Patients with atrial fibrillation and malignancy have increased risks of thromboembolism and bleeding. Evidence comparing apixaban and warfarin in this group remains limited. We aimed to compare effectiveness and safety of apixaban versus warfarin in patients with atrial fibrillation and active malignancy using real-world data from a large multinational cohort. This retrospective cohort study used the TriNetX Global Collaborative Network, de-identified records from 146 healthcare organizations between December 1, 2012, and May 1, 2025. Atrial fibrillation patients with malignancy receiving apixaban or warfarin were matched 1:1 using propensity scores across 74 clinical variables. Outcomes were assessed at 3 months, 6 months, 1 year, and 5 years. Primary endpoints included all-cause mortality, stroke, pulmonary embolism, deep vein thrombosis, gastrointestinal bleeding and intracranial hemorrhage. In this 12.5-year period, 41,764 matched pairs of patients were analyzed. Compared to the warfarin cohort, the apixaban cohort demonstrated lower all-cause mortality at 3 months (OR: 1.05, 95% CI: 1.00–1.10), 6 months (OR: 1.05, 95% CI: 1.01–1.09), 1 year (OR: 1.06, 95% CI: 1.03–1.10), and 5 years (OR: 1.17, 95% CI: 1.13–1.20; all p <0.05). Stroke rates were comparable between groups, while pulmonary embolism, deep vein thrombosis, gastrointestinal bleeding and intracranial hemorrhage were noted less frequent with apixaban. Kaplan–Meier analyses showed early and sustained differences in survival and bleeding outcomes. In conclusion, in atrial fibrillation patients with cancer, apixaban was associated with lower mortality and major bleeding without increasing stroke risk compared to warfarin.
Graphical Abstract
Central Illustration: Efficacy and safety of apixaban versus warfarin in atrial fibrillation patients with malignancy
Atrial fibrillation is the most common cardiac arrhythmia, affecting an estimated 33 million individuals worldwide and contributing significantly to morbidity and mortality from stroke and heart failure. With the rising global burden of cancer and advances in oncologic therapies, an increasing number of patients frequently suffer from malignancy and cardiovascular comorbidities. ,, The coexistence of atrial fibrillation and cancer presents unique therapeutic challenges, particularly regarding anticoagulation. , While vitamin K antagonists, such as warfarin, were long considered standard of care for stroke prevention in atrial fibrillation, direct oral anticoagulants (DOAC), especially apixaban, have become preferred in the general atrial fibrillation population due to their ease of use and favorable safety profile. , However, pivotal randomized trials that established the efficacy of DOAC over warfarin in atrial fibrillation, such as ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) and RE-LY (Randomized Evaluation of Long Term Anticoagulant Therapy With Dabigatran Etexilate), enrolled very few patients with active cancer, typically <3%, thereby limiting the generalizability of their findings to cardio-oncology populations. , Moreover, warfarin management in cancer patients is often complicated by fluctuating nutritional status, hepatic dysfunction, and numerous drug-drug interactions. Although emerging observational studies suggest that apixaban may offer a favorable risk-benefit profile in patients with cancer-associated atrial fibrillation, high-quality comparative data remains limited. , In this context, we conducted a large retrospective cohort study using the TriNetX Global Collaborative Network to compare the effectiveness and safety of apixaban versus warfarin in adults with atrial fibrillation and malignancy.
Methods
Data Source
This study utilized the TriNetX Global Collaborative Network, a federated health research platform comprising real-time, deidentified electronic medical records (EMRs) from 146 healthcare organizations (HCOs) across multiple countries. The database includes structured information on demographics, diagnoses, procedures, medications, and laboratory results. All data are deidentified and compliant with the Section §164.514 of the Health Insurance Portability and Accountability Act (HIPAA) and the U.S. Department of Health and Human Services’ standards for anonymization. No Institutional Review Board approval was required for this analysis of deidentified data.
