ETNA-AF-Europe
Research type
Research Study
Full title
Non-interventional study on Edoxaban treatment in routine clinical practice for patients with non valvular atrial fibrillation
IRAS ID
184456
Contact name
Ameet Bakhai
Contact email
Sponsor organisation
Daiichi Sankyo Europe GmbH
Duration of Study in the UK
5 years, 5 months, 28 days
Research summary
Summary of Research
The primary objective of this study is to collect real-world safety data on bleeding events including intracranial haemorrhage, drug related adverse events such as liver adverse events, cardiovascular (CV) and all-cause mortality in AF patients treated with Edoxaban up to 4 years.
Secondary objective is to assess the effect of Edoxaban on patient reported outcomes as strokes (ischaemic and haemorrhagic), systemic embolic events (SEE), major cardiovascular events (MACE), hospitalisations related to CV condition, persistence to therapy, patient/physician reported outcome, health care utilization and resource use.Summary of Results
Edoxaban was approved in 2015 by the European Commission for the prevention of stroke and systemic embolism in adult subjects with non-valvular atrial fibrillation (NVAF) with 1 or more risk factors, such as congestive heart failure (CHF), hypertension, age ≥75 years, diabetes mellitus, prior stroke or transient ischaemic attack (TIA), as well as for the treatment of deep vein thrombosis (DVT) and pulmonary embolism (PE) and for the prevention of recurrent DVT and PE in adults. As part of an ongoing European risk-management plan, several post-authorisation safety studies (PASSs) have been designed to assess the real-world outcomes of edoxaban-treated subjects in the European Medicines Agency-approved indications. The aim of this non-interventional study on Edoxaban Treatment in routiNe clinical prActice for patients with non-valvular Atrial Fibrillation (ETNA-AF EU) is to understand the risks and benefits of edoxaban use in a real-world clinical setting for the NVAF indication in Europe. Real-world evidence data from routine clinical practice use of edoxaban up to 4 years was collected and evaluated in 13,164 subjects treated by specialised and non-specialised physicians in hospitals and office-based centres.Research Question and Objectives
The primary objective of this study was to collect and evaluate real-world safety data on bleeding events, including intracranial haemorrhage, drug-related adverse events such as liver adverse events, and cardiovascular (CV) and all-cause mortality in NVAF subjects treated with edoxaban up to 4 years with regard to onset (relative to treatment with edoxaban) of the event, duration, severity and outcome.
The secondary objective of this study was to assess the effect of edoxaban on subject-relevant outcomes, including strokes (ischaemic and haemorrhagic), systemic embolic events (SEE), TIAs, major adverse CV events (MACE), venous thromboembolisms (VTEs), acute coronary syndrome (ACS), hospitalisations related to CV condition, and extent of exposure and compliance to edoxaban therapy as well as rate and reasons of permanent discontinuation of edoxaban.
The results relating to safety and effectiveness of this study were set into perspective with the respective results of other anticoagulants obtained from the internal non-interventional disease registry program PREvention oF thromboembolic events-European Registry in Atrial Fibrillation (PREFER in AF program), which includes PREFER in AF, Prolongation of PREFER in AF, and the Extension of the Prolongation of PREFER in AF. The results of the non-selected subject population of the ETNA-AF EU were also set into perspective with the data from European countries obtained in the Phase 3, randomised, double-blind, double-dummy, parallel-group, multicentre, multinational “Effective aNticoaGulation with factor XA next GEneration in Atrial Fibrillation” (ENGAGE AF-TIMI 48 EU) study.
Furthermore, the reported events and adverse drug reactions (ADRs) were set into perspective with external Electronic Medical Records (EMR) data provided by SYNEOS Health and with the spontaneously reported ADRs (only ADRs reported for the AF indication) in the Daiichi Sankyo Europe (DSE) Clinical Safety and Pharmacovigilance (CSPV) global safety database. The timeframe and the included countries DSE CSPV global safety database matched with the data collection in ETNA-AF EU.
