ESTRO 2025 - Abstract Book

S2144

Clinical - Urology

ESTRO 2025

4604

Digital Poster A systematic review of models for predicting PROs and complications among patients with localized prostate cancer undergoing radiotherapy Claudia Cruz Oliveira 1 , Christian Jongen 2 , Wilma Heemsbergen 2 , David van Klaveren 1 1 Department of Public Health, Erasmus MC University Medical Center, Rotterdam, Netherlands. 2 Department of Radiation Oncology, Erasmus MC University Medical Center, Rotterdam, Netherlands Purpose/Objective: Despite significant advances in techniques for curative radiotherapy (RT) of localized prostate cancer (LPC), these treatments are still associated with urinary dysfunction, sexual dysfunction, and bowel dysfunction. Prediction models that incorporate patient characteristics may enable tailored decision-making and provide patients with more realistic expectations. However, it remains unclear which models can better identify patients at higher risk of developing adverse events from curative RT for LPC. Our aim was to systematically review the studies on the development and validation of models for predicting patient-reported outcomes (PROs) and complications following curative RT in males with LPC. Material/Methods: In collaboration with the medical library of Erasmus MC University Medical Center, we developed a search strategy, which included studies published between January 1, 2003, and September 6, 2024. Two independent researchers screened the included studies and extracted data on the types of RT, the predicted outcomes, and the main predictive factors (Figure 1). Studies presenting models that did not include patient or tumor characteristics, or otherwise focused on recurrent prostate cancer or non-curative RT were excluded. The PROBAST tool (A Tool to Assess the Risk of Bias and Applicability of Prediction Model Studies) was used to assess the quality of the included models.

Figure 1: Items extracted from the included study

Results: In total, 2288 titles and abstracts were screened, and 69 studies were included (Figure 2). The included studies provided over 200 prediction models of PROs or complications. Most of the models were developed on patients undergoing different modalities of External-Beam Radiation Therapy (EBRT; 69%), Brachytherapy (BT; 18%), combined BT with EBRT (8%), and Post-Prostatectomy Radiotherapy (PORT; 4%). The models mostly focused on