WELCOME

The Conference would start in some minutes, please wait

for the Moderator to start
Meanwhile keep your audio desactivated

“Material Técnico Científico para uso exclusivo del cuerpo médico”

 Professor John Camm, UK

What RWE is Showing us about NOACs Use

• Honorary consultant physician in the department of cardiology at St
Bartholomew's Hospital and St. George´s Hospital, London.
• Professor of clinical cardiology specializing in cardiac arrhythmias
sudden cardiac death, inherited channel pathyatrial fibrillation,
ventricular tachycardia, cardiomyopathy and pacemakers

“Material Técnico Científico para uso exclusivo del cuerpo médico”

 Basic Navegation buttons

Want to talk with the Want to send a written message/

incorporated audio ? question?
Press the microphone icon and
wait for the generation of the Open the panel chat for write.
green animation.

Want to make a Question?

Press the hand icon and the
moderator would give you the
word

“Material Técnico Científico para uso exclusivo del cuerpo médico”

 What RWE is Showing us About
NOACs Use

John Camm
St. George’s University of London
and Imperial College, United
Kingdom

“Material Técnico Científico para uso exclusivo del cuerpo médico”

 Declaration of Competing Interests
Chairman: ESC Guidelines on Atrial Fibrillation, 2010 and Update, 2012; ACC/AHA/ESC
Guidelines on VAs and SCD; 2006; NICE Guidelines on ACS and NSTEMI, 2012; NICE
Guidelines on Heart Failure, 2008; Member: NICE Guidelines on AF, 2006; ESC VA and
SCD Guidelines, 2015; Reviewer: AHA/ACC/HRS Guidelines on AF, 2014; ACC/AHA/HRS
SVT Guidelines, 2015; ESC AF Guidelines, 2016.

Steering Committees: multiple trials involving antiarrhythmic agents, heart failure drugs
and novel anticoagulants.

DSMBs: multiple trials of devices and drugs.

Events Committees: one trial of novel oral anticoagulants and multiple trials of
miscellaneous agents with CV adverse effects.

Editorial Role: Editor-in-Chief, EP-Europace and Clinical Cardiology; Editor, European

Textbook of Cardiology, European Heart Journal, Electrophysiology of the Heart, and
Evidence Based Cardiology.

Consultant/Advisor/Speaker: Astellas, Astra Zeneca, Gilead, Incarda, Merck, Menarini,

Milestone, Otsuka, Sanofi, Servier, Xention, Bayer, Boehringer Ingelheim, Bristol-Myers
Squibb, Daiichi Sankyo, Pfizer, Boston Scientific, Biotronik, Medtronic, Sorin, St. Jude
Medical, Actelion, GlaxoSmithKline, InfoBionic, Johnson and Johnson, Mitsubishi,
Novartis, Takeda
 The Real-World: State of Play
Proportion of men and
Criteria Patients % women receiving oral
Total number of anticoagulants
patients (Aetna, 70
16.2 million
Humana, Harvard
Pilgrim) 60

231,696
Patients with AF 50
(1.4% of all pts)

