Introduction

Neoadjuvant chemotherapy (NAC) before radical cystectomy improves survival expectations in cT2-T4a bladder cancer (BCa) patients. However, this effect is more evident in patients with cT3 or higher disease. Several studies evaluated factors associated with might help in selecting patients who might benefit more from NAC before RC. However, none of them investigated the role of the completeness of transurethral resection (TUR) before RC which may represents a useful proxy of tumor extension in RC candidates. We therefore evaluated a cohort of patients treated with RC due to BCa stratifying according the completeness of TUR before RC.

Methods

We retrospectively analyzed data of 481 patients treated with RC and bilateral PLND without NAC due to nonmetastatic BCa between 1990 and 2013 and complete data regarding the completeness of pre-cystectomy TUR. TUR were completed when possible, when not feasible physicians reported this aspect in a prospective maintained database. Demographics and preoperative RC data were available for all patients. Univariable and multivariable logistic regression were built predicting the impact of completeness of TUR and adverse pathologic T stage (defined as T3-T4), lymph node invasion (LNI) and positive soft tissue surgical margin (STSM) status. Multivariable models were adjusted for clinical T stage, presence of CIS or lymphovascular invasion at TUR, histology at TUR, age and preoperative idronephrosis. Area under curve (AUC) was calculated predicting adverse pathologic features with and without the completeness of TUR.

Results

Overall, TUR has been completed in 326 (67.8%) patients submitted to RC. Patients where the TUR was not completed had higher cT3-4 disease, higher LVI, higher CIS and higher cN+ diseases. At univariable analyses, incompleteness of TUR was a predictor of_x000D_ LNI (odds ratio [OR]: 1.66, confidence interval [CI]: 1.08-2.55, p=0.02), adverse pathologic stage (OR: 1.78, CI: 1.21-2.62, p=0.003), positive STSM (OR: 2.15, CI: 1.13-4.10, p=0.02). At multivariable analyses, completeness of TUR was a predictor of positive STSM (OR: 2.18, CI: 1.12-4.25, p=0.02) and adverse pathologic T stage (OR: 1.57, CI: 1.03-2.40, p=0.03) but not in the prediction of LNI (p=0.06). The inclusion of completeness of TUR in our preoperative model increases of 3.5% its accuracy (AUC: 76.2 vs. 72.7 with or without completeness of TUR, respectively).

Conclusions

Incompleteness of the TUR before RC represents a predictor of adverse pathologic features at RC. Physicians should consider this aspect in predicting RC patients’ survival and eventually the necessity of neoadjuvant chemotherapy.

Funding

none