Background: Atrial fibrillation (AF) is associated with an increased risk of heart failure, death and thromboembolism. AF is prevalent in patients with cancer. Although current guidelines suggest the application of oral anticoagulants (OACs) for thromboembolic event prevention in high-risk AF patients, owing to the high thromboembolic and bleeding risks of active-cancer patients, there is no consensus on the use of OACs in such a population. Therefore, we conducted this retrospective cohort study to investigate the applicability of the CHA{}_2DS{}_2-VASc score and to evaluate the efficacy and safety outcomes of OAC therapy in active-cancer patients with AF. Methods: This retrospective cohort study enrolled patients diagnosed with cancer at National Cheng Kung University Hospital between November 2012 and August 2019. The primary outcomes included all-cause mortality, thromboembolic events (stroke/transient ischemic attack and systemic emboli), acute myocardial infarction (AMI), hospitalization for HF and major bleeding events. Results: We enrolled 2429 patients with active cancer. Among these patients, 1060 patients (43.6%) had AF. After 1:2 propensity score matching, 690 cancer patients with AF were enrolled for the final analysis, grouped as follows: 225 patients taking OACs and 465 patients without OAC treatment. The OAC-treated group had lower all-cause mortality than the patients without OAC treatment (all-cause mortality rate in OAC treatment vs. non-OAC treatment: 24.4% vs. 37.4%, hazard ratio 0.58 [95% confidence interval (CI) 0.43–0.78], p 0.001). However, there was no difference in thromboembolic events, myocardial infarction or heart failure hospitalization between the OAC-treated and non-OAC-treated groups. Importantly, the risk of major bleeding composition (i.e., major gastrointestinal bleeding and intracranial hemorrhage) was similar between these two groups. Moreover, the CHA{}_2DS{}_2-VASc score could not predict thromboembolic events in the enrolled active-cancer patients with AF (OR 1.23, 95% CI 0.98–1.56). Conclusions: OAC treatment may significantly reduce the risk of death, without safety concerns, in active-cancer patients with AF. OAC treatment may not prevent thromboembolic events in patients with active cancer and AF. However, we found that OAC treatment is associated with improved prognosis without increasing the risks of major bleeding, despite several limitations in this study. Further studies are required to determine the optimal use of anticoagulation therapy in this high-risk population.