AF pts with 201,882 40

CHA2DS2-VASc ≥ 2 (87% of AF pts)
30
Patients with at 105,256
least one oral anti- (52% of AF pts 20
coagulation fill CHA2DS2-VASc ≥ 2)
10
Proportion of days
covered by anti-
32% 0
coagulation in AF
18-34 y/o 35-44 y/o 45-64 y/o 65-74 y/o > 75 y/0
patients
Men Women
Pokorney S, et al. J Am Coll Cardiol 2016;67:886-886
 Real World Data vs. Clinical Trial Data
Real world evidence RCT data
Carefully selected patients who can
Unselected eligible patients,
Enrols cope with either therapy and the
according to SmPC / Drug label
rigors of RCT
Effectiveness, safety Efficacy
Primarily
Management issues Safety
investigates
Persistence Adverse event profile
Only primary outcome(s) are
Based on many variables
Sample size powered. Subgroup outcomes are
(centre size, funding, objectives)
not powered
Patients closely represent Patients are selected and monitored
Key benefits everyday practice guided by carefully.
SmPC / Drug label; no protocol Detailed protocol is followed.
Few if any costs are re- Travel/trial participation costs/trial
Costs
imbursed. Trial drug is not free. drugs and often OMT costs are met.
Care may vary from SmPC
Key Event definitions may vary Do the results translate readily into
limitations between centres and individuals; everyday clinical practice?
potential reporting bias
 Registry/Database Challenges
● Selected patient populations
● May not be generally representative: single country / one health
insurer / only hospitalised patients / particular clinics
● Retrospective studies
● Reliance on coded outcome events; many choices
● Registry designed for another purpose
● Variable quality, and usually unaudited data
● Outcome measures inconsistent, ill-defined - not adjudicated
● No design publication, registry of studies, or prospective
commitment
● Survival bias
● Prospective studies
● Lack of training/quality control, ? recruitment discipline
● Failure to enrol consecutive patients - selection bias on part of
patient or physician
● Incomplete follow-up with many drop outs - under-reporting
 Large AF Registries
Population
Registry Patient enrolment – key design features Follow-up duration
size
GARFIELD-AF1 Target: 57,000 • Prospective patients (approx.52,000) enrolled <6 weeks ≥2 years, up to 8 years
after AF diagnosis in 5 sequential cohorts Garfield-AF unique
• Retrospective patients (approx. 5,000) enrolled 6–24 study design
months after diagnosis
• ≥1 additional investigator-determined stroke risk factor

GLORIA-AF2 Target: 56,000 • Prospective patients enrolled <3 months after AF 0–3 years
diagnosis in 3 phases Phase 1 (pre-NOAC): none
Phase 2 (Dabigatran): 2 years
• CHA2DS2-VASc score ≥1 Phase 3 (VKA/NOAC): 3 years

ORBIT-AF I3 Target 10,000 • Incident or prevalent AF ≥2 years

10,132 enr\olled • Patients excluded if anticipated life expectancy <6
months

ORBIT-AF II4 Target: 15,000 • Prospective patients enrolled <6 months after AF ≤2 years
diagnosis; or enrolled <3 months after initiation or
transition to a NOAC
• Pts excluded if anticipated life expectancy <6 months

PREFER in AF5 Target: 7,000 • Prospective patients enrolled <12 months after AF 1 year
7243 enrolled diagnosis

1. Kakkar AK et al. Am Heart J. 2012;163:13-19 e1; 2. Huisman MV et al. Am Heart J. 2014 Mar;167(3):329-34; 3. Piccini JP et al.
Am Heart J. 2011;162:606-612.e1; 4. Steinberg BA. Am Heart J. 2014;168:160-7. 5. Kirchhof P et al. Europace. 2014;16:6-14.
 Evolution in Baseline Treatment for
Patients Enrolled in GARFIELD-AF
VKA±AP FXA/DTI±AP AP None
100
Proportion of patients on treatment, %

80

60
4,2 13,8 26,3 37,2 43,1
40
60% 71.8%
57.4%

20 40%

0
Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5
2010–2011 2011–2013 2013–2014 2014–2015 2015–2016

Cohorts 1–5, N=51,270

GARFIELD-AF: NOAC High Proportion of OAC
Cohort 5: Data Collected 2015-16
Number % NOAC
% NOAC % NOAC % NOAC
Country Identifier of Cohorts 3-
Cohort 3 Cohort 4 Cohort 5
Patients 5
AUSTRALIA 349 0.571 0.934 0.913 0.829
AUSTRIA 164 0.800 0.891 0.833 0.853
BELGIUM 1074 0.786 0.860 0.893 0.830
CANADA 454 0.593 0.830 0.776 0.709
GERMANY 1510 0.692 0.748 0.828 0.744
JAPAN 3857 0.773 0.828 0.878 0.825
SWITZERLAND 80 . 0.833 0.898 0.877
UNITED ARAB EMIRATES 332 0.603 0.750 0.750 0.694
UNITED STATES 590 0.745 0.749 0.772 0.754
 How Are Low- and High-Risk AF
Patients Managed in Practice?
Contrary to international guideline recommendations:
 28% of high-risk patients (CHA2DS2-VASc ≥2) are not anticoagulated
 51% of very low-risk patients (CHA2DS2-VASc = 0) are anticoagulated

CHA2DS2-VASc 0 1 ≥2
100
90 None
80 AP
Proportion of
patients (%)

