A Population-Based Assessment of Local Therapy for De Novo Oligometastatic Urothelial Carcinoma of the Bladder

Abstract

<h3>Purpose/Objective(s)</h3> The management of oligometastatic bladder cancer is unclear. Prior data implies site and burden of metastases vary in prognoses. This study evaluates treatment trends, examines the effect of local therapy on survival, and explores if immunotherapy influences outcomes in patients with de novo oligometastatic bladder urothelial carcinoma. <h3>Materials/Methods</h3> A retrospective analysis of the National Cancer Database (NCDB) was performed by identifying bladder urothelial carcinoma patients with metastases (M1 stage) at diagnosis between 2004 and 2016. Local therapy was classified as high-intensity [HT] (i.e., definitive surgery, radiation [RT] dose > 50 Gy) or low-intensity [LT] (i.e., non-definitive surgery, RT dose < 50 Gy). Chemotherapy (CT) was classified relative to local therapy as cytoreductive or consolidative. Metastatic burden was determined by cumulative summation of number of metastatic sites involved (i.e., brain, bone, liver, and lung). Clinicopathologic characteristics were analyzed by descriptive statistics and a Chi-square test was used for comparisons across metastatic disease burden. Univariable (UVA) Kaplan-Meier analysis was used to estimate overall survival (OS). A multivariable Cox regression model was used to estimate the risk of poor survival by treatment modality. Also, effect-modification was assessed between treatment intensity and systemic therapy type. <h3>Results</h3> A total of 11,998 patients were identified with metastases at diagnosis. The median age was 72 years (IQR: 62-80), the majority were Caucasian (87.5%) and male (71.2%). The median follow-up time was 5.5 months (IQR: 2.27-12.62). Patients were treated with definitive surgery (n = 849; 7.1%), perioperative CT (n = 404; 3.4%), CT alone (n = 4,030; 33.6%), RT alone (n = 1,026; 8.6%), CTRT (n = 1,451; 12.1%) or no therapy (n = 3,343; 33.6%). HT and LT were given to 1,083 (9%) and 8,867 (73.9%) patients, respectively. Differences were noted in median OS among treatment groups, such as perioperative CT (OS 16.4 months) or CTRT (> 50 Gy) (12.4 months); <i>P</i> < 0.001. HT vs LT improved OS (10.2 vs 5.7 months; <i>P</i> < 0.001). Consolidative vs cytoreductive therapy improved OS (<i>P</i> < 0.001), but within HT/LT comparisons, sequencing was only significant in HT (16.5 vs 12.7 months; <i>P</i> < 0.001), which was confirmed by interaction analysis. A total of 4,315 patients had site-specific data: bone (n = 1,486), lung (n = 1,175), liver (n = 525) and brain (n = 70). Differences were noted in OS between sites (<i>P</i> < 0.001) with liver and brain having the worst OS. Metastatic burden was also associated with OS (<i>P</i> < 0.001). Immunotherapy was given to 174 patients, which improved OS (12.2 vs 5.7 months; <i>P</i> < 0.001), was independent of metastatic burden, and was associated with 41% reduction in mortality (HR: 0.59, 95% CI: 0.49-0.69; <i>P</i> < 0.001). <h3>Conclusion</h3> This population-based study suggests local aggressive therapy, either surgical or RT (> 50 Gy) is associated with an improved OS benefit in patients with de novo metastatic bladder urothelial carcinoma.