Study Design
This is a European, multinational, multicentre, prospective, non-interventional, observational PASS.
Subjects with NVAF from 10 European countries (Austria, Belgium, Germany, Ireland, Italy, the Netherlands, Portugal, Spain, Switzerland, and United Kingdom) were enrolled and followed up for 4 years.Setting
Subjects from 776 sites in 10 European countries were enrolled and followed up for 4 years. To be considered eligible, sites were required to have access to NVAF subjects that may be treated with edoxaban, to be able to access the Electronic Data Capture system, to record data in English, and to provide adequate time and staff to perform all study-related documentation activities (e.g. participate in required trainings, enter study data). They also had to agree to perform follow-up (FU) in subjects for a period of 4 years according to routine clinical practice. The total study period from First Patient In (06 Aug 2015) to Last Patient Out (13 Apr 2022) was 80 months. Subjects who permanently discontinued edoxaban during the observation period had to be followed up annually for further 2 years or until the end of the observational period (whichever came first).
Edoxaban (Marketing Authorization Holder: DSE GmbH) was prescribed for the prevention of stroke and thromboembolism in patients with NVAF. Subjects were either edoxaban naïve or non-naïve at study entry. Treatment pattern and treatment initiation, continuation, or changes were solely at the discretion and responsibility of the physician and the subject. The study did not affect the prescribing behaviour as subjects were included in the study after the treating physician had made the clinical decision to prescribe edoxaban. All medications were prescribed in the usual standard of care and were not provided by the study sponsor. Participation in the study did not influence payment or reimbursement for any treatment received by subjects during the study. No study-specific visits were performed.Subjects and Study Size, Including Dropouts
The subject recruitment period per country was 12 months. After the recruitment, subjects were followed for 48 months.
Subjects were enrolled if they were currently being treated or planned to be treated with edoxaban for NVAF at study entry, provided written informed consent to participate, and were not simultaneously participating in any other interventional study. Simultaneous participation in any other non-interventional study/registry was allowed. Explicit exclusion criteria were not defined.
The sample size justification was based on the incidence rate of intracranial haemorrhages as it is the smallest among the incidence rates of interest (ie, major bleeding events, ischaemic stroke, and intracranial haemorrhage). Taking the real-life situation into account and on the basis of Hart et al., an event rate of 0.35% (0.0035) per year for intracranial haemorrhage was assumed. For 4 years, this would result in a rate of 1.4% (0.014). To estimate the 95% confidence interval (CI) with a precision of ±0.25% for the rate of 1.4% (corresponding to a relative precision of 17.8%), 8485 subjects were needed. Assuming a drop-out rate for the 4 years of about 35%, about 13,100 enrolled subjects would allow for a reasonable estimation of the event rate.
Variables and Data Sources
Parameters recorded during the observation period included the following:
• Subject characteristics (demography, medical history, history/diagnosis and current status of NVAF, risk factors, comorbidities, relevant pre- and concomitant medications, previous interventions for NVAF)
• Efficacy endpoints (stroke [ischaemic or haemorrhagic], SEE, TIA, MACE, VTE occurrences, ACS, CHF, CV events leading to hospitalisation, other AF-relevant clinical events)
• Safety endpoints (major bleeding events, including intracranial haemorrhage, clinically relevant non-major bleeding events, gastrointestinal [GI] bleeding events, ADRs)
• Mortality (all-cause and CV-related)
• Laboratory values (as available in medical records, eg aspartate transaminase, alanine transaminase, bilirubin, serum creatinine, creatinine clearance)
• Use pattern of edoxaban in NVAF (documentation of therapy, dosage, prescription intervals, dose adjustments, permanent discontinuations, switches and reasons for these, concomitant medication, additional therapy/interventions, physician’s judgement on the compliance to edoxaban).Suspected edoxaban-related ADRs were collected and coded using the standardised Medical Dictionary for Regulatory Activities. Physicians were not required to perform any mandatory laboratory assessments for this study.
As this was a non-interventional study, only data from clinical routine practice were documented. To facilitate accurate recording of data, subjects could optionally fill in a memory aid to note important details.
