Comprehensive long-term assessment of outcomes following robot-assisted partial nephrectomy for renal cell carcinoma: the ROMe's achievement and its predicting nomogram.

Abstract

Background: We proposed a new tool (named ROMe's) to summarize long-term outcomes after partial nephrectomy (PN), identified its predictors and generated a predicting nomogram.

Methods: A retrospective analysis of a multicenter dataset of patients with non-metastatic pT1-3a renal cell carcinoma was performed. Baseline demographic, clinical, pathologic and perioperative data were collected. ROMe's was defined as the concomitant lack of cancer-recurrences, death and newly onset Chronic Kidney Disease (CKD), at long term follow-up. Kaplan-Meier method investigated the predictive role of Trifecta on ROMe's achievement. Univariable and multivariable Cox regression analyses identified its predictors. A nomogram was generated and its accuracy was quantified using concordance index (CI). A calibration plot was obtained with 200 bootstraps resampling to explore nomogram performance at 5 years and decision curve analyses (DCA) assessed the net benefit of the model at 12, 36 and 60 months.

Results: We included 927 patients. The rates of ROMe's were 82%, 72% and 56% at 1, 3 and 5 years follow-up. At Kaplan-Meier analysis, patients who achieved Trifecta displayed a significantly higher probability of ROMe's (log rank P<0.001). Young age (OR=0.982; P=0.001), low RENAL score (OR=0.86; P=0.037), high preoperative filtration rate (OR=1.02; P<0.001) and Trifecta achievement (OR=2.03; P=0.015), were independent predictors of ROMe's. The nomogram showed a CI of 0.76 at 60 months. The 5-years calibration plot confirmed a good discrimination accuracy (0.74); on DCA, the net benefit of using the model was evident for probabilities >30%.

Conclusions: We conceived a triad to summarize the main long-term oncologic and functional outcomes after PN and generated a predicting nomogram.