Cardiovascular and Neurological Outcomes in Patients Treated with Edoxaban for Atrial Fibrillation and Characteristics in Patients with Cancer

BACKGROUND: Direct oral anticoagulants (DOACs) outperform warfarin in vascular and bleeding events in atrial fibrillation (AF) patients. Yet, effects of DOACs on congestive heart failure (CHF) and Alzheimer's disease (AD) remain less explored.

Using the Taiwan National Health Insurance Research Database, a nationwide retrospective cohort study was conducted. The study matched 5,683 non-valvular atrial fibrillation (NVAF) edoxaban patients with 11,366 warfarin patients, and 703 NVAF with cancer (NVAF-C) edoxaban patients with 1,406 warfarin patients. Vasular and non-vascular outcomes, with focuses on CHF and AD, were compared between the edoxaban and warfarin users.

Edoxaban significantly lowered adjusted hazrad ratio (aHR) of all-cause mortality, hospitalization for gastrointestinal bleeding, and CHF (0.37, 0.74, and 0.26, respectively, in NVAF; 0.39, 0.67, and 0.31, respectively, in NVAF-C, all p < 0.05), compared to warfarin. Edoxaban was associated with significantly lower aHRs of acute myocardial infarction, peripheral artery disease, venous thromboembolism, pulmonary embolism, and AD (0.71, 0.48, 0.55, 0.20, and 0.66, respectively; all p < 0.05) in NVAF patients versus warfarin. However, edoxaban had higher aHR of hospitalized bleeding (1.19, p = 0.002) than warfarin in NVAF patients, but not in NVAF-C patients.

Edoxaban demonstrated lowered CHF risks in both NVAF and NVAF-C patients, and reduced AD occurrence in NVAF patients versus warfarin. These findings advocate for edoxaban's use in AF cases. (word count: 220)

edoxaban, atrial fibrillation, cancer, Alzheimer‘s disease All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 3 CLINICAL PERSPECTIVE What Is New?  The study reveals that in patients with atrial fibrillation (AF), edoxaban, a direct oral anticoagulant (DOAC), demonstrates significant advantages over warfarin. Notably, edoxaban is associated with a reduced risk of congestive heart failure (CHF) and Alzheimer's disease (AD) when compared to warfarin. Clinical Implications?  These findings have important clinical implications. Edoxaban appears to be a superior anticoagulant choice for AF patients, as it lowers the risk of CHF and AD. This highlights the potential of edoxaban to improve patient outcomes and underscores its relevance for managing AF cases. Nonstandard Abbreviations and Acronyms: AF: Atrial Fibrillation CHF: Congestive Heart Failure AD: Alzheimer's Disease DOACs: Direct Oral Anticoagulants NVAF: Non-Valvular Atrial Fibrillation NHIRD: National Health Insurance Research Database LMWHs: Low-Molecular-Weight Heparins FXa: Factor Xa PARs: Proteinase Activated Receptors RAAS: Renin-Angiotensin-Aldosterone System AMI: Acute Myocardial Infarction PCI: Percutaneous Coronary Intervention CKD: Chronic Kidney Disease VTE: Venous Thromboembolism All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 4 PE: Pulmonary Embolism AKI: Acute Kidney Injury ICD-9-CM: International Classification of Diseases, 9th Revision, Clinical Modification ICD-10-CM: International Classification of Diseases, 10th Revision, Clinical Modification All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 5

Atrial fibrillation (AF) affects over 37 million individuals worldwide. Its prevalence increased by 33% in the last 20 years, and projected to increase by over 60% by 2050.1 AF is responsible for one-sixth of all strokes,2 with a greater impact observed in the East Asian population.3 AF management in cancer patients is especially an important part of cardio-oncology field. Cancer patients face a heightened risk of AF, independent of cancer treatment, and this risk is further amplified by surgery and chemotherapy, underscoring the importance of effective anticoagulation.4 Low-molecular weight heparins (LMWHs) are not approved for thromboprophylaxis in patients with cancer and AF,5 and managing warfarin can be challenging.5 Direct oral anticoagulants (DOACs) are particularly attractive in cancer patients due to their rapid onset/offset action and lower bleeding risk, especially in Asian individuals.5 Sub-analyses of RE-LY, ROCKET AF, ENGAGE AF, and ARISTOTLE trials showed that the incidence of warfarin-associated hemorrhagic stroke was numerically higher in East Asia, underlining the need for alternative treatments.6 A recent nationwide retrospective cohort study from Taiwan National Health Insurance Research Database (NHIRD) revealed that DOACs were associated with a lower risk of major adverse cardiovascular events, major adverse limb events, venous thrombosis, and major bleeding compared with warfarin in patients with AF and history of cancer.7 In addition to the therapeutic challenges in AF patients with cancer, the causal relationship of subclinical thrombosis with cerebral thromboembolism and neurological outcomes remains unclear. Prior research has shown that edoxaban lowers the activation and reduces the expression of proteinase activated receptors (PARs).8 PARs belong to a class of G protein-coupled receptors involved in coagulation, hemostasis, and inflammation, and, recently recognized as important modulators of synaptic efficacy and plasticity.9 The PARs All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 6 pathway also modulates autophagy by facilitating the production of reactive oxygen species (ROS) and beta amyloid Aβ (1-42), the main component of the amyloid plaques found in brains of people with Alzheimer’s disease (AD).10 Inhibition of PAR1 improves cognitive performance and alleviates synaptic plasticity impairments in a rat model of AD.11 With the increased use of edoxaban in Taiwan in recent years, the study comprehensively assesses the vascular, heart failure, bleeding, and neurological outcomes in Taiwanese AF patients treated with edoxaban or warfarin, considering their cancer history. We especially focused on the unexplored impact on congestive heart failure (CHF) and non- vascular neurological outcomes like Alzheimer's disease (AD).