Study Design and Population
This was a retrospective cohort study between December 1, 2012, and May 1, 2025, comparing clinical outcomes in adult patients (≥18 years) with atrial fibrillation and malignancy who were treated with either warfarin or apixaban. Malignancy was defined by International Classification of Diseases 10th edition codes for solid and hematologic cancers occurring within 5 years prior or concurrent with atrial fibrillation diagnosis. The earliest date meeting both atrial fibrillation and malignancy criteria was designated as the index event ( Supplementary Table S1 and S2 ). Patients were followed from the index date until the end of the defined time windows: 3 months, 6 months, 1 year, and 5 years. Patients with outcomes prior to the index event were excluded from analysis for that specific outcome.
Outcomes and Definitions
Clinical outcomes were identified using International Classification of Diseases 10th edition codes and included efficacy outcomes—all-cause mortality, thrombotic stroke, embolic stroke, pulmonary embolism (PE) and deep vein thrombosis (DVT), and safety outcomes—gastrointestinal bleeding, traumatic intracranial hemorrhage (ICH), and nontraumatic ICH ( Supplementary Table S3 ).
Statistical Analysis
To reduce baseline differences and to minimize confounding, 1:1 propensity score matching (PSM) was performed using logistic regression based on 74 baseline variables across 4 domains—demographics (age at index, current age, sex, race), comorbidities (hypertension, diabetes, chronic kidney disease, ischemic heart disease, cardiomyopathy, heart failure, obesity, alcohol/nicotine/cannabis use, respiratory diseases, sleep disorders, malnutrition, body mass index, and detailed malignancy subtypes), medications (anticoagulants, antiplatelets, antihypertensives, and antihyperglycemic agents), and laboratory values and vitals (hemoglobin A1c, troponin, B-type natriuretic peptide, lipid profile, natriuretic peptides, blood pressure, heart rate, and left ventricular ejection fraction). The TriNetX platform utilizes “greedy nearest-neighbor matching” for PSM with a caliper of 0.1 pooled standard deviation of the linear propensity scores to control for differences in the two cohorts. Standardized mean differences were used to assess balance for the covariates, with <0.1 considered as well-balanced ( Table 1 , Supplementary Table S4 and S5 ). The primary analysis was a measure of association comparing event rates between groups at 3 months, 6 months, 1 year, and 5 years postindex. Outcomes were expressed