Bleeding risk and major adverse events in patients with cancer on oral anticoagulation therapy.

BACKGROUND The efficacy of oral anticoagulation therapy (OAT) has not been revealed in atrial fibrillation (AF) patients with newly diagnosed cancers. This study evaluated the thromboembolic and bleeding events in AF patients with malignancies according to OAT. METHODS AND RESULTS In 2168 consecutive non-valvular AF patients with newly diagnosed malignancies, we analyzed the composite endpoints including major adverse cardiac events (MACEs) and major bleeding. Based on a propensity score matching, two groups with 690 matched pairs were created. Patient baseline characteristics were comparable between the matched groups. During a follow-up period of 3.9 ± 2.8 years, 72 (10%) and 65 (9%) patients had MACEs in the propensity score-matched OAT + and OAT − groups, respectively (p = 0.461). There was no significant difference in the major bleeding (10% vs. 8%, p = 0.300) and composite endpoints (18% vs. 16%, p = 0.181) between OAT + and OAT − patients. During the first year after the cancer diagnosis, 66 (48%) MACEs, 52 (41%) major bleedings, and 116 (49%) composite end points of all events occurred. The optimal international normalized ratio (2.0 to 3.0) level was achieved in only 85 (12%) patients. However, 1 year after cancer diagnosis, OAT + patients with the target therapeutic range of ≥ 60% demonstrated better cumulative survival free of composite end point than OAT − patients (p = 0.026). CONCLUSION During the first year after the cancer diagnosis, OAT did not improve the composite end point because of poor INR control caused by cancer treatment. However, after 1 year after diagnosis of cancer, optimal anticoagulation significantly reduced the composite end point.