70 DTI+AP
60 DTI
50
FXaI+AP
40
FXaI
30
VKA+AP
20
10
VKA
0
(n=352) (n=1336) (n=9027)

Camm AJ et al, Heart 2016 (in press)

 How Are Low- and High-Risk AF
Patients Managed in Practice?
Contrary to international guideline recommendations:
 28% of high-risk patients (CHA2DS2-VASc ≥2) are not anticoagulated
 51% of very low-risk patients (CHA2DS2-VASc = 0) are anticoagulated

CHA2DS2-VASc 0 1 ≥2
100
90 None
80 AP
Proportion of
patients (%)

70 DTI+AP
60 DTI
50
FXaI+AP
40
FXaI
30
VKA+AP
20
10
VKA
0
(n=352) (n=1336) (n=9027)

Camm AJ et al, Heart 2016 (in press)

 How Are Low- and High-Risk AF
Patients Managed in Practice?
Contrary to international guideline recommendations:
 28% of high-risk patients (CHA2DS2-VASc ≥2) are not anticoagulated
 51% of very low-risk patients (CHA2DS2-VASc = 0) are anticoagulated

CHA2DS2-VASc 0 1 ≥2
100
90 None
80 AP
Proportion of
patients (%)

70 DTI+AP
60 DTI
50
FXaI+AP
40
FXaI
30
VKA+AP
20
10
VKA
0
(n=352) (n=1336) (n=9027)

Camm AJ et al, Heart 2016 (in press)

 Dabigatran: Favourable Benefit-Fisk Profile
FDA study of >134 000 Medicare patients
Dabigatran was associated with a lower risk of ischaemic stroke,
intracranial haemorrhage and death than warfarin
Incidence rate
per 1000 person-years Adjusted HR
Dabigatran (95% CI)
Warfarin
etexilate

Ischaemic stroke 11.3 13.9 0.80 (0.67–0.96)

Intracranial haemorrhage 3.3 9.6 0.34 (0.26–0.46)

Major GI bleeding 34.2 26.5 1.28 (1.14–1.44)

Acute MI 15.7 16.9 0.92 (0.78–1.08)

Mortality 32.6 37.8 0.86 (0.77–0.96)

Comparison of matched new-user cohorts treated with dabigatran etexilate 150 mg or 75 mg* or
warfarin for non-valvular AF based on 2010–2012 Medicare data *Primary findings are based on
analysis of both doses (no stratification by dose)

Graham DJ, et al. Circulation. 2014 Oct 30. pii: CIRCULATIONAHA.114.012061

Rivaroxaban Safety Profile in Real Life was
Consistent with Findings from ROCKET AF
ROCKET AF1 US DoD PMSS2
Mean CHADS2 score 3.5 Mean CHADS2 score 2.2
N=7111 N=27,467
4 4
Event rate (%/year)

Event rate (%/year)

3,5
3 3.6 3
2,5 2.9
2 2
1,5
1 1
0,5
0 0
Major bleeding* Major bleeding#

Clinical endpoint % (n) Clinical endpoint % (n)

ICH 0.8 (55) ICH 0.1 (36)
Fatal bleeding 0.4 (27) Fatal bleeding <0.1 (14)
Major GI bleeding 3.2 (224) Major GI bleeding 1.5 (423)
Median duration of treatment exposure was 590 days Rivaroxaban users were followed for 455 days
Results are not intended for direct comparison
*Major bleeding definitions according to ISTH; #major bleeding was defined by the Cunningham algorithm3
1. Patel MR et al, N Engl J Med 2011;365:883–891; 2. Tamayo S et al, Clin Cardiol 2015;38:63-68;
3. Cunningham A et al, Pharmacoepidemiol Drug Saf 2011;20:560–566
GI Bleeding: 86% of all Major Bleeds
US DoD Database Analysis
 51,842 NVAF patients taking rivaroxaban were included
MB Cases
N=1613

MB Incidence Rate per 100 person-years (95% CI)* 2.71 (2.58–2.84)

Gastrointestinal 1386 (85.9)

MB Site, n Intracranial 133 (8.2)

(% of those who bled) Genitourinary 14 (0.9)
Other/Unspecified 80 (5.0)

Fatal MB Incidence Rate per 100 person-years (95% CI)# 0.08 (0.06–0.11)