A nationwide analysis was conducted using Taiwan NHIRD, a database of catastrophic illness and medical claims. Taiwan's National Health Insurance system provides medical coverage to all its citizens. The NHIRD contains healthcare data from over 99% of the population. To protect confidentiality, patient identification was encrypted, and authorized researchers were only allowed to link the data. By using scrambled personal identifiers, researchers were able to link files and gather socio-demographic information and longitudinal medical history. Patients with AF (ICD-9-CM 427.31 or ICD-10-CM I48.xx) without valvular heart disease (ICD-9-CM 394.0, 394.2, 396.1, 396.8, 396.9, 746.5 + procedure codes V43.3, 35.02, 35.12, 35.20, 35.22, 35.24, 35.26, 35.28 or ICD-10-CM I05.0, I05.2, I08.0, I08.1, I08.3, I08.8, I08.9, I09.81, I34.2, Q23.2, Q23.8, Q23.9, Z95.2 + procedure codes 02QG, 02RF, 02RG, 02RH, 02RJ) were identified from Jan. 2014 to Dec. 2018. Within the non-valvular AF (NVAF) cohort, patients with co-existent cancer (ICD-9-CM 140-209.3 or ICD-10-CM C00- C97) were identified (NVAF-C). From NVAF and NVAF-C, patients receiving edoxaban or All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 7 warfarin were captured. Patients using these two medications before NVAF diagnosis were excluded. Edoxaban and warfarin cohorts were matched for sex and age (1:2) using propensity score. The study was approved by the Institutional Research Ethics Committee (IRB Number: KMUHIRB-EXEMPT(II)-21090039). Informed consent was waived for secondary data analysis. The data from the National Death Certificate Registry, which includes the death marker and date, was used to calculate overall mortality. We connected the inpatient database to track clinical events that occurred after initiating edoxaban or warfarin, including ischemic stroke/systemic embolism (ISSE), acute myocardial infarction (AMI), congestive heart failure (CHF), intracranial hemorrhage (ICH), hospitalization for gastrointestinal bleeding (GIB), other hospitalized bleeding events (OHB), peripheral artery occlusive disease (PAOD) with gangrene, venous thromboembolism (VTE), pulmonary embolism (PE), AD, and acute kidney injury (AKI). Composite efficacy events included all-cause mortality, ISSE, AMI, and ICH; composite bleeding events included GIB and OHB. Demographic data were presented as mean ± standard deviation (SD) or median and interquartiles. Differences between the edoxaban and warfarin groups were assessed using chi-square test or Fisher's exact test for qualitative variables, Student's t-test for normally distributed quantitative variables, and Wilcoxon rank sum test for non-normally distributed variables. Age and sex were used for one to two matching to alleviate potential confounders between the two groups, with unmatched patients being excluded. Clinical events were reported as adjusted hazard ratio (aHR) with 95% confidence intervals (CI), determined using a Cox proportional hazards regression model. The model was adjusted for various factors including CHA2DS2-VASc score, HAS-BLED score, medical history, and medications. Additionally, a Fine-Gray model was employed to analyze all-cause mortality as a competing All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 8 outcome, presenting results as sub-distribution hazard ratio (SubHR) and adjusted SubHR after further adjustment on covariate to confirm the Cox proportional hazards regression model. Statistical analysis was conducted using SAS 9.1 software (SAS Institute, Inc., Cary, NC), with a significance level set at P < 0.05.

A total of 17,050 patients with NVAF between 2014 and 2018 were included (Figure 1). After matching for age and gender in NVAF patients without cancer, the median age was 73 years, median CHA2DS2-VASc score was 4.5, median HAS-BLED score was 2.7, and 58% of them were male (Table 1). Patients treated with edoxaban had higher CHA2DS2-VASc score, more cardiovascular (CV) risk factors, chronic diseases, medications and PCI than patients treated with warfarin. Conversely, patients treated with edoxaban had less chronic kidney disease (CKD) and coronary artery bypass graft surgery than patients treated with warfarin. Compared with warfarin, edoxaban was associated with significantly reduced composite efficacy events (aHR: 0.07 [0.05-0.08], p < 0.001), composite bleeding events (aHR: 0.13 [0.1-0.16], p < 0.001), all-cause mortality (aHR: 0.37 [0.33-0.42], p < 0.001), panarterial events (aHRs: 0.23 [0.2-0.27], p < 0.001), stroke, AMI, PAOD, venous thrombosis, and pulmonary embolism (Table 2, Figures 2 & 3). Edoxaban also significantly lowered aHR of CHF (0.26 [0.23-0.29], p < 0.001) (Table 2, Figures 2 & 3-6) versus warfarin. For neurological outcomes, edoxaban was associated with a lower aHR of AD (0.66 [0.56-0.78], p < 0.001) (Table 2, Figure 2 & 3-10). Fine-Gray analysis confirmed the results when all-cause mortality was modelled as a competing outcome (Supplemental Table 1). After matching for age and gender in NVAF-C patients, the mean age was 75 years, 59% of patients were male, their median CHA2DS2-VASc score was 4, and the median HAS- All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 9 BLED score was 2.9. After matching for age and gender, NVAF-C patients treated with edoxaban had higher CHA2DS2VASC score, more CV risk factors, chronic diseases, medications than NVAF-C patients treated with warfarin, but less CKD than those treated with warfarin (Table 3). NVAF-C patients treated with edoxaban had lower aHR of composite efficacy events (0.07 [0.04-0.13], p < 0.001), composite bleeding events (0.03 [0.01-0.09], p < 0.001), and overall mortality (0.39 [0.3-0.51], p < 0.001) than other treated with warfarin (Table 4, Figure 2). Additionally, aHRs for ischemic events, congestive heart failure, and GIB were lower in NVAF-C patients treated with edoxaban. The competing risk analysis confirmed the

when all-cause mortality was modelled as a competing outcome (Supplemental Table 2).

