OBJECTIVE: Despite population-level increases in use of direct oral anticoagulants (DOACs) over the last decade, less is known about their use among individuals with cancer and non-valvular atrial fibrillation (NVAF). We quantified DOAC initiation and switching among such individuals.
METHODS: We used Surveillance, Epidemiology, and End Results (SEER)-Medicare data from 2009 to 2016 and included patients aged ≥66 years diagnosed with cancer (breast, bladder, colorectal, esophagus, lung, ovary, kidney, pancreas, prostate, stomach or uterus). We limited the cohort to continuously enrolled beneficiaries with newly diagnosed NVAF following cancer diagnosis (from 2010 to 2016) who did not have any oral anticoagulant use during the 12-months prior to their NVAF diagnosis. We evaluated the initiation of warfarin or DOACs within 3 months of the incident NVAF diagnosis. Among warfarin users, we determined switching rates to DOACs and among DOACs users, we determined switching rates to warfarin. Descriptive statistics were used for all the analyses.
RESULTS: Of 1,028,784 Medicare beneficiaries with cancer, 158,744 (15.4%) were diagnosed with atrial fibrillation. After applying all inclusion criteria, the final study cohort included 19,513 cancer patients diagnosed with incident NVAF. Lung (20.2%), Breast (19.2%), prostate (19.2%) and colorectal (16.0%) cancer accounted for over three-fourths of the cohort. Overall, 7,604 patients (39%) initiated oral anticoagulants. Among these patients, 4,223/7,604 (55.5%) and 3,381/7,604 (44.5%) initiated DOACs and warfarin, respectively. Patients who initiated DOACs had lower Charlson comorbidity score (1.0 ± 1.4 vs. 1.3 ± 1.5) and CHA2DS2-VASc score (4.4 ± 1.7 vs. 4.8 ± 1.7) compared to warfarin users. Initiation of DOACs varied by cancer type; for example, 13.6% of patients with stomach cancer initiated DOACs, whereas 29.8% of patients with prostate cancer initiated DOACs. Among patients who initiated any oral anticoagulants, the use of DOACs increased from 21.6% in 2010 to 76.9% in 2016, whereas warfarin use declined from 78.4% to 23.1%. The switching rate among warfarin initiators to DOACs (623/3381, 18.4%) was higher than the switching rate among DOACs initiators to warfarin (281/4223, 6.7%).
CONCLUSION: Among cancer patients with newly diagnosed NVAF, the use of DOACs has markedly increased from 2010 to 2016, with a corresponding decline in the use of warfarin. Moreover, cancer patients are more likely to switch from warfarin to DOACs. Our findings highlight the dynamic life cycle of these products and the value of real-world assessments of their comparative safety and effectiveness among Medicare beneficiaries.
