Introduction:

Given the high risk of recurrence, patients with BCG-unresponsive non-muscle invasive bladder cancer (NMIBC) are recommended to undergo radical cystectomy (RC) with urinary diversion, which has demonstrated excellent clinical outcomes. These recommendations are based on the FDA and SUO definitions for BCG-unresponsiveness, which includes patients with carcinoma in situ (CIS), Ta, or T1-staged tumors who do not respond to or recur after BCG therapy. Data exists on oncologic outcomes following RC in BCG-unresponsive NMIBC, but there are minimal data on the impact of tumor stage at BCG-unresponsiveness on oncologic outcomes following RC. Furthermore, this data is derived mostly from single institution series. The purpose of this study was to compare oncologic outcomes in patients undergoing upfront RC for T1 or Ta BCG-unresponsive NMIBC with those with pure CIS in a large multi-center cohort.

Methods:

At 10 academic institutions, patients with BCG-unresponsive NMIBC as defined by FDA and SUO criteria were identified and compiled into a single database. Data were collected on demographic variables; clinical tumor stage at time of BCG-unresponsiveness; pathologic outcomes at time of RC; and the incidence and date of metastasis, cancer-specific mortality, and all-cause mortality. The Kaplan-Meier method was utilized to determine survival differences as defined by length of time from BCG-unresponsiveness to event between T1, Ta, and CIS only patients. Statistical differences were determined using log-rank tests. Pathologic outcomes were compared between groups using Chi-squared tests.

Results:

Of 162 patients who underwent upfront RC for BCG-unresponsive NMIBC, 88 were T1, 34 Ta, and 40 CIS alone prior to surgery. Patients with papillary disease were more likely to have pathologic extravesical (> T3) or node positive disease compared to CIS alone (Table). At 60-month median follow up, metastasis occurred in 7%, 8%, and 13% of T1 patients by 12, 24, and 60 months compared to 0%, 12%, and 16% of Ta patients and 2%, 2%, and 9% of CIS alone patients (p=0.488). Cancer-specific mortality occurred in 5%, 10%, and 18% of T1 patients at these timepoints compared to 6%, 6%, and 14% of Ta patients and 5%, 5%, and 5% of CIS alone patients (p=0.511) (Figure). All-cause mortality occurred in 9%, 15%, and 26% of T1 patients compared to 10%, 10%, and 22% of Ta patients and 5%, 8%, and 12% of CIS alone patients (p=0.326).

Conclusion:

The pathologic findings from this study demonstrate that patients with concomitant papillary disease have greater rates of extravesical and node-positive disease compared to CIS alone. In the context of a patient cohort which entirely underwent upfront RC, these findings are unsurprising as patients with higher clinically staged tumors are also expected to have concordantly worse pathologic outcomes at the same time points. However, interestingly, oncologic outcomes did not statistically differ between T1, Ta, and CIS alone groups, suggesting that RC is equally efficacious in preventing further disease progression regardless of tumor stage at BCG-unresponsiveness. The findings from this cohort of select, non-randomized patients reaffirm the long term oncologic durability of upfront RC despite stage at time of NMIBC in patients with BCG-unresponsive disease.

Funding: N/A