*The MB incidence rate was calculated using person-time for the denominator value (exposure time at risk) for all first major bleeding events within the study
period; #Occurred during hospitalization for the MB event

Tamayo S et al, Circulation 2016:134:A15047

Real-world Comparison of Bleeding in NVAF
Apixaban, Dabigatran, Rivaroxaban
Data source: MarketScan Earlyview insurance claims database,
Eligible patients: NVAF patients ≥18 years who received NOAC or switched from
warfarin to NOAC from 01/01/2013-31/10/2014 NOAC were excluded.
Apixaban: 8,785
Dabigatran: 20,963
Rivaroxaban: 30,529
Follow-up: Patients were followed up to 6 months until bleeding, discontinuation/switch of
therapy, disenrollment, or end of the study.

Tepper P, et al. ESC abstract 2015

Real-world Comparison of Bleeding in NVAF
Apixaban, Dabigatran, Rivaroxaban
Data source: MarketScan Earlyview insurance claims database,
Eligible patients: NVAF patients ≥18 years who received NOAC or switched from
warfarin to NOAC from 01/01/2013-31/10/2014 NOAC were excluded.
Apixaban: 8,785
Dabigatran: 20,963
Rivaroxaban: 30,529
Follow-up: Patients were followed up to 6 months until bleeding, discontinuation/switch of
therapy, disenrollment, or end of the study.

Tepper P, et al. ESC abstract 2015

 Major Bleeding with NOACs
● Truven MarketScan® US
claims database
● NVAF patients
● Aged ≥18 years
● Newly prescribed oral
anticoagulant
● 01Jan2013–31Dec2014
● 1-year baseline period
● Major bleeding: bleeding
requiring hospitalization
● Propensity score
matching
● No efficacy data

Lip GYH, et al. Thromb Haemost 2016; 116:

 Major Bleeding in Selected Real-World
Rivaroxaban Studies
XANTUS Dresden REVISIT Tamayo Laliberté Tepper
20151 20142 20163 20154 20145 20156
Prospective Prospective Retrospective Retrospective Retrospective Retrospective
Design
registry registry claims claims claims claims
Definition of
FDA mini- Cunningham Author
major ISTH ISTH ?
sentinel algorithm determined
bleeding
Primary
ICD-9/10 Primary and
N/A N/A Primary (occasionally ?
code position secondary
secondary)
Rate of major
bleeding, 2.1 3.1 NR 2.86 12.79 20.2
%/year
Rate of ICH,
0.4 NR 0.49 0.22 1.8 2.4
%/year
Rate of GI
bleeding, 0.9 NR NR 2.53 9.5 6.2
%/year
Adjudicated events FDA, US Food and Drug Administration; GI, gastrointestinal; ICH, intracranial haemorrhage; N/A, not applicable; NR, not reported

1. Camm AJ et al, Eur Heart J 2016;37:1145–1153; 2. Beyer-Westendorf J et al, Blood 2014;124:955–962; 3. Coleman CI et al, JICE
2016;45:253 3–5; abstract 15-48; 4. Tamayo S et al, Clin Cardiol 2015;38:63–68; 5. Laliberté F et al, Curr Med Res Opin
2014;30:1317–1325; 6. Tepper P et al, Eur Heart J 2015;36:338; abstract 1975
 Inconsistency in Major Bleeding
Definitions
 Schemas to identify bleeding- 5
Proportion %
related hospitalizations in 4,66
4,5
claims data differ in both the
specific codes used and 4

