Combined reporting of surgical quality, cancer control and functional outcomes of robot-assisted radical cystectomy with intracorporeal orthotopic neobladder into a novel trifecta.

BACKGROUND

To optimize outcomes reporting after robot-assisted radical cystectomy (RARC) with intracorporeal orthotopic neobladder (iON), we propose a novel trifecta, aimed to overcome the available pentafectas that neglect functional outcomes.

METHODS

A retrospective analysis of prospectively maintained RARC-iON dataset was performed. Baseline demographic, clinical, pathologic, perioperative and follow-up data were collected. Trifecta was defined as the coexistence of daytime urinary continence, Clavien-Dindo ≥3 complication-free and recurrence-free status, all assessed at 1 year. Kaplan-Meier method assessed the role of trifecta achievement in predicting overall survival (OS) probabilities. Univariable and multivariable regression analyses identified predictors of trifecta achievement. The predictive accuracy of trifecta and other pentafectas on 3-year OS probability was plotted with receiver operating characteristic (ROC) analyses.

RESULTS

The trifecta rate was 53% while 62% and 47% of patients achieved the University of Southern California (USC-P) and PROMETRICS group (PROM-P) pentafectas, respectively. On tertile analysis, both trifecta (P=0.011) and USC-P (P=0.043) rates significantly increased with surgical experience, while PROM-P did not (P=0.204). On ROC analysis, trifecta was the only significant predictor of 3-year OS probability (AUC=0.69, 95% CI:0.55-0.82; P=0.018). On Kaplan-Meier analysis, patients achieving trifecta displayed significantly higher OS probability (P=0.032); adjusted for covariates, each incremental case was associated with a 2% increased probability (OR: 1.02) of achieving trifecta on multivariable regression model.

CONCLUSIONS

Learning curve in RARC-iON has a significant impact on achievement of trifecta. This tool combines accurate measurement of surgical proficiency with an independent predictive value in assessing 3-year OS probabilities.