology, Milan, Italy; 8 European Institute of Oncology IRCCS, Division of Urology, Milan, Italy; 9 European Institute of Oncology IRCCS, Scientific Directorate, Milan, Italy; 10 University of Milan, Department of Oncology and Hemato-Oncology, Milan, Italy; 11 Politecnico di Milano, Department of Electronics, Information and Bioengineering, Milan, Italy Purpose or Objective Prostate cancer (PCa) is the 2nd most common cancer in men. Several treatment options are available, making an accurate diagnostic and risk stratification essential to maximize the therapeutic benefits. Several oncological and imaging-based scores are conventionally used for PCa characterization, such as the Gleason Score (GS), National Comprehensive Cancer Network risk class (NCCN risk class), T-stage, extracapsular extension (ECE) score, and Prostate Imaging Reporting and Data System v2 (PIRADS). In this context, multiparametric MRI is showing promising results for extracting information sensitive to pathological differences in PCa, providing useful tools for non-invasive tumour characterization. The aim of this study is