ESTRO 2025 - Abstract Book

S1684

Clinical - Sarcoma & skin cancer & malignant melanoma

ESTRO 2025

2199

Digital Poster Nomogram predicting wound complications in patients with soft tissue sarcomas treated with surgery and radiotherapy Paul-Adrien GUIGUE 1 , Benoit ALLIGNET 2,3 , Paul FROBERT 4 , Waisse WAISSI 2 , Marie-Pierre SUNYACH 2 , Gualter VAZ 5 , François GOUIN 5 1 Faculté de Médecine et de Maïeutique Lyon Sud - Charles Mérieux, Université Claude Bernard Lyon 1, Lyon, France. 2 Department of Radiation Oncology, Centre Léon Bérard, Lyon, France. 3 CREATIS, University Claude Bernard Lyon 1, Villeurbanne, France. 4 Department of Plastic and Reconstructive Surgery, Centre Léon Bérard, Lyon, France. 5 Orthopedic oncologic surgery department, Centre Léon Bérard, Lyon, France Purpose/Objective: Localized soft tissue sarcomas of limbs and trunk (STS) are treated with surgery and neoadjuvant or adjuvant radiotherapy (RT). Major wound complications (MWC) lead to prolonged care, iterative surgeries, infections, and may delay or prevent adjuvant oncology treatment. While predicting the risk of MWC could guide therapeutic strategies, no tool currently exists for this purpose. The aim of this study was to construct a predictive nomogram for this risk based on pragmatic data available before every treatment. Material/Methods: Retrospective case-control study including all STS patients who underwent surgery and RT in the Centre Léon Bérard (Lyon, France) from January 2018 to December 2022. The primary outcome was the occurrence of at least one MWC, defined as the need for wound care beyond the 28 th day after surgery. Evaluated variables included clinical data such as age, BMI, tumor localization, tumor long axis, history of diabetes and smoking and lesion depth, and the timing of RT (neoadjuvant vs adjuvant). Correlations were first analyzed with univariate logistic regression. Variables with p<0.05 and those identified in the literature were included in multivariate regression. The final model selected minimized Akaike’s criterion and excluded multicorrelated variables, defined as VIF>2. Results: Among the 191 patients included, 64 (33.5%) experienced MWC. 101 (52,8%) received neoadjuvant RT. The average age was 62 years. In univariate analysis, MWC was associated with neoadjuvant RT, age, lower limb localization, and tumor length. BMI was included based on its potential impact on MWC as reported in the literature. In multivariate regression, MWC tended to be correlated with age (HR = 1.01, p=0.056), BMI (HR = 1.04, p=0.17), and was significantly correlated with lower limb localization (HR = 2.12, p=0.03) and neoadjuvant RT (HR = 2.48, p=0.006) (Figure 1). The model’s ability to predict MWC was good, with an area under curve of 0.72 (Figure 2). Finally, in a pragmatic approach, an interactive nomogram was produced to assess individual patient risk (https://pauladrienguigue.shinyapps.io/Nomograme/).