coding positions allowed 3,5

 Within MarketScan claims 3

2,65
data, identified adults with 2,5

NVAF and newly started on an 2 1,87

OAC from 1/2012 to 6/2015 1,5

 151,738 new users of OACs 1

with NVAF (median CHA2DS2- 0,5
VASc score=3, HAS-BLED
0
score=3)
Cunningham FDA Yao
Coleman CI et al, Data on file 1.81-1.94 2.57-2.74 4.55-4.76
 NSAIDs and Upper Gastrointestinal
Bleeding Risk: Danish Registry Data
Periods of drug use Number Person- Observed Expected O/E 95% CI
years (O) (E)
Current use of NSAID 156,138 107,305 515 124.9 4.12 3.8–4.5
NSAID only 152,882 94,987 365 101.2 3.61 3.3–4.0
NSAID + glucocorticoids 17,875 5908 58 8.0 7.24 5.5–9.4
NSAID + glucocorticoids + other drug* 1593 464 7 1.1 6.41 2.6–13.2
NSAID + anticoagulants 1001 340 8 0.7 11.46 4.9–22.6
NSAID + anticoagulants + other drug# 178 35 0 0.1 - -
NSAID + glucocorticoids +
154 29 1 0.1 18.74 0.2–10.4
anticoagulants ± other drugs
NSAID ± low-dose ASA ± high-dose ASA 10,246 5542 76 13.8 5.52 4.3–6.9
Former use of NSAID 145,877§ 314,278 370 264.4 1.40 1.3–1.5
Non-use of any other drug 144,584 294,676 267 224.9 1.19 1.0–1.3
Current use of any other drug‡ (not
28,455 19,062 103 39.6 2.60 2.1–3.2
NSAID)

*Not anticoagulants; #Not glucocorticoids; ‡Includes drugs suspected to predispose to UGIB (low and high dose ASA, anticoagulants and glucocorticoids; §Among the 156,138
NSAID users, 145,877 were followed during periods of non-use ≥90 days after a non-renewed prescription

Mellemkjaer L et al, Br J Clin Pharmacol 2002:53:173–181

Predictors of Gastrointestinal Bleeding
Patients with Atrial Fibrillation
Characteristic Adjusted Hazard Ratio (95%CI)

Age, years, (reference: <55 years) ---

55–64 1.54 (0.89–2.68)
65–74 2.72 (1.59–4.65)
≥75 4.52 (2.68–7.64)
Heart Failure 1.25 (1.01–1.56)
Coronary artery disease 1.37 (1.10–1.69)
Renal impairment 1.67 (1.24–2.25)
Prior Bleeding 1.32 (1.01–1.72)
Alcohol abuse 2.57 (1.52–4.35)
Helicobacter pylori infection 4.75 (1.93–11.68)
Corticosteroids (and stress) 1.17 (0.95–1.45)

Lauffenburger JC et al, Pharmacotherapy 2015:5:560–568

 Gastro-Intestinal Bleeding
Rivaroxaban and Warfarin
Rivaroxaban Warfarin
HR (95% CI)
HR (95% CI)
Rivaroxaban vs
Rivaroxaban vs warfarin
warfarin
Rate (%/year) Rate (%/year)

Abraham et al,
2.84 3.06 0.93 (0.69–1.25)
20151

Yao et al, 20162 3.26 2.53 1.21 (1.02–1.43)

Hohnloser et al,
4.5 3.5 1.39 (1.20–1.59)
20173
0,125 0,25 0,5 1 2 4

 Only Abraham et al used propensity score matching based on ICD-9-CM code-

identified comorbid conditions that are known predictors of GI bleeding, including1
 History of previous GI bleed, diverticulosis and Helicobacter pylori infection, and
specific concomitant medications including NSAIDs
1. Abraham NS et al, BMJ 2015:350:h1857
2. Yao X et al, J Am Heart Assoc 2016:5:pii: e003725
3. Hohnloser SH et al, Clin Res Cardiol 2017 DOI 10.1007/s00392-017-1098-x
 REVISIT-US Study Design
Real-world EVIdence on Stroke prevention In patients with
aTrial Fibrillation in the United States

Optimize “Internal Validity”

Combined Endpoint of Ischemic Stroke and ICH
● Most likely to be coded accurately and with less variability in
claims data and of equal importance to allow for benefit/risk
assessment
● Used validated ICD-9 coding algorithms and restricted
codes to the primary diagnosis code position
 May miss cases, but greater robustness in those identified
● Data source: MarketScan Earlyview insurance claims
database
Coleman CI et al, ECAS 2016: Oral presentation
 REVISIT-US
Apixaban Rate Warfarin Rate HR (95% CI) HR (95% Cl)
(%/year) (%/year) apixaban vs. warfarin apixaban vs. warfarin

ICH 0.38 0.97 0.38 (0.17–0.88)

Ischemic stroke 0.56 0.51 1.13 (0.49–2.63)
Combined 0.89 1.44 0.63 (0.35–1.12)

