Pathological downstaging as a novel endpoint for the development of neoadjuvant chemotherapy for upper tract urothelial carcinoma

Objectives To evaluate whether pathological downstaging (pDS) was more informative in predicting overall survival (OS) than pathological complete response (pCR) in patients treated with neoadjuvant chemotherapy (NAC) for upper tract urothelial carcinoma (UTUC). Patients and Methods The National Cancer Database was queried for patients with high-grade cN0M0 disease who had received NAC. pDS was defined as a decrease of at least one stage from cT to pT stage along with pN0, including pCR. A multivariable Cox model predicting OS was generated by fitting alternatively either pDS or pCR, and adjusted for potential confounders. The discrimination of the Cox models for predicting OS was evaluated using Harrell's C-index. The analyses were repeated in patients diagnosed as having cT2-4N0M0 disease. Results Among 264 patients meeting the inclusion criteria, 72 (27%) and 39 (15%) achieved pDS and pCR, respectively. On multivariable analysis, both pDS (hazard ratio [HR] 0.24, 95% confidence interval [CI] 0.13, 0.45; P < 0.001) and pCR (HR 0.37, 95% CI 0.18, 0.79; P = 0.01) were associated with OS. The model including pDS achieved better discrimination with respect to the model including pCR: C-index 76.4 vs 72.7, respectively. In the 128 patients diagnosed with cT2-4 disease, both pDS (HR 0.19, 95% CI 0.09, 0.40; P < 0.001) and pCR (HR 0.31, 95% CI 0.11, 0.85; P = 0.023) were confirmed as predictors of OS. The model including pDS was confirmed to discriminate better than the model including pCR: C-index 75 vs 68.9, respectively. Conclusion The study showed that pDS after NAC for UTUC was more informative than pCR when predicting OS. These findings, although requiring prospective validation, can aid in the design of clinical trials seeking to refine the use of chemotherapy and other systemic therapies in this setting.