In this study, edoxaban demonstrated greater beneficial effects than warfarin in NVAF patients with and without cancer, based on a lower aHR of thrombotic events, CHF, and mortality. Edoxaban was also associated with significantly reduced risks of AD versus warfarin in NVAF patients without cancer. Besides, edoxaban showed a significantly better safety profile than warfarin in composite bleeding events in NVAF patients with and without cancer. Surprisingly, this study revealed some noteworthy findings regarding impacts of DOACs on AD and CHF that had not been previously discussed. Besides anticoagulation properties, emerging data highlight additional benefits of DOACs through PARs inhibition. FXa, both in vitro and in vivo studies, has been found to promote atherosclerosis through PARs pathways12 and animal studies demonstrated that FXa inhibition reduced atherosclerosis, promoted plaque regression, reduced macrophage infiltration, enhanced collagen deposition, and decreased necrotic core size through PAR-1 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 10 and PAR-2 inhibition.12 Rivaroxaban also inhibited atherogenesis by suppressing macrophage autophagy and inflammasome activity mediated by FXa-PAR2.13 DOACs inhibited FXa and had anti-inflammatory effects, which reduced the progression of cardiac valve calcification and deterioration of renal function.14 In contrast, warfarin inhibited matrix gamma- carboxyglutamate Gla protein (MGP), promoting systemic calcification in the coronary and peripheral vasculature, valve, and heart.15 Moreover, a higher burden of coronary artery calcium was found to be closely linked to a higher risk of sudden cardiac death.16 These mechanisms may contribute to the lower incidence of thrombotic events and the higher survival associated with edoxaban compared to warfarin. Our study is the first to report that edoxaban is linked to a lower rate of CHF admissions compared to warfarin in patients with NVAF, both with and without cancer. AF frequently coexists with CHF, and the joint occurrence of AF and CHF may further worsen prognosis.17 There are some plausible explanations for edoxaban’s benefits on CHF. First, edoxaban effectively reduced AMI events, a crucial risk factor for CHF.18,19 Furthermore, the renin- angiotensin-aldosterone system (RAAS) is essential in the pathophysiology of heart failure.20 The blockade of PAR-1 signaling, such as that induced by DOAC, was recently shown to reverse RAAS activation and attenuate cardiac fibrosis and hypertrophy.19 PAR-2 signaling, which can be directly activated by FXa, also contributes to the pathogenesis of hypertrophy and CHF.21 Finally, edoxaban decreases the incidence of AKI, which may contribute to CHF. Therefore, edoxaban might significantly reduce CHF through reduction of AMI and AKI, and by inhibiting PARs. Our results showed that edoxaban, compared to warfarin, significantly lowered risks of AD in patients with NVAF, which was in line with previous studies. A recent review reported potential positive effects of anticoagulant on vascular component of AD and dementia.22 Studies in mice models also showed that normalizing the procoagulant state in AD led to a All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 11 reduction in AD pathological features, suggesting the impact of coagulation on AD pathogenesis.23 Furthermore, DOACs diminish oxidative stress, neuroinflammation and blood–brain barrier dysfunction, which, in turn, ameliorate memory loss, and the amyloid load in AD mouse models.24,25 Therefore, edoxaban, as an DOAC, may prevent dementia by a direct effect on the development of AD, and by protective effects on vascular and embolic brain impairment. An association between AF and AD has been reported, with an increase in risks of AD by 1.5-2.5 folds in AF patients.26 AF patients exhibit elevated thrombin level which can activate PAR-1 and cause nerve fiber conduction block.27 Moreover, PARs influence autophagy by affecting ROS production and the degradation of beta amyloid Aβ (1-42).10 Edoxaban may ameliorate cognitive deterioration of AD11 by reducing the expression of PARs, proinflammatory and profibrotic genes.28 Autopsy studies found small strokes in AD patients19,29 and microemboli from cardiac and vascular sources can cause brain structural changes.29 A recent study showed that in AF patients, even 89.9% were anticoagulated, 5.5% had a new brain infarct on MRI after 2 years and most infarcts were clinically silent.30 Both overt and silent brain infarcts led to cognitive decline.30 Therefore, differentiating vascular dementia from AD is challenging, leading to potential inclusion of vascular dementia cases within AD. In a previous study, comparing DOACs with warfarin in AF patients with cancer, apixaban showed lower risks of ISSE and major bleeding, while dabigatran and rivaroxaban showed similar risks. 31 Our study provides new insights that edoxaban significantly reduced ISSE and GIB in both NVAF patients with and without cancer. Cancer is a heterogeneous disease; therefore, results should be interpreted with caution. However, we still observed significantly greater clinical benefits on stroke and MI prevention All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 12 for edoxaban over warfarin, with preserved safety profile for edoxaban. In evaluating risks and benefits, edoxaban seemed to be preferable to warfarin for NVAF patients with cancer. There are several limitations. First, the study was retrospective and could be biased by confounding factors; therefore, we used propensity score matching to balance baseline characteristics between two groups. Second, the study was conducted in Taiwan, and results may not be generalizable to other populations with different races and healthcare systems. Third, the study did not adjust confounding factors such as smoking status and body mass index, which may have influenced outcomes. Lastly, the follow-up period was relatively short (up to 2 years), and long-term outcomes was not fully captured. In Taiwanese patients with NVAF, edoxaban, as compared with warfarin, was associated with reduced CHF in both NVAF patients with and without cancer, and lower risks of AD in NVAF patients without cancer. All-cause mortality, ISSE, and GIB events were also significantly reduced with edoxaban, along with preserved safety profile as reflected by lower composite bleeding events in both NVAF patients with and without cancer. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 13

