Back to the list
Congress: ECR25
Poster Number: C-16003
Type: Poster: EPOS Radiologist (scientific)
Authorblock: F. Palmeri, D. Caivano, A. Del Gaudio, M. Polici, S. Nardacci, M. Zerunian, F. Pucciarelli, D. Caruso, A. Laghi; Roma/IT
Disclosures:
Federica Palmeri: Nothing to disclose
Donatella Caivano: Nothing to disclose
Antonella Del Gaudio: Nothing to disclose
Michela Polici: Nothing to disclose
Stefano Nardacci: Nothing to disclose
Marta Zerunian: Nothing to disclose
Francesco Pucciarelli: Nothing to disclose
Damiano Caruso: Nothing to disclose
Andrea Laghi: Nothing to disclose
Keywords: Colon, Computer applications, Lung, CT, Stereotactic radiotherapy, Cancer, Multidisciplinary cancer care
Results

The final study population comprised 123 lung metastases from 80 patients (mean age, 71 ± 10 years, with 60 men). Of the 123 lesions, 31 (39%) progressed to polymetastatic disease. A multivariate analysis identified three clinical variables—age, neoadjuvant treatment, and adjuvant treatment—as significant predictors of progression (all p < 0.05), alongside five radiomics features. The clinical model achieved an area under the curve (AUC) of 0.69 (95% CI, 0.60–0.77), with a correct classification rate of 65.8%. In comparison, the radiomics model demonstrated an AUC of 0.65 (95% CI, 0.53–0.73) and a correct classification rate of 63.4%. The combined model, which integrated both radiomics and clinical features, yielded an AUC of 0.73 (95% CI, 0.65–0.81), improving the correct classification rate to 70%.

GALLERY