0,125 0,25 0,5 1 2 4

Rivaroxaban Rate Warfarin Rate HR (95% CI) HR (95% CI)

(%/year) (%/year) rivaroxaban vs. warfarin rivaroxaban vs. warfarin

ICH 0.49 0.96 0.53 (0.35-0.79)

Ischemic stroke 0.54 0.83 0.71 (0.47-1.07)
Combined 0.95 1.60 0.61 (0.45-0.82)

0,125 0,25 0,5 1 2 4

Dabigatran Rate Warfarin Rate HR (95% CI) HR (95% Cl)

Favours Favours warfarin
apixaban vs. warfarin
(%/year) (%/year) dabigatran vs. warfarin dabigatran

ICH 0.46 0.66 0.71 (0.49-1.02)

Ischemic stroke 0.60 0.70 0.87 (0.63-1.22)
Combined 1.00 1.30 0.79 (0.62-1.02)

0,125 0,25 0,5 1 2 4

Coleman CI et al, presented at ESC 2016
Favours NOAC
 Outcome Events in New-User Cohorts of
Dabigatran and Rivaroxaban for NVAF
Outcome event counts, and crude and adjusted hazard ratios comparing inverse probability of treatment-
weighted new-user cohorts of dabigatran and rivaroxaban for non-valvular atrial fibrillation a

Crude (Unadjusted) Incidence

Hazard Ratio Hazard Ratio (95% CI)
Rate per 1,000 person-years
(95% CI) Adjusted Data
(No.of Events)
Outcome
Dabigatran Rivaroxaban
Crude Adjusted p Value
(n=52,240) (n=66,651)

Primary Outcomes
Thromboembolic stroke 9.7 (150) 7.7 (156) 0.80 (0.64‒1.00) 0.81 (0.65‒1.01) 0.07

Intracranial hemorrhage 3.7 (58) 5.8 (118) 1.58 (1.15‒2.16) 1.65 (1.20‒2.26) 0.002

Major extracranial bleeding 26.6 (413) 39.4 (796) 1.47 (1.31‒1.66) 1.48 (1.32‒1.67) <0.001

Gastrointestinal 23.3 (362) 32.5 (656) 1.39 (1.22‒1.58) 1.40 (1.23‒1.59) <0.001

Mortality 22.2 (346) 24.7 (500) 1.12 (0.98‒1.29) 1.15 (1.00‒1.32) 0.051

Secondary Outcomes
Hospitalized extracranial bleeds 39.2 (608) 54.0 (1,091) 1.38 (1.25‒1.52) 1.39 (1.25‒1.53) <0.001
Acute myocardial infarction 12.9 (200) 11.0 (223) 0.86 (0.71‒1.05) 0.88 (0.72‒1.06) 0.18
a Dabigatran served as the reference group.

Graham DJ, et al. JAMA Intern Med. doi:10.1001/jamainternmed.2016.5954

Prescribing Patterns of NOACs Globally
4th Quarter 2015
Apixaban1 Dabigatran1 Rivaroxaban1
Country
2.5 mg 5 mg 75 mg 110 mg 150 mg 10 mg 15 mg 20 mg
United States 25 75 16 0 84 5 19 76
Japan 46 54 28 72 0 45 55 0
Germany 43 57 1 60 39 3 33 63
Canada 37 63 1 50 49 4 25 71
Australia 40 60 0 63 36 2 29 69
United Kingdom 37 62 2 51 48 4 21 75
Spain 39 61 2 59 39 4 31 65
Belgium 31 69 0 59 41 2 40 58
Italy 37 63 0 62 37 1 36 63
% ARISTOTLE2 RE-LY3 ROCKET AF4
Phase III studies 4.7 50 (29) 21
Real world 37 50 31

In practice, prescriptions for apixaban at the lower 2.5 mg dose are disproportionately
high. Similar but less-marked patterns are seen with dabigatran and rivaroxaban

1. IMS MIDAS; 2. Granger CB et al, N Engl J Med 2011;365:981-92; 3. Connolly SJ et al, N Engl J
Med 2009;361:1139–1151; 4. Fox KAA et al, Eur Heart J 2011;32:2387–2394
Outcomes Associated with Reduced Dose
NOAC Treatment in Focused Populations