as event proportions and compared using odds ratios (OR) with 95% confidence intervals (CI). Statistical significance was defined as a two-sided p-value <0.05. Survival analyses were conducted using Kaplan–Meier estimates with censoring at last known follow-up. All analyses were performed using the in-built TriNetX Analytics Platform (TriNetX, Inc.; Cambridge, MA, USA), version as of June 11, 2025.
Table 1
Baseline characteristics of atrial fibrillation patients with malignancy on warfarin versus apixaban
| Characteristic | Before propensity score matching | After propensity score matching | ||||||
|---|---|---|---|---|---|---|---|---|
| Warfarin ( n = 66,321) | Apixaban ( n = 144,485) | p | Standard difference | Warfarin ( n = 41,764) | Apixaban ( n = 41,764) | p | Standard difference | |
| Demographics | ||||||||
| Current Age (y) | 81.5 ± 8.8 | 78.6 ± 9.1 | <0.01 | 0.329 | 80.7 ± 9.3 | 80.7 ± 8.2 | 0.55 | 0.004 |
| Age at Index (y) | 75.3 ± 9.6 | 75.1 ± 9.4 | <0.01 | 0.024 | 75.7 ± 10.1 | 75.6 ± 8.6 | 0.75 | 0.002 |
| White race | 54,427 (79.2%) | 113,023 (76.3%) | <0.01 | 0.071 | 32,693 (78.3%) | 32,862 (78.7%) | 0.15 | 0.01 |
| African American race | 6,121 (8.9%) | 13,729 (9.3%) | <0.01 | 0.012 | 3,731 (8.9) | 3,774 (9.0) | 0.60 | 0.004 |
| Asian race | 1,493 (2.2%) | 4,144 (2.8%) | <0.01 | 0.04 | 926 (2.2%) | 884 (2.1%) | 0.32 | 0.007 |
| Female sex | 24,458 (35.6%) | 54,661 (36.9%) | <0.01 | 0.027 | 15,094 (36.1%) | 15,189 (36.4%) | 0.5 | 0.005 |
| Male sex | 41,917 (61.0%) | 87,766 (59.2%) | <0.01 | 0.037 | 24,943 (59.7%) | 24,912 (59.6%) | 0.88 | 0.002 |
| Comorbidities | ||||||||
| Thyroid disorders | 17,257 (25.1%) | 39,088 (26.4%) | <0.01 | 0.029 | 10,749 (25.7%) | 10,894 (26.1%) | 0.25 | 0.008 |
| Diabetes mellitus | 25,030 (36.4%) | 52,686 (35.6%) | <0.01 | 0.018 | 15,097 (36.1%) | 15,138 (36.2%) | 0.77 | 0.002 |
| Overweight/ obesity | 17,827 (26.0%) | 44,019 (29.7%) | <0.01 | 0.084 | 11,402 (27.3%) | 11,595 (27.8%) | 0.14 | 0.01 |
| Hypertension | 51,970 (75.7%) | 113,099 (76.3%) | 0.01 | 0.016 | 31,637 (75.8%) | 31,633 (75.7%) | 0.97 | <0.001 |
| Ischemic heart disease | 32,171 (46.8%) | 65,929 (44.5%) | <0.01 | 0.047 | 19,438 (46.5%) | 19,712 (47.2%) | 0.06 | 0.013 |
| Cardiomyopathy | 11,970 (17.4%) | 21,517 (14.5%) | <0.01 | 0.079 | 6,859 (16.4%) | 7,114 (17.0%) | 0.02 | 0.016 |
| Heart failure | 29,636 (43.1%) | 57,048 (38.5%) | <0.01 | 0.095 | 17,681 (42.3%) | 18,055 (43.2%) | <0.01 | 0.018 |
| Cerebral infarction | 8,659 (12.6%) | 18,870 (12.7%) | 0.4 | 0.004 | 5,424 (13.0%) | 5,547 (13.3%) | 0.21 | 0.009 |