Authors thank Daiichi Sankyo Taiwan Ltd for the financial support and assistance in publication. Sources of Funding The study was supported by Daiichi Sankyo Taiwan Ltd and NIH-NHLBI R01HL130539 (to MS-C). Disclosures The study was supported by Daiichi Sankyo Taiwan Ltd. MS-C has received grant funding from NIH (grant n umber NIH -NHLBI R01HL130539). All authors have nothing else to be disclosed. Data availability & Supplementary material Data are available in the article and its online Supplementary material. Hung-Pin Tu had full access to all the data in the study and takes responsibility for its integrity and the data analysis. The deidentified participant data will not be shared. Declaration of Helsinki The study complied with the Declaration of Helsinki, with ethical approval from the Institutional Research Ethics Committee of Kaohsiung Medical University Hospital (IRB Number: KMUHIRB-EXEMPT(II)-21090039). All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 14

1. Lippi G, Sanchis-Gomar F, Cervellin G. Global epidemiology of atrial fibrillation: An increasing epidemic and public health challenge. Int J Stroke. 2021;16:217-221. doi: 10.1177/1747493019897870 2. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22:983-988. doi: 10.1161/01.str.22.8.983 3. Suzuki S, Yamashita T, Okumura K, Atarashi H, Akao M, Ogawa H, Inoue H. Incidence of ischemic stroke in Japanese patients with atrial fibrillation not receiving anticoagulation therapy--pooled analysis of the Shinken Database, J-RHYTHM Registry, and Fushimi AF Registry. Circ J. 2015;79:432-438. doi: 10.1253/circj.CJ- 14-1131 4. Fitzpatrick T, Carrier M, Le Gal G. Cancer, atrial fibrillation, and stroke. Thromb Res. 2017;155:101-105. doi: 10.1016/j.thromres.2017.05.006 5. Menichelli D, Vicario T, Ameri P, Toma M, Violi F, Pignatelli P, Pastori D. Cancer and atrial fibrillation: Epidemiology, mechanisms, and anticoagulation treatment. Prog Cardiovasc Dis. 2021;66:28-36. doi: 10.1016/j.pcad.2021.04.004 6. Chiang CE, Okumura K, Zhang S, Chao TF, Siu CW, Wei Lim T, Saxena A, Takahashi Y, Siong Teo W. 2017 consensus of the Asia Pacific Heart Rhythm Society on stroke prevention in atrial fibrillation. J Arrhythm. 2017;33:345-367. doi: 10.1016/j.joa.2017.05.004 7. Chan YH, Chao TF, Lee HF, Chen SW, Li PR, Liu JR, Wu LS, Chang SH, Yeh YH, Kuo CT, et al. Clinical Outcomes in Atrial Fibrillation Patients With a History of Cancer Treated With Non-Vitamin K Antagonist Oral Anticoagulants: A Nationwide Cohort Study. Stroke. 2021;52:3132-3141. doi: 10.1161/STROKEAHA.120.033470 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 15 8. Posma JJ, Grover SP, Hisada Y, Owens AP, 3rd, Antoniak S, Spronk HM, Mackman N. Roles of Coagulation Proteases and PARs (Protease-Activated Receptors) in Mouse Models of Inflammatory Diseases. Arterioscler Thromb Vasc Biol. 2019;39:13-24. doi: 10.1161/ATVBAHA.118.311655 9. Price R, Mercuri NB, Ledonne A. Emerging Roles of Protease-Activated Receptors (PARs) in the Modulation of Synaptic Transmission and Plasticity. Int J Mol Sci. 2021;22. doi: 10.3390/ijms22020869 10. Matej R, Rohan Z, Holada K, Olejar T. The contribution of proteinase-activated receptors to intracellular signaling, transcellular transport and autophagy in Alzheimer's disease. Curr Alzheimer Res. 2015;12:2-12. doi: 10.2174/1567205012666141218123202 11. Zare D, Rajizadeh MA, Maneshian M, Jonaidi H, Sheibani V, Asadi-Shekaari M, Yousefi M, Esmaeilpour K. Inhibition of protease-activated receptor 1 (PAR1) ameliorates cognitive performance and synaptic plasticity impairments in animal model of Alzheimer's diseases. Psychopharmacology (Berl). 2021;238:1645-1656. doi: 10.1007/s00213-021-05798-8 12. Posthuma JJ, Posma JJN, van Oerle R, Leenders P, van Gorp RH, Jaminon AMG, Mackman N, Heitmeier S, Schurgers LJ, Ten Cate H, et al. Targeting Coagulation Factor Xa Promotes Regression of Advanced Atherosclerosis in Apolipoprotein-E Deficient Mice. Sci Rep. 2019;9:3909. doi: 10.1038/s41598-019-40602-w 13. Ito Y, Maejima Y, Nakagama S, Shiheido-Watanabe Y, Tamura N, Sasano T. Rivaroxaban, a Direct Oral Factor Xa Inhibitor, Attenuates Atherosclerosis by Alleviating Factor Xa-PAR2-Mediated Autophagy Suppression. JACC Basic Transl Sci. 2021;6:964-980. doi: 10.1016/j.jacbts.2021.09.010 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 16 14. Di Lullo L, Lavalle C, Magnocavallo M, Mariani MV, Della Rocca DG, Severino P, Di Iorio BR, Russo D, Summaria F, Forleo GB, et al. New evidence of direct oral anticoagulation therapy on cardiac valve calcifications, renal preservation and inflammatory modulation. Int J Cardiol. 