Efficacy of Preoperative Chemotherapy on Outcomes of High-risk Upper Tract Urothelial Carcinoma (UTUC)

View Poster


The effect of preoperative chemotherapy (CHT) on disease outcomes in patients with upper urinary tract urothelial carcinoma (UTUC) is unclear yet. The insufficient data on this relatively rare disease and the lack of prospective trials preclude definitive conclusions about the effect of CHT. We assessed the rates of complete pathologic response (CPR; pT0N0) and partial pathologic response (PPR; ≤pT1N0) rates after different preoperative CHT regimens (MVAC, GC and others) in patients with high-risk clinically non-metastatic UTUC treated with radical nephroureterectomy (RNU). We also assessed predictive factors of response.


This was an international multicenter (n=10) retrospective study of patients who underwent preoperative CHT and RNU for clinically non-metastatic UTUC between 2002 and 2017. Univariable and multivariable logistic regression analyses were performed to identify predictors of pT0N0 and ≤pT1N0 stages at RNU. We investigated primarily the number of cycles and different therapy regimens such as Methotrexate, Vinblastine, Adriamycin and Cisplatin (MVAC), Gemcitabine and Cisplatin (GC), Carboplatin-based regimens and others.


A total of 212 patients met the inclusion criteria and were eligible for the study. The majority received Cisplatin-based CHT (n=167, 78.8%), followed by other regimens (n=28, 13.2%) and Carboplatin-based CHT (n=17, 8.0%). Overall, 82 patients (38.7%) were treated with MVAC regimens (dose-dense included) and 84 (39.6%) with GC regimens. Thirty patients (14.2%) received 1-2 cycles, 165 (77.8%) 3-4 and 17 (8.0%) 5-6 cycles. The rates of CPR and PPR were 9.9% and 30.7%, respectively. Among patients treated with MVAC, the rates of CPR and PPR were 15.9% and 36.6% compared to 7.1% and 26.2% with GC and 4.3% and 28.3% with other regimens, respectively. In multivariable analysis, GC (OR 0.43, p=0.1) and other regimens (OR 0.25, p=0.07) were not inferior to MVAC for CPR, after adjusting for received cycles. Also for prediction of PPR, the regimen choice such as GC (OR 0.71, p=0.3) and other regimens (OR 0.75, p=0.5) showed no significant difference to MVAC. Administration of three or four cycles was not independently associated with CPR or PPR (OR 1.16, p=0.9 and OR 1.18, p=0.8) compared to one or two cycles. We observed similar results in patients who received five or six cycles (OR 2.8, p=0.1 and OR 2.7, p=0.08).


Although the rates of CPR and PPR were higher for MVAC compared to GC and other regimens, we did not observe a significant difference in multivariable analysis. In addition, the number of cycles did not appear to have an influence on pathologic response. Further studies are needed to identify the best preoperative chemotherapy regimens for high-risk UTUC patients.

Funding: none