Recurrence- and progression-free survival in intermediate-risk non-muscle-invasive bladder cancer: the impact of conditional evaluation and subclassification.

Abstract

To assess the change in rates of recurrence-free survival (RFS) and progression-free survival (PFS) based on the duration of survival without recurrence or progression among patients with intermediate-risk (IR) non-muscle-invasive bladder cancer (NMIBC), and to examine the predictive factors for recurrence at different time points by assessing conditional RFS and PFS. A cohort of 602 patients treated with transurethral resection of bladder tumour and histopathologically diagnosed with IR NMIBC was included in this retrospective study. The conditional RFS rate at 1, 2, 3, 4 and 5years improved with increased duration of RFS; however, the conditional PFS rate did not improve over time. Multivariable analyses showed that recurrent tumour, multiple tumours, tumour size (>3cm), immediate postoperative instillation of chemotherapy, and administration of BCG were independent predictive factors for recurrence at baseline. The predictive ability of these factors disappeared with increasing recurrence-free survivorship. Subclassification of these patients with IR NMIBC into three groups using clinicopathological factors (recurrent tumour, multiple tumours, tumour size) demonstrated that the high IR group (two factors) had significantly worse RFS than the intermediate (one factor, P < 0.001) and low IR groups (no factor, P = 0.005) at baseline. This subclassification stratified conditional risk of RFS also at 1, 3 and 5 years, which provides the basis for distinct surveillance protocols among patients with IR NMIBC. Conditional survival analyses of patients with IR NMIBC demonstrate that RFS changes over time, while PFS does not change. These data support distinct surveillance protocols based on the subclassification of IR NMIBC.