(N = 3,449)

(N = 5,347)

(N = 2,020)

(N = 417)

(N = 350)

(N = 315)

(N = 3,313)

(N = 5,159)

(N = 1,585)

Nielsen PB et al. BMJ 2017; 356: j510

 XANTUS: Study Objective and Design
• To collect real-life data on adverse events in patients with non-valvular AF
treated with rivaroxaban to determine the safety profile of rivaroxaban across the
broad range of patient risk profiles encountered in routine clinical practice
 Primary outcomes: major bleeding (ISTH definition),
all-cause mortality, any other adverse events
 Secondary outcomes: stroke and non-CNS systemic embolism
N=6,784 Data collection at initial
visit, hospital discharge
Population:
Rivaroxaban; (if applicable) and quarterly*
Adult patients with
treatment
non-valvular AF receiving 1 year
duration and
rivaroxaban for stroke/
dose at
non-CNS systemic embolism
physician’s
prevention, who had provided
discretion
written informed consent

Prospective, single-arm, observational, non-interventional phase IV Final visit:

study statistical analyses were descriptive and exploratory in nature 1 year#

AF=Atrial fibrillation; CNS=Central nervous system; ISTH=International Society on Thrombosis and Haemostasis
*Exact referral dates for follow-up visits not defined (every 3 months recommended); #for rivaroxaban discontinuation ≤1 year,
observation period ends 30 days after last dose. Observational design means no interference with clinical practice was allowed
Camm AJ, et al. Vasc Health Risk Manag. 2014;10:425–34; ClinicalTrials.gov: NCT01606995
 XANTUS: Patient Disposition
Major events, Screened 4149 patients excluded*
specifically major (N=10,934) Patient decision (n=1222)
bleeding, stroke, SE, Administrative reason (n=456)
TIA and MI,
Availability of drug (n=18)
adjudicated centrally
by an Adjudication Medical guidelines (n=399)
Committee blinded to Price of drug (n=473)
individual patient data Medical reasons (n=442)
Internal hospital guidelines (n=30)
Type of health insurance (n=183)
Enrolled Other (n=1454)
(N=6785)
Primary analysis
population: defined as 1 patient
all patients who had Did not take any rivaroxaban (n=1)
taken at least one dose Safety population
of rivaroxaban (N=6784)

Rivaroxaban 20 mg Rivaroxaban 15 mg Another dose

(n=5336) (n=1410) (n=35)#
*Reasons for not continuing in the study included, but were not limited to, patient decision, administrative or medical reasons. Some patients could have
more than one reason for exclusion; #other dose includes any initial daily rivaroxaban dose besides 15/20 mg od (excluding missing information, n=3)

Camm AJ, et al. Eur Heart J 2015; doi: 10.1093/eurheartj/ehv466

 XANTUS: Patient Disposition
Major events, Screened 4149 patients excluded*
specifically major (N=10,934) Patient decision (n=1222)
bleeding, stroke, SE, Administrative reason (n=456)
TIA and MI,
Availability of drug (n=18)
adjudicated centrally
by an Adjudication Medical guidelines (n=399)
Committee blinded to Price of drug (n=473)
individual patient data Medical reasons (n=442)
Internal hospital guidelines (n=30)
Type of health insurance (n=183)
Enrolled Other (n=1454)
(N=6785)
Primary analysis
population: defined as 1 patient
all patients who had Did not take any rivaroxaban (n=1)
taken at least one dose Safety population
of rivaroxaban (N=6784)

Rivaroxaban 20 mg Rivaroxaban 15 mg Another dose

(n=5336) (n=1410) (n=35)#
*Reasons for not continuing in the study included, but were not limited to, patient decision, administrative or medical reasons. Some patients could have
more than one reason for exclusion; #other dose includes any initial daily rivaroxaban dose besides 15/20 mg od (excluding missing information, n=3)

