Does time to recurrence after radical cystectomy for bladder cancer impact subsequent survival?

Abstract

874 Background: It has been long theorized that patients with earlier recurrence after radical cystectomy (RC) for bladder cancer have worse subsequent survival. However, it remains unclear whether time to recurrence is an independent predictor of survival or merely a surrogate for underlying factors such as adverse pathology. In this study, we assessed the impact of time to recurrence on the subsequent survival following RC for bladder cancer. Methods: We reviewed our IRB-approved RC database for patients who experienced recurrence following RC for bladder cancer between 1983 and 2024. Time to recurrence was categorized into four quartiles. Cox regression analyses were used to assess predictors of subsequent survival post-recurrence. Results: 935 patients who experienced recurrence post-cystectomy were included in the analysis. Time to recurrence was categorized into four quartiles: ≤ 4 months (Q1), 4-9 months (Q2), 9-20 months (Q3), and > 20 months (Q4). Patients who recurred later were more likely to be younger, male, and present with organ confined disease. Adverse pathological features, including variant histology and lymphovascular invasion, were less commonly observed among those with later recurrence. Univariate cox regression showed that recurrence between 9-20 months or later than 20 months was associated with better survival compared to recurrence within the first 4 months (HR: 0.77 and HR: 0.67, respectively.) Multivariate analysis demonstrated that, after adjusting for confounders including age at recurrence, pathological staging, and lymphovascular invasion, patients with recurrence after 20 months had a 19% decreased risk of death following recurrence (HR: 0.81, 95% CI: 0.66 – 0.99, p = 0.049). Conclusions: Recurrence after 20 months following RC is independently associated with superior survival, even after adjusting for major confounding factors. This may suggest that late recurrences possess distinct biological characteristics that influence their behavior. Demographics and perioperative characteristics stratified by time to recurrence quartiles. Patient characteristics Q1 (N=234) Q2 (N=243) Q3 (N=231) Q4 (N=175) P value Age at Recurrence (yrs.) 70.0 ± 10.1 68.1 ± 11.3 69.2 ± 10.8 69.9 ± 10.0 0.203 Sex (Male) 164 (70.1%) 176 (72.4%) 186 (80.5%) 175 (77.1%) 0.043 Neoadjuvant chemotherapy 65 (27.8%) 83 (34.2%) 45 (19.5%) 33 (14.5%) <0.001 Pathological Staging Organ-confined (≤ (y)pT2N0) 22 (9.4%) 54 (22.2%) 66 (28.6%) 104 (45.8%) <0.001 Extravesical (> (y)pT2N0) 66 (28.2%) 73 (30.0%) 43 (18.6%) 41 (18.1%) Lymph-node positive (pN+) 146 (62.4%) 116 (47.7%) 122 (52.8%) 82 (36.1%) Total Positive Lymph Nodes, median (IQR) 2.0 (0.0, 9.0) 0.0 (0.0, 3.0) 1.0 (0.0, 5.0) 0.0 (0.0, 1.0) <0.001 Histologic subtype (Variant histology) 82 (35.0%) 79 (32.5%) 57 (24.7%) 47 (20.7%) 0.002 Presence of lymphovascular invasion 155 (66.2%) 118 (48.6%) 123 (53.3%) 99 (43.6%) <0.001