| Gastroesophageal reflux disease | 21,836 (31.8%) | 53,454 (36.1%) | <0.01 | 0.091 | 13,973 (33.5%) | 14,288 (34.2%) | 0.02 | 0.016 |
| Acute kidney injury | 18,329 (26.7%) | 42,680 (28.8%) | <0.01 | 0.047 | 11,760 (28.2%) | 12,223 (29.3%) | <0.01 | 0.025 |
| Chronic kidney disease | 22,571 (32.9%) | 46,311 (31.3%) | <0.01 | 0.034 | 13,852 (33.2%) | 14,076 (33.7%) | 0.1 | 0.011 |
| Alcohol disorders | 3,140 (4.6%) | 8,484 (5.7%) | <0.01 | 0.052 | 1,971 (4.7%) | 2,087 (5.0%) | 0.06 | 0.013 |
| Cannabis disorders | 609 (0.9%) | 2,056 (1.4%) | <0.01 | 0.047 | 438 (1.0%) | 463 (1.1%) | 0.40 | 0.006 |
| Nicotine dependence | 9,033 (13.2%) | 23,892 (16.1%) | <0.01 | 0.084 | 5,735 (13.7%) | 5,794 (13.9%) | 0.55 | 0.004 |
| Lower respiratory diseases | 24,008 (35.0%) | 53,801 (36.3%) | <0.01 | 0.028 | 14,704 (35.2%) | 14,743 (35.3%) | 0.78 | 0.002 |
| Osteoarthritis | 23,094 (33.6%) | 53,667 (36.2%) | <0.01 | 0.054 | 14,535 (34.8%) | 14,859 (35.6%) | 0.02 | 0.016 |
| Spondylopathies | 14,118 (20.6%) | 35,284 (23.8%) | <0.01 | 0.078 | 9,078 (21.7%) | 9,273 (22.2%) | 0.10 | 0.011 |
| Malnutrition | 6,003 (8.7%) | 17,944 (12.1%) | <0.01 | 0.11 | 4,233 (10.1%) | 4,306 (10.3%) | 0.40 | 0.006 |
| Acute respiratory failure | 5,000 (7.3%) | 7,726 (5.2%) | <0.01 | 0.085 | 2,732 (6.5%) | 2,729 (6.5%) | 0.97 | <0.001 |
| Sleep disorders | 19,768 (28.8%) | 45,223 (30.5%) | <0.01 | 0.038 | 12,240 (29.3%) | 12,501 (29.9%) | 0.05 | 0.014 |
| Medications | ||||||||
| Beta blockers | 51,118 (74.4%) | 111,651 (75.4%) | <0.01 | 0.021 | 30,365 (72.7%) | 30,480 (73.0%) | 0.37 | 0.006 |
| Calcium channel blockers | 33,191 (48.3%) | 77,048 (52.0%) | <0.01 | 0.074 | 20,301 (48.6%) | 20,710 (49.6%) | <0.01 | 0.02 |
| Antianginals | 20,779 (30.3%) | 41,154 (27.8%) | <0.01 | 0.055 | 12,157 (29.1%) | 12,538 (30.0%) | <0.01 | 0.02 |
| Antiarrhythmics | 43,583 (63.5%) | 105,537 (71.2%) | <0.01 | 0.166 | 27,411 (65.6%) | 27,556 (66.0%) | 0.29 | 0.007 |
| Antiepileptics | 43,108 (62.8%) | 95,080 (64.2%) | <0.01 | 0.029 | 25,557 (61.2%) | 25,681 (61.5%) | 0.38 | 0.006 |
| Diuretics | 44,619 (65.0%) | 89,957 (60.7%) | <0.01 | 0.088 | 25,882 (62.0%) | 26,142 (62.6%) | 0.06 | 0.013 |
| ACE (angiotensin converting enzyme) inhibitors | 26,047 (37.9%) | 49,959 (33.7%) | <0.01 | 0.088 | 14,532 (34.8%) | 14,749 (35.3%) | 0.12 | 0.011 |
| Angiotensin inhibitors | 16,661 (24.3%) | 43,714 (29.5%) | <0.01 | 0.119 | 10,337 (24.8%) | 10,518 (25.2%) | 0.15 | 0.01 |
| Direct renin inhibitors | 151 (0.2%) | 224 (0.2%) | <0.01 | 0.016 | 80 (0.2%) | 89 (0.2%) | 0.49 | 0.005 |
