Perirenal Fat CT Radiomics-Based Survival Model for Upper Tract Urothelial Carcinoma: Integrating Texture Features...

Area of Science:

• Urology
• Radiology
• Oncology

Background:

• Upper tract urothelial carcinoma (UTUC) is rare and challenging to prognosticate.
• Traditional prognostic factors for UTUC are limited.
• Novel imaging biomarkers are needed for improved risk stratification.

Purpose of the Study:

• To develop and validate a prognostic model for UTUC using perirenal fat (PRF) radiomics and clinical factors.
• To compare the performance of a combined model against radiomics-only and clinical-only models.
• To assess the prognostic value of PRF radiomics features in UTUC.

Main Methods:

• Retrospective analysis of 103 UTUC patients undergoing radical nephroureterectomy.
• Extraction of PRF radiomics features from preoperative CT scans using semi-automated segmentation.
• Development and validation of clinical, radiomics, and combined prognostic models using C-index, time-dependent AUC, and Brier score.

Main Results:

• The combined model showed superior prognostic performance (C-index: 0.784) compared to radiomics (0.759) and clinical (0.653) models.
• Radiomics model excelled in short-term prognosis (12-month AUC: 0.9281), while the combined model was better for long-term predictions (60-month AUC: 0.8403).
• PRF radiomics features demonstrated stronger prognostic value than traditional clinical factors.

Conclusions:

• Integration of PRF radiomics with clinical data significantly enhances UTUC prognostic accuracy.
• This approach offers deeper insights into the tumor microenvironment and early invasion.
• The semi-automated PRF segmentation method is reproducible and potentially implementable in clinical practice.