Radiomics-Based Computed Tomography Urogram Approach for the Prediction of Survival and Recurrence in Upper Urinary...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Upper tract urothelial carcinoma (UTUC) is an aggressive cancer with a poor prognosis.
• Accurate prediction of survival and recurrence is vital for risk stratification and treatment decisions in UTUC patients.

Purpose of the Study:

• To develop and validate a model combining radiomics and clinicopathological features for predicting survival and recurrence in UTUC.
• To assess the prognostic performance of CT-urogram-derived radiomics features in UTUC.

Main Methods:

• Analysis of preoperative CT urograms and clinical data from 106 UTUC patients.
• Extraction of radiomics features and selection using the LASSO method.
• Development of multivariable Cox models integrating radiomics and clinical factors.

Main Results:

• The combined radiomics and clinical model demonstrated superior predictive accuracy for survival (C-index: 0.73) and recurrence (C-index: 0.84).
• Kaplan-Meier analysis revealed significant differences in survival (p < 0.0001) and recurrence (p < 0.002) probabilities.
• Multivariable Cox models identified significant predictors for UTUC outcomes.

Conclusions:

• CT-urogram-based radiomics models effectively predict survival and recurrence in UTUC.
• Combining radiomics features with clinical factors significantly enhances prognostic performance for UTUC.