Development of an Enomogram to Predict the Rate of Loco-Regional Control After Radio-Chemotherapy and Interventional...

Objective: This study aimed to explore the association between magnetic resonance imaging (MRI)-derived volumetric parameters and oncological outcomes, and to develop an exploratory predictive model based on these variables in patients treated with radio-chemotherapy followed by interventional radiotherapy (modern brachytherapy). Methods: Between 2021 and 2024, 300 patients with cervical cancer were included. Treatment was pelvic external beam radiotherapy with platinum-based chemotherapy followed by interventional radiotherapy boost. Volumetric MRI variables for each patient were collected. Time-to-event analyses were performed using Cox proportional hazards regression models. Model performance was assessed using Harrell's concordance index (C-index). Internal validation was performed using bootstrap resampling. Based on the final multivariable Cox models, an interactive web-based nomogram was developed as an exploratory tool to visualize model-derived associations. Results: Median tumor volume decreased from 69.4 cm³ at diagnosis to 2.2 cm³ at the time of pre-interventional radiotherapy MRI, with a median reduction rate of 96.5%. Tumor volume at diagnosis, pre-interventional radiotherapy residual tumor volume, and tumor volume reduction rate were significantly associated with loco-regional relapse and distant metastases in Cox regression analyses. These findings were consistent across univariate and multivariable models. Internal validation confirmed the stability of the model estimates. Conclusions: MRI-derived volumetric parameters are associated with oncological outcomes in patients with locally advanced cervical cancer and may contribute to early risk stratification. The proposed model should be considered exploratory and hypothesis-generating and requires external validation before any potential clinical application.