Adverse events of special interest (major bleedings, clinically relevant non-major bleedings, strokes, SEE, and deaths) were adjudicated by the clinical event adjudication committee. Assessment was made by consensus.
Results
ETNA-AF EU included a total of 13,632 subjects across 10 European countries, of which 13,164 were included in the analysis and presented in this report. Subjects had a median age of 75.0 years, and 56.7% were male. A total of 53.9% of subjects had paroxysmal NVAF, 24.2% had persistent NVAF, and 22.0% had long-standing persistent or permanent NVAF. At Baseline, 75.9% of subjects were prescribed 60 mg and 24.0% were prescribed 30 mg edoxaban. Dosing of edoxaban was largely adequate according to the recommendations of the summary of product characteristics (SmPC), with 69.2% (N=9354) of the subjects receiving the correct dose (60 mg: 53.7% [N=7264] and 30 mg: 15.5% [N=2090]).
The rate of adjudicated all-cause mortality within the total 48-month FU period was 14.0% (95% CI 13.43, 14.62), CV-related mortality 3.4% (95% CI 3.14, 3.77), and subjects with at least 1 stroke event 2.1% (95% CI 1.84, 2.33). The rate of all-cause mortality was higher among subjects with long-standing persistent or permanent NVAF (20.6%) compared with subjects with paroxysmal NVAF (11.4%) and persistent NVAF (14.0%). Among the non-adjudicated events, VTE occurred in 0.3% of subjects (95% CI 0.20, 0.40), MACE including bleeding in 5.1% (95% CI 4.70, 5.46), and 25.4% (95% CI 24.71, 26.20) of subjects were hospitalised due to CV-related reasons. A total of 3.3% of the subjects developed CHF during the FU, which resulted in an overall proportion of 18.8% (95% CI 18.13, 19.47) of subjects with CHF when considering those with CHF already present at Baseline (15.5%).
During edoxaban treatment, all-cause mortality was reported in 10.8% (95% CI 10.2, 11.3) of subjects and CV-related mortality in 2.5% (95% CI 2.2, 2.7). Stroke was documented in 1.7% (95% CI 1.5, 2.0) of subjects and ischaemic stroke in 1.4% (95% CI 1.2, 1.6). The rates of SEE, VTE, and ACS were low. MACE including bleeding occurred in 4.1% (95% CI 3.8, 4.5) of subjects and 23.7% (95% CI 23.0, 24.4) of subjects were hospitalised due to CV-related reasons.
For bleeding events, only events occurring on-edoxaban during the 48-month FU were presented. The rate of subjects with at least 1 bleeding event was 8.9% (95% CI 8.39, 9.37). This rate consisted of the following (adjudicated) events: 2.6% of the subjects (95% CI 2.30, 2.84) had major bleeding events, 2.4% (95% CI 2.12, 2.64) had clinically relevant non-major bleeding events, 0.6% (95% 0.43, 0.70) had intracranial haemorrhage events, and 3.2% (95% CI 2.93, 3.54) had GI bleeding events.
There was no major difference in bleeding complications by the type of NVAF.A total of 332 ADRs were recorded between Baseline and Month 48, and 3 further ADRs were recorded beyond this prespecified time-window but within the study period. Two hundred and seventy-four subjects (2.0% of all subjects) had least 1 ADR and 156 subjects (1.2% of all subjects) had at least 1 serious ADR Thirteen ADRs (4.1% of all ADRs) had a fatal outcome. When ADRs were categorised by PT, the most frequently observed were haemorrhage (0.3%), GI haemorrhage (0.3%), and pruritus (0.2%). Apart from fatigue (0.1%), all reported ADRs are expected reactions according to the SmPC of edoxaban. It is important to note that due to the long time between 2 data collection points, some data may have been underreported by participating patients, or there may be potential chances of misclassification of treatment changes and concomitant drug exposure. Therefore, the rates of nonserious events and ADRs, especially after 3 or 4 years, may rather represent the lower bound of the true rate observed over the long-term. In addition, bleeding events may only have been reported as clinical events and not as ADRs, which further reduces the ADR rate.