2021;345:90-97. doi: 10.1016/j.ijcard.2021.10.025 15. Poterucha TJ, Goldhaber SZ. Warfarin and Vascular Calcification. Am J Med. 2016;129:635 e631-634. doi: 10.1016/j.amjmed.2015.11.032 16. Razavi AC, Uddin SMI, Dardari ZA, Berman DS, Budoff MJ, Miedema MD, Osei AD, Obisesan OH, Nasir K, Rozanski A, et al. Coronary Artery Calcium for Risk Stratification of Sudden Cardiac Death: The Coronary Artery Calcium Consortium. JACC: Cardiovascular Imaging.0. doi: doi:10.1016/j.jcmg.2022.02.011 17. Aspromonte N, Colivicchi F. Clinical relevance of pharmacokinetic and pharmacodynamic properties of edoxaban when treating patients with atrial fibrillation and heart failure. Expert Opin Drug Metab Toxicol. 2017;13:113-122. doi: 10.1080/17425255.2017.1258054 18. Berezin AE, Berezin AA. Adverse Cardiac Remodelling after Acute Myocardial Infarction: Old and New Biomarkers. Dis Markers. 2020;2020:1215802. doi: 10.1155/2020/1215802 19. Callizo J, Feltgen N, Ammermann A, Ganser J, Bemme S, Bertelmann T, Pfeiffer S, Duvinage A, Groschel K, Hoerauf H, et al. Atrial fibrillation in retinal vascular occlusion disease and non-arteritic anterior ischemic optic neuropathy. PLoS One. 2017;12:e0181766. doi: 10.1371/journal.pone.0181766 20. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE, Jr., Colvin MM, Drazner MH, Filippatos GS, Fonarow GC, Givertz MM, et al. 2017 ACC/AHA/HFSA Focused Update of the 2013 ACCF/AHA Guideline for the Management of Heart Failure: A All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 17 Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. Circulation. 2017;136:e137-e161. doi: 10.1161/CIR.0000000000000509 21. Antoniak S, Sparkenbaugh EM, Tencati M, Rojas M, Mackman N, Pawlinski R. Protease activated receptor-2 contributes to heart failure. PLoS One. 2013;8:e81733. doi: 10.1371/journal.pone.0081733 22. Toribio-Fernandez R, Ceron C, Tristao-Pereira C, Fernandez-Nueda I, Perez-Castillo A, Fernandez-Ferro J, Moro MA, Ibanez B, Fuster V, Cortes-Canteli M. Oral anticoagulants: A plausible new treatment for Alzheimer's disease? Br J Pharmacol. 2023. doi: 10.1111/bph.16032 23. Cortes-Canteli M, Mattei L, Richards AT, Norris EH, Strickland S. Fibrin deposited in the Alzheimer's disease brain promotes neuronal degeneration. Neurobiol Aging. 2015;36:608-617. doi: 10.1016/j.neurobiolaging.2014.10.030 24. Tripathy D, Sanchez A, Yin X, Luo J, Martinez J, Grammas P. Thrombin, a mediator of cerebrovascular inflammation in AD and hypoxia. Front Aging Neurosci. 2013;5:19. doi: 10.3389/fnagi.2013.00019 25. Grossmann K. Direct Oral Anticoagulants (DOACs) for Therapeutic Targeting of Thrombin, a Key Mediator of Cerebrovascular and Neuronal Dysfunction in Alzheimer's Disease. Biomedicines. 2022;10. doi: 10.3390/biomedicines10081890 26. Ihara M, Washida K. Linking Atrial Fibrillation with Alzheimer's Disease: Epidemiological, Pathological, and Mechanistic Evidence. J Alzheimers Dis. 2018;62:61-72. doi: 10.3233/JAD-170970 27. Pompili E, De Franchis V, Giampietri C, Leone S, De Santis E, Fornai F, Fumagalli L, Fabrizi C. Protease Activated Receptor 1 and Its Ligands as Main Regulators of the All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 18 Regeneration of Peripheral Nerves. Biomolecules. 2021;11. doi: 10.3390/biom11111668 28. Oe Y, Hayashi S, Fushima T, Sato E, Kisu K, Sato H, Ito S, Takahashi N. Coagulation Factor Xa and Protease-Activated Receptor 2 as Novel Therapeutic Targets for Diabetic Nephropathy. Arterioscler Thromb Vasc Biol. 2016;36:1525-1533. doi: 10.1161/ATVBAHA.116.307883 29. Goldberg I, Auriel E, Russell D, Korczyn AD. Microembolism, silent brain infarcts and dementia. J Neurol Sci. 2012;322:250-253. doi: 10.1016/j.jns.2012.02.021 30. Kuhne M, Krisai P, Coslovsky M, Rodondi N, Muller A, Beer JH, Ammann P, Auricchio A, Moschovitis G, Hayoz D, et al. Silent brain infarcts impact on cognitive function in atrial fibrillation. Eur Heart J. 2022. doi: 10.1093/eurheartj/ehac020 31. Deitelzweig S, Keshishian AV, Zhang Y, Kang A, Dhamane AD, Luo X, Klem C, Ferri M, Jiang J, Yuce H, et al. Effectiveness and Safety of Oral Anticoagulants Among Nonvalvular Atrial Fibrillation Patients With Active Cancer. JACC CardioOncol. 2021;3:411-424. doi: 10.1016/j.jaccao.2021.06.004 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint 19 Figure Legends: Figure 1: Study flow - Selection of study patients from the National Health Insurance Research Database Figure 2: Outcomes of NVAF patients (with and without cancer) on Edoxaban or Warfarin to after 1:2 matching for sex and age Figure 3: 1) Primary composite efficacy events; 2) All-cause mortality; 3) Ischemic stroke/systemic embolism; 4) Overall venous thromboembolism; 5) Pulmonary embolism; 6) Congestive heart failure; 7) Hospitalized composite bleeding events; 8) Hospitalization, Gastrointestinal bleeding; 9) Other hospitalized bleeding events; 10) Alzheimer's disease All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Table 1. NV AF with Edoxaban vs Warfarin after 1:2 matching for sex and age Entire Cohort N = 17049 Edoxaban N = 5683 Warfarin N = 11366 p-value Age, mean (SD), years 72.8 (10.2) 72.8 (10.2) 72.8 (10.2) 0.9655 Age groups, n (%), years < 55 858 (5.0) 286 (5.0) 572 (5.0) 0.9998 55-64 2965 (17.4) 984 (17.3) 1981 (17.4) 65–74 5483 (32.2) 1830 (32.2) 3653 (32.1) 75–84 5935 (34.8) 1981 (34.9) 3954 (34.8) 85-100 1808 (10.6) 602 (10.6) 1206 (10.6) Male, n (%) 9879 (57.9) 3293 (57.9) 6586 (57.9) 1.0000 CHA2DS2-V ASc score*, mean (SD) 4.20 (1.8) 4.63 (1.8) 4.11 (1.8) < 0.0001 HAS-BLED score†, mean (SD) 2.72 (1.3) 3.14 (1.2) 2.69 (1.4) < 0.0001 Previous stroke, n (%) 20 (0.1) 17 (0.3) 3 (0.0) < 0.0001 Previous TIA, n (%) 6 (0.0) 6 (0.1) 0 (0.0) 0.0005 Comorbidity, n (%) All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Chronic kidney disease 4871 (28.6) 1510 (26.6) 3361 (29.6) <0.0001 Chronic liver disease 2389 (14.0) 898 (15.8) 1491 (13.1) <0.0001 Congestive heart failure 7674 (45.0) 2555 (45.0) 5119 (45.0) 0.922 