Computed tomography-based nomogram for estimating progression-free survival probability in bladder cancer patients undergoing partial cystectomy

Abstract

Purpose

To establish a prognostic model to estimate progression-free survival (PFS) probability in bladder cancer (BCa) patients undergoing partial cystectomy.

Material and Methods

Consecutive patients who underwent partial cystectomy between August 2012 and April 2021 were enrolled. The primary endpoint was PFS during the follow-up. The following features were assessed: tumor location, amount, size, tortuous blood vessels around or within the lesions, perivesical fat stranding, stalk, computed tomography (CT) enhancement, calcification, cystic degeneration, CT reported lymph nodes status, and presence of hydronephrosis. Univariate Cox regression and LASSO regression, followed by backward stepwise multivariable Cox, were used to construct the nomogram.

Results

A total of 106 patients were enrolled. Gender, histology, tortuous blood vessels, and perivesical fat stranding were used to fit the nomogram. The overall Harrell’s concordance index (C-index) was 0.752. The area under the receiver operator characteristic curves (AUC) at 1-year, 2-year and 3-year were 0.733, 0.789 and 0.833, respectively. The calibration curves showed remarkable consistency.

Conclusion

A nomogram model constructed based on the CT features and clinical risk factors is potentially feasible for predicting the PFS within 3 years after PC for BCa, which can assist in the choice of treatment and follow-up scheduling.