Clinical and therapeutic factors associated with adverse pathological outcomes in clinically node-negative patients treated with neoadjuvant cisplatin-based chemotherapy and radical cystectomy

Purpose: Several disease characteristics have been identified as potential predictors for pathological node involvement (pN+) following radical cystectomy (RC). However, these have not been assessed in patients treated with neoadjuvant chemotherapy (NAC). We endeavored to assess factors predicting adverse pathology in clinically node-negative patients treated with NAC and RC. Methods: Patients from four North American institutions with cT2-4aN0M0 UC who received three or four cycles of NAC followed by RC were selected. Logistic regression was used to predict pN+, <pT2 and pT4 disease. Results: One hundred and ninety-six patients were included. The clinical stage was cT2 in 115 (61 %), cT3 in 62 (33 %) and cT4 in 12 (6 %) cases. NAC regiments were gemcitabine–cisplatin (GC)-4 cycles 57 (29 %), GC-3 cycles 77 (39 %), methotrexate, vinblastine, adriamycin, cisplatin (MVAC)-3 cycle 22 (11 %) and MVAC-4 cycles 40 (21 %). pN+ was seen in 35 (18 %) patients. In the logistic regression analysis, cT4 stage (OR 7.50; 95 % CI 1.58–33.3) and three compared to four cycles of GC (OR 3.44; 95 % CI 1.09–10.9) were significant predictors of pN+ status. Additionally, when controlling for clinical stage, three cycles of GC, compared to four, were significantly associated with higher rates of pT4 disease and lower rates of downstaging to non-muscle-invasive disease. Conclusions: The results suggest that four cycles of neoadjuvant GC may be superior to three cycles, and the latter regimen may be associated with adverse pathological findings. Although this would require validation in a prospective trial, it does encourage the completion of the conventional four cycles GC whenever possible.