Background: Patients with cancer are at a higher risk for developing venous thromboembolisms (VTEs), leading to further complications. The National Comprehensive Cancer Network (NCCN) guidelines currently recommend using low-molecular-weight heparin (LMWH) or direct oral anticoagulants (DOACs) for the treatment of VTEs in oncology patients. Other trials compare DOACs with LMWH to determine the risk of bleeding. The Select-D trial found that rivaroxaban had a higher risk of bleeding in patients with gastrointestinal (GI) cancer as compared to LMWH. However, the Caravaggio trial found that apixaban had no increase of bleeding for patients with GI cancer. Due to the inherent risk of major bleeding shown in the subgroup analysis of the trials above, the guidelines recommend LMWH as the preferred agent for patients with GI cancers. At Winship Cancer Institute, DOACs are used for the treatment of cancer-associated VTE for patients with GI malignancies. The purpose of this study was to assess real-world safety outcomes of patients with GI cancer who were treated with apixaban or rivaroxaban for VTE. Methods: This study is a single-center, retrospective chart review of patients 18 years or older with GI cancer receiving treatment for VTE with either apixaban or rivaroxaban. The study looks to review bleeding outcomes in patients who started their VTE treatment from January 1, 2017 – December 31, 2021. The primary outcome of this study was to evaluate the rates of major bleeding between apixaban and rivaroxaban. Major bleeding was defined as a decrease in hemoglobin of ≥ 2g/dL over 24 hours, transfusion of ≥2 units of red blood cells, bleeding that requires surgical intervention, or bleeding at critical sites. Secondary outcomes included clinically relevant non-major bleeding (CRNMB) defined by bleeding compromising hemodynamics, hematemesis, hemoptysis, rectal bleeding, or hematuria, VTE recurrence, and death from bleeding. Data collected to measure these outcomes included patient characteristics such as renal function, weight, primary cancer site, medication dosing, and duration and indication of treatment. Results: A total of 300 patients qualified for inclusion of this study. Of these, 89 were in the apixaban group and 211 in the rivaroxaban group. Major bleeding occurred in 3 patients in the apixaban group (3.4%) and 18 in the rivaroxaban group (8.5%). CRNMB occurred in 2 patients in the apixaban group (2.2%) and 16 patients in the rivaroxaban group (7.6%). VTE recurrence occurred in 4 patients in the apixaban group (4.5%) and 9 patients in the rivaroxaban group (4.3%). Conclusions: This is the only study to date to evaluate DOACs head to head and their bleeding risk in cancer patients. Apixaban was associated with lower rates of major bleeds and CRNMB, but relatively similar rates of VTE recurrence. The results of this study warrant consideration of using DOACs for patients with GI cancers in the upfront setting.