Purpose: Data from the pre-neoadjuvant chemotherapy (NAC) era suggests patients who progress on bacillus Calmette-Guérin (BCG) to muscle-invasive bladder cancer (P-MIBC) exhibit worse outcomes compared to de novo MIBC (D-MIBC). Herein, we investigate whether P-MIBC is an independent poor risk factor in the setting of contemporary NAC use.
Materials and methods: A review of patients who underwent radical cystectomy (RC) for cT2-3 MIBC from 2005 to 2018 was performed. Patients were stratified into high risk (lymphovascular invasion, variant histology, hydronephrosis, cT3b) vs low risk (no risk factors) and P-MIBC (≤pT1 treated with at least induction BCG who progressed to ≥cT2) vs D-MIBC.
Results: Among 801 patients who underwent RC 20.3% had P-MIBC and 79.7% had D-MIBC. In low-risk patients treated without NAC, P-MIBC was associated with pathological upstaging (64.9% vs 42.7%, p=0.004) and worse overall (OS, p=0.006) and cancer-specific survival (CSS, p=0.001) compared to D-MIBC. P-MIBC status conferred uniformly poor survival outcomes to patients who did not receive NAC compared to D-MIBC without NAC (median OS 51.5 months [95% CI 40.0-81.0] vs 85.1 months [95% CI 62.8-96.0], p=0.040; median CSS not reached, p=0.014). However, P-MIBC status did not remain a negative prognostic factor in the setting of NAC (median OS 90.5 months [95% CI 34.0-not estimable] vs 87.8 months [95% CI 68.7-not estimable], p=0.606; median CSS not reached, p=0.448).
Conclusions: P-MIBC confers a poor prognosis when managed with RC alone. Treatment with NAC results in equivalent pathological response and survival outcomes compared to D-MIBC. P-MIBC should be included in risk-stratified approaches to NAC selection.
Keywords: cystectomy; mycobacterium bovis; neoadjuvant therapy; urinary bladder neoplasms.