Chronic lung disease 4214 (24.7) 1630 (28.7) 2584 (22.7) < 0.0001 Hypertension 11835 (69.4) 4308 (75.8) 7527 (66.2) < 0.0001 Hyperlipidemia 5475 (32.1) 2150 (37.8) 3325 (29.3) < 0.0001 Diabetes mellitus 12890 (75.6) 4597 (80.9) 8293 (73.0) < 0.0001 Myocardial infarction 789 (4.6) 287 (5.1) 502 (4.4) 0.0635 PAOD 1384 (8.1) 459 (8.1) 925 (8.1) 0.8896 History of bleeding 3257 (19.1) 1287 (22.6) 1970 (17.3) < 0.0001 Medications, n (%) Antiplatelet agents 8190 (48.0) 3580 (63.0) 4610 (40.6) < 0.0001 PPIs 6045 (35.5) 2194 (38.6) 3851 (33.9) < 0.0001 H2 blocker 10289 (60.3) 3836 (67.5) 6453 (56.8) < 0.0001 Statin 7285 (42.7) 2708 (47.7) 4577 (40.3) < 0.0001 Amiodarone/dronedarone 5245 (30.8) 2162 (38.0) 3083 (27.1) < 0.0001 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Beta-blocker 11333 (66.5) 4321 (76.0) 7012 (61.7) < 0.0001 Diltiazem/verapamil 868 (5.1) 360(6.3) 508 (4.5) < 0.0001 Digoxin 4980 (29.2) 1541 (27.1) 3439 (30.3) < 0.0001 PCI, n (%) 492 (2.9) 206 (3.6) 286 (2.5) < 0.0001 CABG, n (%) 52 (0.3) 4 (0.1) 48 (0.4) < 0.0001 *CHA2DS2-V ASc score assigns points for congestive heart failure, hypertension, diabetes mellitus, vascular disease, age s, and female sex and previous stroke or TIA. †HAS-BLED score assigns points for hypertension, abnormal renal or liver function, stroke, bleeding history, labile INR, age 65 years or older, and antiplatelet drug or alcohol use (Labile INR was excluded due to unavailability in claims data.) CABG: coronary artery bypass graft; INR: international normalized ratio; PAOD: peripheral arterial occlusive disease; PCI: percutaneous coronary intervention; PPI: proton pump inhibitor; TIA: transient ischemic attack; SD: standard deviation; NVAF: non-valvular atrial fibrillation. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Table 2 Matched NV AF patient outcomes: Edoxaban vs. Warfarin in 2-year follow-up Median (IQR) Event/total, (%) HR (95% CI) p-value Adjusted HR (95% CI) p-value Primary composite efficacy events a Warfarin 2.00 (1.35-2.00) 1790/3907, 45.82 1.00 1.00 Edoxaban 2.00 (2.00-2.00) 80/3907, 2.05 0.03 (0.03-0.04) < 0.0001 0.07 (0.05-0.08) < 0.0001 All-cause mortality Warfarin 1.16 (0.56-1.83) 2220/3907, 56.82 1.00 1.00 Edoxaban 0.98 (0.37-1.59) 394/3907, 10.08 0.21 (0.19-0.23) < 0.0001 0.37 (0.33-0.42) < 0.0001 Ischemic stroke/systemic embolism Warfarin 2.00 (1.99-2.00) 1816/11366, 15.98 1.00 1.00 Edoxaban 1.11 (0.55-1.58) 152/5683, 2.67 0.26 (0.22-0.31) < 0.0001 0.23 (0.2-0.27) < 0.0001 Acute myocardial infarction Warfarin 2.00 (2.00-2.00) 346/11366, 3.04 1.00 1.00 Edoxaban 1.12 (0.55-1.58) 90/5683, 1.58 0.91 (0.72-1.15) 0.4386 0.71 (0.56-0.9) 0.0051 PAD-gangrene/hospitalization/foot ulcers Warfarin 2.00 (2.00-2.00) 594/11366, 5.23 1.00 1.00 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Edoxaban 1.12 (0.55-1.58) 97/5683, 1.71 0.55 (0.44-0.69) < 0.0001 0.48 (0.39-0.60) < 0.0001 Overall venous thromboembolism Warfarin 2.00 (2.00-2.00) 338/11366, 2.97 1.00 1.00 Edoxaban 1.13 (0.56-1.59) 47/5683, 0.83 0.66 (0.48-0.91) 0.0108 0.55 (0.4-0.76) 0.0003 Pulmonary embolism Warfarin 2.00 (2.00-2.00) 87/11366, 0.77 1.00 1.00 Edoxaban 1.13 (0.56-1.59) 6/5683, 0.11 0.23(0.1-0.53) 0.0006 0.20 (0.08-0.46) 0.0002 Congestive heart failure Warfarin 2.00 (1.06-2.00) 3432/11366, 30.20 1.00 1.00 Edoxaban 1.07 (0.50-1.55) 346/5683, 6.09 0.30 (0.27-0.34) < 0.0001 0.26 (0.23-0.29) < 0.0001 Intracranial hemorrhage Warfarin 2.00 (2.00-2.00) 138/11366, 1.21 1.00 1.00 Edoxaban 1.13 (0.56-1.59) 50/5683, 0.88 1.27 (0.91-1.77) 0.1546 1.01 (0.72-1.41) 0.9749 Hospitalized composite bleeding events b Warfarin 2.00 (2.00-2.00) 1015/11366,8.93 1.00 1.00 Edoxaban 2.00 (2.00-2.00) 86/5683,1.51 0.17 (0.14-0.21) < 0.0001 0.13 (0.1-0.16) < 0.0001 Hospitalization, gastrointestinal bleeding All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Warfarin 2.00 (2.00-2.00) 1069/11366, 9.41 1.00 1.00 Edoxaban 1.09 (0.52-1.56) 273/5683, 4.80 0.98 (0.86-1.12) 0.7583 0.74 (0.64-0.85) < 0.0001 Other hospitalized bleeding events Warfarin 2.00 (2.00-2.00) 1432/11366, 12.60 1.00 1.00 Edoxaban 1.05 (0.48-1.53) 507/5683, 8.92 1.40 (1.26-1.56) < 0.0001 1.19 (1.07-1.32) 0.0016 Alzheimer's disease Warfarin 2.00 (2.00-2.00) 892/11366, 7.85 1.00 1.00 Edoxaban 1.10 (0.54-1.57) 184/5683, 3.24 0.71 (0.61-0.83) < 0.0001 0.66 (0.56-0.78) < 0.0001 Acute kidney injury Warfarin 2.00 (2.00-2.00) 502/11366, 4.42 1.00 1.00 Edoxaban 1.11 (0.55-1.58) 145/5683, 2.55 1.01 (0.84-1.22) 0.9083 1.06 (0.88-1.29) 0.5364 Adjusted HR was calculated after considering covariates, including previous stroke, previous TIA, comorbidities (chronic kidney disease, chronic liver disease, congestive heart failure, chronic lung disease, hypertension, hyperlipidemia , diabetes mellitus, myocardial infarction, PAOD), history of bleed ing (The factors related to the CHA2DS2 -V ASc score and HAS-BLED score are already included in the information provided above), and medications. a Primary composite efficacy events included all-cause mortality, ischemic stroke/systemic embolism, and acute myocardial infarction. b Hospitalized composite bleeding events included hospitalized gastrointestinal bleeding and other hospitalized bleeding events, except for ICH. