Postoperative nomogram to predict cancer-specific survival after radical nephroureterectomy in patients with localised and/or locally advanced upper tract urothelial carcinoma without metastasis.

OBJECTIVE

To propose and validate a nomogram to predict cancer-specific survival (CSS) after radical nephroureterectomy (RNU) in patients with pT1-3/N0-x upper tract urothelial carcinoma (UTUC).

PATIENTS AND METHODS

The international and the French national collaborative groups on UTUC pooled data from 3387 patients treated with RNU. Only 2233 chemotherapy naïve pT1-3/N0-x patients were included in the present study. The population was randomly split into the development cohort (1563) and the external validation cohort (670). To build the nomogram, logistic regressions were used for univariable and multivariable analyses. Different models were generated. The most accurate model was assessed using Harrell's concordance index and decision curve analysis (DCA). Internal validation was then performed by bootstrapping. Finally, the nomogram was calibrated and externally validated in the external dataset.

RESULTS

Of the 1563 patients in the nomogram development cohort, 309 (19.7%) died during follow-up from UTUC. The actuarial CSS probability at 5 years was 75.7% (95% confidence interval [CI] 73.2-78.6%). DCA revealed that the use of the best model was associated with benefit gains relative to prediction of CSS. The optimised nomogram included only six variables associated with CSS in multivariable analysis: age (P < 0.001), pT stage (P < 0.001), grade (P < 0.02), location (P < 0.001), architecture (P < 0.001) and lymphovascular invasion (P < 0.001). The accuracy of the nomogram was 0.81 (95% CI, 0.78-0.85). Limitations included the retrospective study design and the lack of a central pathological review.

CONCLUSION

An accurate postoperative nomogram was developed to predict CSS after RNU only in locally and/or locally advanced UTUC without metastasis, where the decision for adjuvant treatment is controversial but crucial for the oncological outcome.