MSPP05 Presentation Time: 4:40 PM

Purpose Unresectable cholangiocarcinoma is managed with any combination of systemic therapy and local therapies such as radiotherapy (RT). The purpose of this study is to use a large national database to assess the utilization rate and prognostic impact of brachytherapy (BT) in this setting. Materials and Methods The National Cancer Database (NCDB) was queried (2004-2017) for patients with unresected cholangiocarcinoma. Pearson chi-square testing was used to compare categorical frequencies between patients who received external beam RT without BT (EBRT alone) and patients who received BT. Propensity score matching (PSM) was performed to create a 1:1 matched cohort of patients who received EBRT alone v. BT. Kaplan-Meier analysis was used to evaluate overall survival (OS). Univariate (UVA) and multivariate (MVA) analyses were conducted using Cox proportional hazard models to determine which demographic and clinical factors were prognostic for OS. Results A total of 40,187 patients with median age 69 (IQR 60- 78) were included in this study. A total of 15,089 (37.5%) were extrahepatic cholangiocarcinoma and 25,098 (62.5%) were intrahepatic cholangiocarcinoma. A total of 19,397 (49.8%) received chemotherapy (CTX) while 19,589 (50.2%) did not. A total of 35,023 (87.2%) received no RT, 4,234 (10.5%) received EBRT alone, and 930 (2.3%) received BT. EBRT doses ranged from palliative doses of 20 Gy to definitive doses of 60 Gy (median 45Gy, IQR 30- 50.4). BT doses were not available in >90% of cases. Compared to EBRT alone, patients who received BT were more likely diagnosed in 2013 or later (p< 0.001), more likely white race (p= 0.049), more likely treated at an academic center (p < 0.001), more likely insured (p< 0.001), more likely to travel farther to treatment center (p< 0.001), more likely to have intrahepatic disease (p< 0.001), more likely to have earlier stage disease (p< 0.001), and less likely to receive chemotherapy (p< 0.001). After PSM, there were no differences in demographic or clinical factors between patients who received EBRT alone v. BT. After PSM, the median OS was 9.2 months for EBRT alone and 14.7 months for BT (p< 0.001). On UVA after PSM, BT (HR 0.637, p< 0.001; 95% CI 0.560- 0.725), CTX (HR 0.844, p=0.11; 95% CI 0.741- 0.962), female sex (HR 0.841, p=0.008; 95% CI 0.739- 0.956), academic facility (HR 0.804, p=0.001; 95% CI 0.707- 0.914), and further distance from treatment center (HR 0.804, p=0.008; 95% CI 0.686- 0.944) were positive prognostic factor while higher stage (HR 1.814, p< 0.001; 95% CI 1.591- 2.067) was a negative prognostic factor. All prognostic factors remained significant on MVA. Conclusions The utilization rate of BT has increased and certain demographic and clinical factors are associated with its use. BT has a positive prognostic impact relative to EBRT alone and a future prospective randomized trial should be performed to confirm this.