425 Background: Data guiding selection of optimal nonsurgical therapies for management of localized hepatocellular carcinoma (HCC) are lacking. Because there are limited prospective comparative data for these treatment modalities, we aimed to compare the effectiveness of chemotherapy (systemic or arterial directed) versus radiofrequency ablation (RFA) versus radioembolization (RE) in nonsurgically managed patients with stage I and II HCC using the National Cancer Database. Methods: We identified patients who received chemotherapy, RFA or RE for nonsurgically managed Stage I (T1N0M0) and II (T2N0M0) HCC between 2008 and 2013. Patients excluded included those who received upfront lobectomy or resection or extended lobectomy or hepatectomy or transplant, and if they had Ishak fibrosis score of 5-6 or total bilirubin > 3 mg/dl or international normalized ratio (INR) of > 3. Overall survival (OS) was compared between treatment groups using propensity score matched (1:1:1) and weighted analyses. We also verified balance of all available confounders, and assessed sensitivity to unobserved confounding. Results: Overall 2622 (73.2%), 837 (23.3%) and 125 (3.5%) patients with nonsurgically managed stage I and II HCC received chemotherapy, RFA and RE, respectively. RE treated patients tend to be older, live in the areas with a higher median income and higher percentage of high school- educated residents, and have stage II disease. The propensity matched cohort included 303 patients with baseline characteristics well balance between all arms. After propensity matching, 5-year OS was 17.4% (95% CI, 6.6% to 32.3%) in the chemotherapy group, 27.2% (95% CI, 14.2% to 42%) in the RFA group and 36.6% (95% CI, 23.1% to 50.1%) in RE group (p = 0.48). Conclusions: To our knowledge, this is the first study comparing various different nonsurgical therapies for localized HCC. All chemotherapy, RFA and RE are effective treatment options for nonsurgically managed patients with stage I and II HCC. Although these data are retrospective, RE appears to be a reasonable first-line treatment of nonsurgically managed stage I and II HCC in carefully selected patients.