ESTRO 2023 - Abstract Book

S1025

Digital Posters

ESTRO 2023

factors were analyzed: age, family history, clinical presentation, nuclear grade, pathological tumor , multifocality, comedonecrosis, estrogen and progesterone receptors, final surgical resection margin, number of reinterventions, adjuvant treatments with radiotherapy and hormone therapy. Clinicopathologic and treatment parameters of all patients were used to create a multivariable model and the model of MSKCC was tested. The predictive value of the model was evaluated using Software for Statistics and Data Science release 15.1 (STATA). Statistically significant differences among variables were evaluated as independent risk factors using multivariate analysis with logistic regression. Results The median follows up time was 10.3 years (1-17), with a mean age of 58 years (42-75). The external validation of the MSKCC nomogram was performed with a total of 207 patients not reaching statistical significance in the studied population to predict LR (p= 0.40). In 7 patients (3.4%) follow-up was lost. The risk factors analyzed are described in table 1. Non expression of estrogen receptors was significantly associated with an increased risk of LR (p= 0.004). The omission of adjuvant endocrine therapy presented a slight tendency (OR 0.39), but without significance, to an increased risk of LR. The overall local relapse rate was 9.6% (20 patients) during the study period. At the end of the study, 94.6% (196 patients) were alive and 5.4% (11 patients) were dead.