Risk Factors for Unfavorable Pathological Types of Intravesical Recurrence in Patients With Upper Urinary Tract Urothelial Carcinoma Following Radical Nephroureterectomy

BACKGROUND: Numerous studies have investigated the risk factors of intravesical recurrence (IVR) after radical nephroureterectomy (RNU) in patients with upper urinary tract urothelial carcinoma (UTUC). However, few studies explore the predictors for unfavorable pathological types of IVR following RNU. METHODS: We retrospectively reviewed 155 patients diagnosed with bladder cancer (BC) following RNU. Binary logistic regression was used for the univariable and multivariable analyses. Nomograms were developed based on the multivariable analysis. The concordance index (C-index) was used to assess the performance of the nomograms. We performed internal validation by generating calibration plots. RESULTS: Muscle-invasive BC (MIBC) was significantly correlated with operation interval (p = 0.004) and UTUC T-stage (p = 0.016). Operation interval (p = 0.002) and UTUC T-stage (p = 0.028) were also risk factors for BC > 3 cm. UTUC grade (p = 0.015), operation interval (p = 0.003), and hydronephrosis (p = 0.049) were independent predictors for high-grade BC (HGBC). MIBC (p = 0.018) and surgical approach (p = 0.003) were associated with multifocal IVR. Besides, MIBC and HGBC were associated with UTUC grade (p = 0.009), operation interval (p = 0.001), and hydronephrosis (p = 0.023). Moreover, only operation interval (p = 0.036) was a predictor for BC with at least one unfavorable pathological type. We developed nomograms for MIBC, HGBC, BC > 3 cm, and MIBC and/or HGBC. The calibration curves of the nomograms showed good agreement between the observation and prediction cases. The C-indexes of the nomograms were 0.820 (95% CI, 0.747–0.894), 0.728 (95% CI, 0.649–0.809), 0.770 (95% CI, 0.679–0.861), and 0.749 (95% CI, 0.671–0.827), respectively. CONCLUSIONS: The current study found that operation interval, UTUC T-stage, UTUC grade, surgical approach, and hydronephrosis are independent predictors for unfavorable pathological types of IVR following RNU. Nomograms based on these predictors were developed and internally validated to assess the risk of developing unfavorable pathological types of IVR. Furthermore, patients at high risk of developing unfavorable pathological types BC may benefit from more active follow-up 1 year after RNU by early detection of IVR.