BCG response and oncological outcomes in high risk nonmuscle invasive bladder cancer following previously treated upper tract urothelial carcinoma: A propensity-matched analysis

Abstract

Introduction

Metachronous bladder recurrences after prior treatment for primary upper tract urothelial carcinoma (UTUC) can occur in ∼3% to 50% of patients. Because UTUC demonstrated distinct molecular alterations, bladder recurrences in these patients may be molecularly and phenotypically different compared to primary bladder carcinoma. We aim to study the BCG efficacy in patients with primary high risk nonmuscle invasive bladder cancer (P-NMIBC) and metachronous bladder recurrences after previous nephroureterectomy for UTUC (M-NMIBC).

Methods

We reviewed an IRB-approved prospective uro-oncology database of patients who underwent resection followed by BCG therapy for high grade NMIBC from 2017 to 2021. Clinicopathological parameters, intravesical therapies and the oncological outcomes were analyzed. Patients in the P-NMIBC group were matched to patients in the M-NMIBC cohort (control) via propensity score matching (PSM) to adjust for potential clinicopathological confounders. Nearest-neighbor PSM targeting a 4:1 ratio of study to control subjects was performed using a caliper of 0.2, aiming for an absolute standardized mean difference of <0.1 across key covariates. Secondary outcomes were progression to distant metastasis and overall survival. Logistic and cox regression analyses were performed to elucidate independent variables associated with intravesical recurrences and disease progression.

Results

Of the 183 patients diagnosed with NMIBC, 35 patients were identified to have a history of UTUC with radical nephroureterectomy. EAU risk stratification revealed 50 (27.3%) intermediate risk, 107 (58.5%) high risk and 26 (14.2%) very high risk groups. P-NMIBC patients were more likely to have symptomatic presentation (79.7% vs. 23.9%), and a larger mean tumor size (25.7 mm vs. 15.4 mm) than M-NMIBC. The mean follow-up duration for the study was 34.0 months. In the unmatched analysis, M-NMIBC was associated with increased risk of HG intravesical recurrence post BCG compared to P-NMIBC (54.3% vs. 28.4%, P = 0.006, HR 2.14, 95% CI: 1.25–3.65) and increased risk of progression to MIBC (28.6% vs. 4.7%, P = 0.007, HR 4.19, 95% CI: 1.47–11.95). For the propensity-matched analysis, the control group consisted of 35 M-NMIBC matched to 123 P-NMIBC patients for similar demographics, EAU risk score and BCG doses. M-NMIBC again demonstrated a higher HG intravesical recurrence rate (54.3% vs. 22.8%, P = 0.001, HR 2.67, 95% CI: 1.50–4.77), progression to MIBC (28.6% vs. 5.7%, P = 0.022, HR 3.42, 95% CI: 1.20–9.75) and progression to distant metastasis (20.0% vs. 6.5%, P = 0.033, HR 3.02, 95% CI: 1.09–8.35). Overall survival in both groups were not significantly different in both unmatched and matched analysis.

Conclusions

Our study indicates that BCG treatment may be less effective for NMIBC patients with a history of UTUC, with a higher risk of intravesical recurrences and disease progression. This is an important consideration when counselling patients for BCG treatment and overall prognostication.