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Note: All-cause mortality and primary composite efficacy events were matched for medication and follow -up period due to the shorter follow-up period for edoxaban in this database. HR: hazard ratio; IQR: interquartile range; CI: confidence of interval; PAD: peripheral arterial disease; NV AF: non-valvular atrial fibrillation All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Table 3. NV AF Cancer patients with Edoxaban vs Warfarin after 1:2 matching for sex and age Entire Cohort N = 2109 Edoxaban N = 703 Warfarin N = 1406 p-value Age, mean (SD), years 75.1 (9.0) 75.1 (9.0) 75.1 (8.9) 0.9249 Age groups, n (%), years < 55 30 (1.4) 10 (1.4) 20 (1.4) 1.0000 55-64 276 (13.1) 92 (13.1) 184 (13.1) 65–74 678 (32.1) 226 (32.1) 452 (32.1) 75–84 840 (39.8) 280 (39.8) 560 (39.8) 85-100 285 (13.5) 95 (13.5) 190 (13.5) Male, n (%) 1452 (68.8) 484 (68.8) 968 (68.8) 1.0000 CHA2DS2-V ASc score*, mean (SD) 4.2 (1.5) 4.3 (1.4) 4.1(1.5) 0.0368 HAS-BLED score†, mean (SD) 2.9 (1.3) 3.1 (1.2) 2.9 (1.3) 0.0002 Previous stroke, n (%) 1 (0.0) 1 (0.1) 0 (0.0) 0.1572 Previous TIA, n (%) 1 (0.0) 1 (0.1) 0 (0.0) 0.1572 Comorbidity, n (%) All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Chronic kidney disease 752 (35.7) 210 (29.9) 542 (38.5) < 0.0001 Chronic liver disease 452 (21.4) 154 (21.9) 298 (21.2) 0.7075 Congestive heart failure 1006 (47.7) 332 (47.2) 674 (47.9) 0.7579 Chronic lung disease 651 (30.9) 254 (36.1) 397 (28.2) 0.0002 Hypertension 1607 (76.2) 567 (80.7) 1040 (74.0) 0.0007 Hyperlipidemia 665 (31.5) 231 (32.9) 434 (30.9) 0.3535 Diabetes mellitus 1769 (83.9) 614 (87.3) 1155 (82.1) 0.0022 Myocardial infarction 96 (4.6) 34 (4.8) 62 (4.4) 0.6576 PAOD 186 (8.8) 59 (8.4) 127 (9.0) 0.6251 History of bleeding 455 (21.6) 174 (24.8) 281 (20.0) 0.0121 Medications, n (%) Anthracycline 76 (3.6) 23 (3.3) 53 (3.8) 0.5631 Doxorubicin 41 (1.9) 15 (2.1) 26 (1.8) 0.6555 Epirubicin 35 (1.7) 8 (1.1) 27 (1.9) 0.1849 Zorubicin 1 (0.1) 0 (0.0) 1 (0.1) 1.0000 Idarubicin 1 (0.1) 0 (0.0) 1 (0.1) 1.0000 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Antiplatelet agents 1056 (50.1) 427 (60.7) 629 (44.7) < 0.0001 PPIs 1093 (51.8) 363 (51.6) 730 (51.9) 0.9019 H2 blocker 1454 (68.9) 526 (74.8) 928 (66.0) < 0.0001 Statin 876 (41.5) 297 (42.2) 579 (41.2) 0.6393 Amiodarone/dronedarone 680 (32.2) 269 (38.3) 411 (29.2) < 0.0001 Beta-blocker 1476 (70.0) 533 (75.8) 943 (67.1) < 0.0001 Diltiazem/verapamil 137 (6.5) 44 (6.3) 93 (6.6) 0.7548 Digoxin 651 (30.9) 205 (29.2) 446 (31.7) 0.2302 PCI, n (%) 64 (3.0) 22 (3.1) 42 (3.0) 0.8575 CABG, n (%) 4 (0.2) 0 (0.0) 4 (0.3) 0.1569 *CHA2DS2-V ASc score assigns points for congestive heart failure, hypertension, diabetes mellitus, vascular disease, ages, and female sex and previous stroke or TIA. †HAS-BLED score assigns points for hypertension, abnormal renal or liver function, stroke, bleeding history, labile INR, age 65 years or older, and antiplatelet drug or alcohol use (Labile INR was excluded due to unavailability in claims data.) CABG: coronary artery bypass graft; INR: international normalized ratio; PAOD: peripheral arterial occlusive disease; PCI: pe rcutaneous coronary intervention; PPI: proton pump inhibitor; TIA: transient ischemic attack; SD: standard deviation; NVAF: non-valvular atrial fibrillation. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Table 4. Matched NV AF cancer patient outcomes: Edoxaban vs. Warfarin in 2-year follow-up Median (IQR) Event/total, (%) HR (95% CI) p-value Adjusted HR (95% CI) p-value Primary composite efficacy events a Warfarin 2.00 (1.77-2.00) 172/489, 35.17 1.00 1.00 Edoxaban 2.00 (2.00-2.00) 11/489, 2.25 0.05 (0.03-0.1) < 0.0001 0.07 (0.04-0.13) < 0.0001 All-cause mortality Warfarin 1.43 (0.89-1.90) 289/489, 59.1 1.00 1.00 Edoxaban 1.22 (0.64-1.72) 77/489, 15.75 0.32 (0.25-0.41) < 0.0001 0.39 (0.3-0.51) < 0.0001 Ischemic stroke/systemic embolism Warfarin 2.00 (2.00-2.00) 229/1406, 16.29 1.00 1.00 Edoxaban 1.16 (0.57-1.64) 23/703, 3.27 0.30 (0.20-0.47) < 0.0001 0.30 (0.19-0.46) < 0.0001 Acute myocardial infarction Warfarin 2.00 (2.00-2.00) 46/1406, 3.27 1.00 1.00 Edoxaban 1.18 (0.57-1.65) 10/703, 1.42 0.76 (0.38-1.52) 0.4431 0.48 (0.21-1.08) 0.0773 PAD-gangrene/hospitalization/foot ulcers Warfarin 2.00 (2.00-2.00) 76/1406, 5.41 1.00 1.00 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Edoxaban 1.17 (0.55-1.66) 16/703, 2.28 0.65 (0.38-1.13) 0.1260 0.74 (0.41-1.32) 0.3016 Overall venous thromboembolism Warfarin 2.00 (2.00-2.00) 63/1406, 4.48 1.00 1.00 Edoxaban 1.18 (0.59-1.66) 8/703, 1.14 0.53(0.25-1.12) 0.1009 0.53 (0.25-1.12) 0.1009 Pulmonary embolism Warfarin 2.00 (2.00-2.00) 16/1406, 1.14 1.00 1.00 Edoxaban 1.18 (0.59-1.66) 2/703, 0.28 0.4(0.09-1.76) 0.2870 0.4 (0.09-1.76) 0.2870 Congestive heart failure Warfarin 2.00 (1.09-2.00) 431/1406, 30.65 1.00 1.00 Edoxaban 1.10 (0.53-1.59) 53/703, 7.54 0.35 (0.27-0.47) < 0.0001 0.31 (0.23-0.41) < 0.0001 Intracranial hemorrhage Warfarin 2.00 (2.00-2.00) 14/1406, 1.00 1.00 1.00 Edoxaban 1.19 (0.57-1.66) 7/703, 1.00 1.67 (0.66-4.23) 0.2822 1.26 (0.48-3.29) 0.6417 Hospitalized composite bleeding events b Warfarin 2.00 (2.00-2.00) 195/1406, 13.87 1.00 1.00 Edoxaban 2.00 (2.00-2.00) 4/703, 0.57 0.04 (0.01-0.1) < 0.0001 0.03 (0.01-0.09) < 0.0001 Hospitalization, gastrointestinal bleeding All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Warfarin 2.00 (2.00-2.00) 166/1406, 11.81 1.00 1.00 Edoxaban 1.15 (0.55-1.64) 38/703, 5.41 0.84 (0.58-1.19) 0.3246 0.67 (0.46-0.98) 0.0384 Other hospitalized bleeding events Warfarin 2.00 (1.83-2.00) 273/1406, 19.42 1.00 1.00 Edoxaban 1.07 (0.48-1.55) 83/703, 11.81 1.13 (0.88-1.45) 0.3435 1.09 (0.84-1.41) 0.5193 Alzheimer's disease Warfarin 2.00 (2.00-2.00) 137/1406, 9.74 1.00 1.00 Edoxaban 1.16 (0.57-1.64) 28/703, 3.98 0.68 (0.45-1.02) 0.0642 0.73 (0.48-1.12) 0.1496 Acute kidney injury Warfarin 2.00 (2.00-2.00) 98/1406, 6.97 1.00 1.00 Edoxaban 1.16 (0.57-1.65) 29/703, 4.13 0.98 (0.64-1.49) 0.9132 1.16 (0.75-1.79) 0.4954 Adjusted HR was calculated after considering covariates, including previous stroke, previous TIA, comorbidities (chronic kidney disease, chronic liver disease, congestive heart failure, chronic lung