75 / 210
Ponencias
75
Real-world evidence for non-vitamin K antagonist oral anticoagulants in atrial fibrillation
J. Beyer-Westendorf
Center for Vascular Medicine and Department of Medicine III. Division of Angiology, Thrombosis Research Unit. University Hospital “Carl Gustav Carus” Dresden. Dresden,
Germany
Introduction
The non-vitamin K antagonist oral anticoagulants (NOACs)
apixaban, dabigatran, edoxaban and rivaroxaban are approved for
the prevention of stroke and systemic embolism in patients with
non-valvular atrial fibrillation (AF) and one or more risk factors. The
approvals of the NOACs in this setting were based on the results of
large phase III clinical trials, which demonstrated similar or impro-
ved efficacy compared with warfarin, an acceptable safety profile
and reductions in the rate of intracranial haemorrhage (ICH).
Randomized controlled trials (RCTs) are the most scientifically
rigorous method of hypothesis testing available and are considered
the gold standard for evaluating the efficacy of a particular interven-
tion. By necessity, RCTs have strict study protocols and designs,
with well-defined inclusion and exclusion criteria and adjudicated
endpoints. However, both the risk profile and the management of
patients enrolled in clinical trials may differ from that of patients
seen in routine care, and this can impact on event rates and reduce
the applicability of clinical trial data to a broader patient population.
This is where real-world data are of value. Definitions of what
constitutes real-world data can vary, but it is widely accepted as
healthcare data that have been collected outside the context of
RCTs
1
. Real-world studies are important because they can pro-
vide information about the benefits and harms of therapy in a
wider population and establish whether outcomes seen in clinical
trials are replicated in routine clinical practice. This is particu-
larly important for safety outcomes. Even in large clinical trials,
outcome rates, particularly for adverse events, can be so low or
follow-up so short that rare events and complications may not be
identified. Because real-world studies, and especially retrospective
healthcare database analyses, can include a much larger number
of patients and can cover a long period of follow-up, they provide
the opportunity to detect safety signals that may not have been
apparent in clinical trials. Real-world evidence is also important
to gather information on the management and outcomes of patient
groups that may have been under-represented or not even included
in RCTs because of the study inclusion and exclusion criteria. Fur-
thermore, real-life studies can evaluate treatment decision patterns
or management strategies for treatment complications. Because of
the randomization process and strict guidance from a trial protocol,
such behavioural assessments cannot derived from RCTs.
Real-world evidence for stroke prevention in atrial
fibrillation with NOAC
Use of the NOACs in clinical practice continues to grow and
there is accumulating real-world evidence relating to patient adhe-
rence to and persistence with these drugs and their safety and
effectiveness in routine care. Several different sources of real-
world data exist, including prospective studies, registry data and
database analyses, but interpretation of these data is not always
straightforward. At present, the majority of real-world studies have
evaluated the use of rivaroxaban or dabigatran, and real-world data
on apixaban and edoxaban are limited at present.
Safety
In clinical trials, all of the NOACs had similar or lower rates
of major bleeding compared with warfarin, with significant reduc-
tions in ICH. A meta-analysis of the four pivotal NOAC trials con-
firmed these findings but highlighted an increased risk of gastroin-
testinal (GI) bleeding with the NOACs compared with warfarin
2
.
Several real-world studies have evaluated major bleeding rates
in patients taking NOACs and it is reassuring to note that irrespec-
tive of the type of study or the source of the data, patients in the
NOAC cohorts consistently had similar or lower major bleeding
rates compared with the respective VKA cohorts. For rivaroxaban,
major bleeding rates in real-world studies were generally lower
than those seen in ROCKET AF (2.1-3.3 per 100 patient-years
compared with 3.6 per 100 patient-years)
3,8
. This is probably
in part because ROCKET AF enrolled a moderate-to-high-risk
patient population (mean CHADS
2
score of 3.5) at increased risk
of bleeding. In contrast, the range of major bleeding rates for
dabigatran in real-world studies is consistent with findings from
the RE-LY trial (2.1-4.3 per 100 patient-years compared with 2.9
[dabigatran 110 mg] and 3.4 [dabigatran 150 mg] per 100 patient-
years)
9-12
. There is a lack of apixaban real-world data, but one early
evaluation of bleeding-related hospital readmission data from two
US hospital claims databases indicated that among patients with
non-valvular AF taking NOACs, apixaban reduced the risk of ble-
eding-related readmissions compared with rivaroxaban
13
.
In the Graham et al.
study, the rates of bleeding for both dabi-
gatran and warfarin were comparatively higher than the corres-
ponding rates seen in other contemporary real-world studies
11
.
This may be explained by the fact that the Graham study included
only elderly patients withAF, increasing the potential for bleeding
events with the use of anticoagulant therapy
11
. Another example
could be the substantially lower rates of major bleeding seen in
the Fontaine et al.
study, which may be attributed to a bleeding
definition that differed from those used in other studies
14
. In the
Fontaine study, the definition of major bleeding was similar to the
definition of critical bleeding used in ROCKET AF. In contrast,
the International Society on Thrombosis and Haemostasis (ISTH)
major bleeding definition used in several of the other real-world