Login to Access Video or Poster Abstract: MP21-02
Sources of Funding: none

Introduction

Around one third of bladder cancers (BCs) are high-grade non-muscle invasive tumors (HGNMI). Current guidelines advocate early re-resection for these cancers, although the benefits are unclear and the uniform need has been questioned. Here we compare the outcomes in patients with and without re-resection using a large single-centre cohort._x000D_

Methods

We identified all patients with new HGNMI BC treated between 1994-2009 in Sheffield. We annotated these with hospital, pharmacy and cancer registry records. Primary outcomes were disease specific and overall survival. Secondary outcomes were the findings at re-resection, rates of muscle invasion and radical treatment. Statistical tests were two-tailed and significance defined as p<0.05._x000D_ _x000D_

Results

We identified 932 eligible patients, including 229 (25%) who underwent re-resection within 12 weeks and 234 (25%) 3-6 months after diagnosis. Patients with and without re-resection were broadly similar for clinicopathological criteria. Re-resection was normal in 91 (20%) and contained BC in 138 (30%) patients. Of these, 15(10.8%) cancers were low grade, 85(61.6%) high-grade NMI and 38 muscle invasive (28%). During follow up, patients with re-resection were more frequently diagnosed with muscle invasion (126 (27%) vs. 49 (11%), Chi sq. p<0.001) and more commonly underwent radical treatment (127 (27%) vs. 35 (8%), p<0.001) than those without re-resection. In total, 528 (57%) patients died during follow up. Patients with re-resection had a significantly higher disease-specific (179 (78%) vs. 518 (76%), log rank p=0.05) and overall survival (119 (53%) vs. 251 (37%), log rank p<0.001) than those without re-resection._x000D_

Conclusions

We found that patients undergoing early re-resection were more likely to be diagnosed with muscle invasion, more likely to undergo radical treatment and had a higher disease-specific and overall survival. The differences were greatest in patients with lamina propria invasion at diagnosis. Limitations of our work include retrospective design and potential selection bias.

Funding

none