disease, hypertension, hyperlipidemia, diabetes mellitus, myocardial infarction, PAOD), history of bleeding (The factors related to the CHA2DS2 -V ASc score and HAS-BLED score are already included in the information provided above), and medications. a Primary composite efficacy events included all-cause mortality, ischemic stroke/systemic embolism, and acute myocardial infarction. b Hospitalized composite bleeding events included hospitalized gastrointestinal bleeding and other hospitalized bleeding events, except for ICH. Note: All-cause mortality and primary composite efficacy events were matched for medication and follow -up period due to the shorter follow-up All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint period for edoxaban in this database. HR: hazard ratio; IQR: interquartile range; CI: confidence of interval; PAD: peripheral arterial disease; NV AF: non-valvular atrial fibrillation All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint No cancerEdoxabanvs. Warfarin Cancer Non-Valvular Atrial Fibrillation (NV AF) (1:2 propensity matching) Primary composite efficacy events aHR:0.07 P<0.001Hospitalized composite bleeding eventsaHR:0.13P<0.001 Primary composite efficacy events aHR:0.07 P<0.001Hospitalized composite bleeding eventsaHR:0.03P<0.001Edoxabanvs. Warfarin All-cause mortality Ischemic stroke/systemic embolismAcute myocardial infarctionPeripheral arterial diseaseVenous thrombo-embolismCongestive heart failureAlzheimer's disease 63% 77% 29% 52%45% 74% …? 34% All-cause mortality Ischemic stroke/systemic embolismAcute myocardial infarctionPeripheral arterial diseaseVenous thrombo-embolismCongestive heart failureAlzheimer's disease 61% 70% 52% 26%47% 69% …? 27% **** *** * * ** p<0.05 Follow-up period: 2 years N =17,049 N =2,109 Statistically significant novel findings All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 1 : Study flow -Selection of study patients from the National Health Insurance Research Database All NHIRD beneficiaries between 2014-2018N ≥ 23,403,635 Cases of atrial fibrillation (AF) between 2014 and 2018 N = 314,774Non-Valvular Atrial Fibrillation (NV AF) patients between 2014 and 2018N = 174,501*No edoxaban/warfarin usedN = 134,620 Dates of edoxaban/warfarin used unknownN = 723 Edoxaban/Warfarin discontinuedN = 11,894 Incident edoxabanused (n = 5,684) or incident warfarin used (n = 21,580) in NV AF patientsN = 27,264 NV AF patients on edoxabanN = 5,684NV AF patients on warfarinN = 11,366 Matching 1:2 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 2 : Outcomes of NVAF patients (with and without cancer) on Edoxabanor Warfarin to after 1:2 matching for sex and age Edoxabanbetter Warfarin better 00.511.522.533.5Acute kidney injury(Malignancy)Acute kidney injury(No malignancy)Alzheimer's disease(Malignancy)Alzheimer's disease(No malignancy)Other hospitalized bleeding events (Malignancy)Other hospitalized bleeding events (No malignancy)Hospitalization, Gastrointestinal bleeding(Malignancy)Hospitalization, Gastrointestinal bleeding(No malignancy)Hospitalized composite bleeding events(Malignancy)Hospitalized composite bleeding events(No malignancy)Intracranial hemorrhage(Malignancy)Intracranial hemorrhage(No malignancy)Congestive heart failur(Malignancy)Congestive heart failur(No malignancy)Pulmonary embolism(Malignancy)Pulmonary embolism(No malignancy)Overall venous thromboembolism(Malignancy)Overall venous thromboembolism(No malignancy)PAOD Gangrene/hospitalization/Foot ulcers(Malignancy)PAOD Gangrene/hospitalization/Foot ulcers(No malignancy)Acute myocardial infarction(Malignancy)Acute myocardial infarction(No malignancy)Ischemic stroke/systemic embolism(Malignancy)Ischemic stroke/systemic embolism(No malignancy)All-cause mortality (Malignancy)All-cause mortality (No malignancy)Primary composite efficacy events (Malignancy)Primary composite efficacy events (No malignancy)Primary composite efficacy events No cancerCancer All-cause mortality No cancerCancer Ischemic stroke/systemic embolism No cancerCancer Acute myocardial infarction No cancerCancer PAOD/hospitalization/Foot ulcers No cancerCancer Venous thromboembolism No cancerCancer Pulmonary embolism No cancerCancer Congestive heart failure No cancerCancer Intracranial hemorrhage No cancerCancer Hospitalized composite bleeding No cancerCancer Hospitalization, GI bleeding No cancerCancer Other hospitalized bleeding events No cancerCancer Alzheimer's disease No cancerCancer Acute kidney injury No cancerCancer All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-1 Primary composite efficacy eventsall-cause mortality+ ischemic stroke/systemic embolism + acute myocardial infarction All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-2 All-cause mortality All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-3 Ischemic stroke/systemic embolism All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-4 Overall venous thromboembolism All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-5 Pulmonary embolism All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-6 Congestive heart failure All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-7 Hospitalized composite bleeding eventshospitalized gastrointestinal bleeding + other hospitalized bleeding events All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-8 Hospitalization, Gastrointestinal bleeding All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-9 Other hospitalized bleeding events All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint Figure 3-10 Alzheimer's disease All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for thisthis version posted October 26, 2023. ; https://doi.org/10.1101/2023.10.25.23297577doi: medRxiv preprint