Clinical effectiveness events in ETNA-AF EU as compared to the overall PREFER in AF program both on NOACs and VKA were, in general, comparable. The major difference was observed in the annual rates of CV-related hospitalisations, which were higher in ETNA-AF EU than in PREFER in AF program during VKA/NOAC treatment in all populations (except VKA 1-year FU population due to unavailable data). The annual rate of major bleeding events was slightly lower in ETNA-AF EU compared to the VKA population in PREFER in AF program. The rate of GI bleedings was also lower in ETNA-AF EU compared to VKA subjects at 1-year FU and those with long-term VKA use but higher than in NOAC subjects at 1-year FU and matched NOAC/VKA subjects.
When comparing the results of ETNA-AF EU with ENGAGE AF-TIMI 48 EU, the annual rates of ischaemic stroke, TIA, VTE occurrences, and MACE in ETNA-AF EU subjects were lower compared to the ENGAGE AF-TIMI 48 EU population. On the other hand, the annual rate of all-cause death was higher in ETNA-AF EU compared to ENGAGE AF-TIMI 48 EU (both edoxaban and warfarin). This might be due to the older population in ETNA-AF EU. However, the annual rate of CV-related mortality was lower, which might suggest a population at lower risk of ischaemic events in ETNA-AF EU as already reflected in a lower CHA2DS2–VASc score. As for bleeding events during edoxaban treatment, the rates of major bleeding events, clinically relevant non-major bleeding, intracranial haemorrhage events, and GI bleeding events (including major GI bleeding events) were lower in ETNA-AF EU compared to ENGAGE AF-TIMI 48 EU.
Compared to the EMR study, the annual rates of all-cause and CV-related mortality during the total observational period were lower in ETNA-AF EU, particularly when looking at German and Spanish populations in the EMR study. When considering events occurring on-edoxaban, the annual rates of all-cause and CV-related mortality as well stroke, including ischaemic stroke, MACE, and hospitalisations related to CV reasons were also lower in ETNA-AF EU compared to EMR study, while the rates of TIA and SEE were comparable. The annual rate of bleeding events during the edoxaban treatment in ETNA-AF EU, including major bleeding events and clinically relevant non-major bleeding events, were lower compared to the corresponding treatment period in the EMR study. However, interpretations should be made with caution due to the high heterogeneity observed among the countries in EMR study caused primarily by the use of different types of data collection within the local databases.
The proportion of subjects with at least 1 ADR was substantially lower in ETNA-AF EU (ADR 2.0%) than the percentage in ENGAGE AF-TIMI 48 EU (7.7%). Furthermore, the proportion of subjects with at least 1 serious ADR in ETNA-AF EU (1.2%) was higher compared with ENGAGE AF-TIMI 48 EU (0.5%), while the rate of death due to ADR was comparable (0.1% vs 0.09%). The total number of prespecified drug-related ADRs was 335 in ETNA-AF EU (N=13,520) and 84 in ENGAGE AF-TIMI 48 EU edoxaban group (N=3449), and the percentage of subjects with any prespecified ADRs was comparable (2.0% and 2.1%, respectively).
Despite low numbers in ETNA-AF EU, the trends of ADRs were similar in spontaneous reporting and ETNA-AF EU. Main ADRs were related to bleeding events, allergic reactions, or general events, such as dizziness and nausea. In addition, except for fatigue, no new ADRs were reported in ETNA-AF EU, which confirms the safety profile of edoxaban.
Conclusion
Overall, ETNA-AF provides robust evidence for the clinical effectiveness and safety, specifically in terms of bleeding complications for edoxaban in real-world clinical practice in NVAF patients when treated with edoxaban up to 48 months.
REC name
North West - Greater Manchester East Research Ethics Committee
REC reference
15/NW/0586
Date of REC Opinion
27 Jul 2015
REC opinion
Favourable Opinion