| Rivaroxaban | 4,338 (6.3%) | 12,362 (8.3%) | <0.01 | 0.078 | 3,361 (8.0%) | 3,633 (8.7%) | <0.01 | 0.024 |
| Heparin | 31,876 (46.4%) | 76,659 (51.7%) | <0.01 | 0.107 | 20,506 (49.1%) | 20,866 (50.0%) | 0.01 | 0.017 |
| Aspirin | 35,581 (51.8%) | 79,565 (53.7%) | <0.01 | 0.038 | 21,645 (51.8%) | 22,047 (52.8%) | <0.01 | 0.019 |
| Clopidogrel | 9,564 (13.9%) | 23,569 (15.9%) | <0.01 | 0.056 | 6,039 (14.5%) | 6,256 (15.0%) | 0.03 | 0.015 |
| Insulin | 23,559 (34.3%) | 48,401 (32.7%) | <0.01 | 0.035 | 14,274 (34.2%) | 14,393 (34.5%) | 0.39 | 0.006 |
| Oral hypoglycemic agents | 14,452 (21.0%) | 32,397 (21.9%) | <0.01 | 0.02 | 8,424 (20.2%) | 8,517 (20.4%) | 0.42 | 0.006 |
| SGLT2i (Sodium glucose cotransporter-2 inhibitor) | 1,602 (2.3%) | 8,229 (5.6%) | <0.01 | 0.166 | 1,325 (3.2%) | 1,386 (3.3%) | 0.23 | 0.008 |
| Laboratory parameters | ||||||||
| Left Ventricular Ejection Fraction (%) | 54.0 ± 14.8 | 55.5 ± 13.6 | <0.01 | 0.109 | 54.3 ± 14.8 | 54.4 ± 14.5 | 0.72 | 0.007 |
| Cholesterol (mg/dL) | 152.9 ± 45.4 | 152.2 ± 45.0 | 0.01 | 0.015 | 152.4 ± 45.8 | 151.5 ± 45.2 | 0.03 | 0.02 |
| Low density lipoprotein (mg/dL) | 82.0 ± 35.0 | 81.3 ± 35.1 | <0.01 | 0.02 | 81.6 ± 35.4 | 80.8 ± 35.0 | 0.01 | 0.024 |
| High density lipoprotein (mg/dL) | 45.1 ± 17.9 | 45.5 ± 19.7 | <0.01 | 0.02 | 44.9 ± 18.4 | 45.3 ± 19.2 | 0.02 | 0.023 |
| Triglycerides (mg/dL) | 127.4 ± 89.4 | 121.9 ± 91.7 | <0.01 | 0.061 | 126.6 ± 88.4 | 122.8 ± 85.8 | <0.01 | 0.044 |
| Troponin I (ng/mL) | 0.7 ± 6.1 | 0.6 ± 5.7 | 0.18 | 0.013 | 0.7 ± 6.6 | 0.6 ± 5.7 | 0.13 | 0.022 |
| B-type natriuretic peptide (pg/mL) | 961.9 ± 3246.2 | 921.3 ± 3183.2 | 0.15 | 0.013 | 1102.8 ± 3745.0 | 1077.6 ± 3679.7 | 0.61 | 0.007 |
| Hemoglobin A1c (%) | 6.3 ± 1.5 | 6.3 ± 1.5 | 0.31 | 0.007 | 6.3 ± 1.5 | 6.3 ± 1.5 | 0.07 | 0.018 |
| Respiratory rate (breaths/min) | 17.3 ± 3.1 | 17.2 ± 3.0 | 0.43 | 0.005 | 17.3 ± 3.2 | 17.2 ± 2.9 | 0.13 | 0.015 |
| Heart rate (beats/min) | 76.7 ± 17.2 | 77.8 ± 18.6 | <0.01 | 0.062 | 77.2 ± 17.7 | 77.4 ± 18.1 | 0.25 | 0.01 |
| Body weight (lb.) | 190.4 ± 51.3 | 186.8 ± 49.7 | <0.01 | 0.071 | 189.3 ± 51.7 | 188.3 ± 49.2 | 0.03 | 0.019 |
| Body mass index (kg/m 2) | 29.4 ± 7.1 | 28.8 ± 6.9 | <0.01 | 0.087 | 29.3 ± 7.1 | 29.0 ± 6.9 | <0.01 | 0.033 |
| Blood Pressure, Systolic (mmHg) | 123.2 ± 22.9 | 124.0 ± 22.7 | <0.01 | 0.037 | 122.7 ± 23.7 | 124.0 ± 22.3 | <0.01 | 0.055 |
| Blood Pressure, Diastolic (mmHg) | 67.9 ± 13.8 | 69.0 ± 13.5 | <0.01 | 0.081 | 67.5 ± 14.2 | 68.6 ± 13.4 | <0.01 | 0.082 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
