AASLD Abstracts Forrest Plots Forrest Plot and Funnel Plot Mo1409 SURVIVAL OF COMBINED HEPATOCELLULAR AND CHOLANGIOCARCINOMA DIFFERS BY RACE: A SEER DATABASE ANALYSIS Tomoki Sempokuya, Linda L. Wong Introduction Hepatobiliary cancer is one of a few cancers in the U.S. that increasing incidence and mortality. Hepatocellular carcinoma is the most common form but combined hepatocellular and cholangiocarcinoma (HCC/CC) occurs rarely and is generally believed to have poor prognosis. Prognosis on this variant is unclear as it has only been studied in small series. We thus aimed to explore if there were racial differences in cancer Methods This is a retrospective study of The Surveillance, Epidemiology, and End Results (SEER) database obtained from SEER*Stat 8.3.6 software. We utilized previously defined histology code 8180 for the International Classification of Disease for Oncology, 3rd edition, to identify patients with combined HCC/CC from 2004 to 2015. We obtained demographic information, state of residence, year of diagnosis, stage, treatment, tumor characteristics, and survival information. We analyzed this data stratified by race: Whites, Blacks, Hispanics, Asians/ Pacific Islanders (PI). Chi-square test was used to compare categorical variables. The Kruskal- Wallis test was used to compare age, which did not show normal distribution by the Kolmogorov–Smirnov test. Kaplan-Meier Survival analysis was used to compare survival by race. Results After exclusion of seven patients for unclear race, 490 patients were identified. Overall, the mean age was 62.4 (Standard deviation (S.D.): 11.3) and 149 (30.4%) were female. Median survival was 7 months (95% confidence interval (CI): 5-9). Racial distribution was as follows: 276 (56.3%) Whites, 84 (17.1%) Asians/PIs, 77 (15.7%) Hispanics, and 53 (10.8%) Blacks. Whites had the highest mean age (63.5 ± 10.6 years), and Blacks had the lowest mean age (59.8± 9.8 years), but this was not significant. There was no difference by race in status on distant and regional lymph nodes positivity, male to female ratio, size of the tumor, distribution of stages by American Joint Committee on Cancer or SEER staging, and treatment. Marital status showed a significant difference among race (P<0.001). California had a high proportion of Asians and Hispanics, and all patients in Hawaii were Asians. The difference in racial distribution was statistically significant (P<0.001). Hispanics and Asians had a higher median survival of 12 months (CI: 4-18) and 11 months (CI: 6-19), respectively (P=0.02). Conclusion Although age, stage, tumor characteristics and treatment did not differ by race, there were distinct differences in survival by racial groups with Hispanics and Asians having significantly higher median survival than Whites or Blacks. Future studies should explore possible differences in comorbidities or tumor biology to explain these racial differ- ences. S-1398 AASLD Abstracts Kaplan–Meier Survival Curve by Race. Mo1410 CURATIVE-INTENT CHEMORADIOTHERAPY FOR PATIENTS WITH LOCALLY ADVANCED, UNRESECTABLE, EXTRA-HEPATIC CHOLANGIOCARCINOMA Krishan Jethwa, Shilpa Sannapaneni, Trey Mullikin, Phanindra Antharam, William S. Harmsen, Molly M. Petersen, Amit Mahipal, Thorvardur R. Halfdanarson, Kenneth Merrell, Michelle Neben-Wittich, Michael G. Haddock, Christopher Hallemeier Background: Unresectable extra-hepatic cholangiocarcinoma (ECCA) is a highly fatal disease with limited curative treatment options. The purpose of this study was to evaluate the efficacy and adverse events (AEs) associated with curative-intent radiotherapy (RT) for patients with unresectable ECCA and identify variables associated with overall survival (OS) and progression-free survival (PFS). Methods: This was a retrospective cohort study of patients diagnosed with localized, unresectable ECCA between 1998 and 2018. Patients were included in the analysis if they were at least 18 years of age, had biopsy or cytology proven ECCA, had no evidence of distant metastatic disease, were deemed ineligible for any form of curative surgical resection including radical resection or radical resection with liver transplantation, and received curative-intent RT. Patients received routine oncologic surveillance at 3 to 6-month intervals following RT with history and physical examination, laboratory evaluation, and diagnostic imaging. AEs were assessed and attributed per common terminology criteria for AEs version 4.0. The Kaplan-Meier method was used to estimate OS and PFS. The cumulative incidence of locoregional recurrence (LRR) and late AEs were reported using the competing risk model with death as a competing risk. Baseline patient, tumor, and treatment characteristics associated with OS and PFS were assessed using a univariate Cox proportional hazards regression model. Biologically effective dose (BED) was calculated using the linear-quadratic model assuming an a/b of 10 Gy. Results: A total of 48 patients were included for analysis. Table 1 demonstrates baseline patient, tumor, and treatment characteristics. The median follow-up duration was 13 months (IQR 6-29 m). The 2-year OS, PFS, and LRR were 33% [95% confidence interval (CI): 22-50%], 21% (95% CI: 12-36%), and 31% (95% CI: 20-48%), respectively. On UVA, BED > 59.5 Gy (equivalent to the median dose of 50.4 Gy in 28 fractions) was associated with improved OS [hazard ratio (HR) 0.41, 95% CI: 0.18-0.92, p=0.03] and PFS (HR 0.37, 95% CI: 0.16-0.84, p= 0.02), where-as increased primary tumor size (per 1 cm) was associated with worsened PFS (HR 1.29, 95% CI: 1.02-1.63, p=0.04). Figure 1 demonstrates the Kaplan-Meier OS and PFS survival estimates for the cohort stratified by BED > 59.5 Gy vs 59.5 Gy. Treatment- related grade 3 or higher acute and late gastrointestinal AEs occurred in 13% and 17% of patients, respectively. Conclusions: Curative-intent RT is associated with long term survivor- ship in a subset of patients with unresectable ECCA. Further exploration of the role of RT as part of a multi-modality curative treatment strategy for this cohort is warranted.