S1449
Interdisciplinary - Other
ESTRO 2026
and, indirectly, of quality of care (1,2). This study aims to evaluate the adequacy of palliative RT among patients treated at Instituto Português de Oncologia- Porto (IPO-Porto) who died within 30 days after RT initiation, characterizing their clinical profile and
2023;66(6):638–46.3. Park KR et al. Palliative radiation therapy in the last 30 days of life: A systematic review. Radiother Oncol. 2017;125(2):193–9. Keywords: Paliative, Radiation Therapy, End of Life
treatment patterns. Material/Methods:
Digital Poster Highlight 3137 A data-driven model for identifying endophenotypes associated with prostate cancer incidence Lucas Uglione Da Ros 1,2 , Daniel Arnold 2 , Eduardo Rigon Zimmer 2 , Wyllians Vendramini Borelli 2 1 Radiotherapy, HCPA, Porto Alegre, Brazil. 2 PPGFT, UFRGS, Porto Alegre, Brazil Purpose/Objective: Prostate cancer (PC) is the most common malignancy in men and the fifth leading cause of cancer-specific deaths. Despite its high prevalence, PC incidence varies by over 50-fold across different populations, with established risk factors largely limited to age, race, and family history. While PC screening has been
A retrospective study was conducted including all adult patients who initiated palliative RT at IPO-Porto in 2023. Demographic, clinical and treatment data were collected from electronic medical records. Patients who survived ≤30 days from the start of RT were analysed separately. Variables included age, sex, ECOG Performance Status (PS), primary tumour, RT intent, fractionation and treatment completion. Descriptive statistics were applied to assess patterns of care and patient characteristics. Results: Of 667 patients treated with RT in 2023, 94 (14%) died within 30 days. Of these patients median age was 67.5 years (range 37–94), and 62% were male. Median survival from RT start was 19 days (range 1–30). ECOG PS was 1 in 15%, 2 in 32%, 3 in 26%, and 4 in 12% of patients (unknown 16%).The primary tumours among patients were: lung (28%), head and neck (9.6%), colorectal (9.6%), gastric (9.6%), breast (7.4%) and sarcomas (7.4%). The indications for palliative RT were pain relief (49%), haemostatic (20%), brain metastasis control (11%), nodal metastasis control (10%), spinal cord compression (4%), superior vena cava syndrome (SVCS) (3%) and others. Fractionation regimens included single-fraction (39%), 5 fractions (30%), 10 fractions (10%), and 12 fractions (2%).Overall, 83% of patients completed the planned RT. Seventeen percent of patients discontinued treatment. Among these, half discontinued due to death, most frequently in the context of superior vena cava syndrome (33.3%), whole-brain radiotherapy (WBRT) for brain metastases (30%), and nodal metastases (22.2%). The remaining half stopped treatment due to clinical deterioration, most commonly during WBRT (40%) and nodal irradiation (22.2%). Conclusion: Our results, showing that 14% of patients receiving radiotherapy died within 30 days, are in line with the literature. This is consistent with reports that around 16% of patients with advanced cancer undergoing palliative radiotherapy die within 30 days, a rate suggested as a benchmark for global quality metrics in radiation oncology practice audits (1). References: 1. Kutzko JH et al. Defining the expected 30-day mortality for patients undergoing palliative radiotherapy: A meta-analysis. Radiother Oncol. 2022;168:147–210.2. Davis MP et al. Radiation therapy in the last month of life: Association with aggressive care at the end of life. J Pain Symptom Manage.
extensively investigated, its population-level implementation remains controversial, most
guidelines recommend shared decision-making, but some high risk population may benefit the most from this approach. The present study aimed to investigate how endophenotypes correlate with prostate cancer incidence using a data-driven modeling approach. Material/Methods: Data from the UK Biobank were utilized. A comprehensive set of variables was gathered, including demographics (e.g., age), health status (e.g., blood pressure, blood glucose levels), environmental exposures (e.g., air pollution), sociodemographic factors (e.g., Townsend Deprivation Index), behavioral factors (e.g., physical activity, alcohol intake), functionality (e.g., intelligence score), and prostate cancer polygenic risk scores (PRS). To address the high-dimensional nature of the dataset, Uniform Manifold Approximation and Projection was applied for dimensionality reduction, transforming the multivariate data into a lower-dimensional space while preserving the intrinsic structure and relationships between variables. Following dimension reduction, Hierarchical Density-Based Spatial Clustering of Applications with Noise was employed to identify distinct population clusters based on the reduced- dimensional representation. The clustering approach allows for the identification of groups with similar characteristics without predetermining the number of clusters.The outcome was the occurrence of PC, those diagnosed with PC prior to baseline assessment were excluded. Results: The model effectively clustered the population into 12
Made with FlippingBook - Share PDF online