Camm AJ, et al. Eur Heart J 2015; doi: 10.1093/eurheartj/ehv466

 XANTUS: Baseline Demographics and
Clinical Characteristics
Rivaroxaban Rivaroxaban
(N=6,784) (N=6,784)
Age (years) VKA experienced 3,089 (45.5)
Mean±SD 71.5±10.0 VKA naïve 3,695 (54.5)
Age < 65, n (%) 1,478 (21.8) Creatinine clearance, n (%)
Age ≥ 65 – ≤ 75, n (%) 2,782 (41.0) <15 ml/min 20 (0.3)
Age > 75, n (%) 2,524 (37.2) ≥ 15 – < 30 ml/min 75 (1.1)
Gender (male), n (%) 4,016 (59.2) ≥ 30 – < 50 ml/min 545 (8.0)
Weight (kg), mean±SD 83.0±17.3 ≥ 50 – ≤ 80 ml/min 2,354 (34.7)
BMI (kg/m2), mean±SD 28.3±5.0 > 80 ml/min 1,458 (21.5)
CHADS2 score, mean±SD 2.0±1.3 Missing 2,332 (34.4)
CHA2DS2-VASc score, mean±SD 3.4±1.7 Co-morbidities, n (%)
AF, n (%) Hypertension 5,065 (74.7)
First diagnosed 1,253 (18.5) Diabetes mellitus 1,333 (19.6)
Paroxysmal 2,757 (40.6) Prior stroke/non-CNS SE/TIA 1,291 (19.0)
Persistent 923 (13.6) Congestive HF 1,265 (18.6)
Permanent 1,835 (27.0) Prior MI 688 (10.1)
Missing 16 (0.2) Hospitalization at baseline, n (%) 1,226 (18.1)
Camm AJ, et al. Eur Heart J. 2016;37:1145–53
 Cumulative Rates (Kaplan–Meier) for
Treatment-Emergent Primary Outcomes

0.05 All-cause death In total, 6,522 (96.1%) patients did not

Major bleeding experience any of the outcomes of
Stroke/SE treatment-emergent all-cause death,
Cumulative event rate

0.04
major bleeding or stroke/SE
0.03

0.02

0.01

0
0 30 60 90 120 150 180 210 240 270 300 330 360 390 420
Time to event (days)
Patients at risk
All-cause death 6,784 6,530 6,349 6,211 6,054 5,938 5,853 5,754 5,679 5,597 5,512 5,295 4,307 1,153 514
Major bleeding 6,784 6,522 6,340 6,197 6,033 5,909 5,824 5,726 5,649 5,559 5,471 5,256 4,273 1,144 513
Stroke/SE 6,784 6,532 6,353 6,216 6,053 5,933 5,848 5,752 5,674 5,587 5,499 5,282 4,296 1,149 513

Camm AJ, et al. Eur Heart J. 2016;37:1145–53

 Treatment Persistence and
Patient Satisfaction
● Persistence with rivaroxaban in XANTUS was 80% at 1 year
● Over 75% of patients were ‘very satisfied/satisfied’ with their treatment

Rivaroxaban
0.45
N=6,784
(discontinuation of treatment)

0.40 n (%)
Cumulative event rate

0.35
0.30 Very satisfied 2,247 (33.1)

0.25
Satisfied 2,849 (42.0)
0.20
0.15 Neutral 877 (12.9)
0.10
0.05 Unsatisfied 340 (5.0)

0
Very
0 30 60 90 120 150 180 210 240 270 300 330 360 116 (1.7)
Time to discontinuation (days) unsatisfied

Patients at risk Missing 355 (5.2)

6,784 6,521 6,344 6,214 6,058 5,938 5,853 5,759 5,682 5,596 5,507 5,295 4,308

Camm AJ, et al. Eur Heart J. 2016;37:1145–53; Data on file

 XANTUS: Treatment-Emergent
Bleeding Events
Rivaroxaban (N=6,784)
Incidence Incidence rate,
proportion, n (%) %/year (95% CI)*
Major bleeding 128 (1.9) 2.1 (1.8–2.5)
Fatal 12 (0.2) 0.2 (0.1–0.3)
Critical organ bleeding 43 (0.6) 0.7 (0.5–0.9)
Intracranial haemorrhage 26 (0.4) 0.4 (0.3–0.6)
Mucosal bleeding# 60 (0.9) 1.0 (0.7–1.3)
Gastrointestinal 52 (0.8) 0.9 (0.6–1.1)
Haemoglobin decrease ≥ 2 g/dl‡ 52 (0.8) 0.9 (0.6–1.1)
Transfusion of ≥ 2 units of packed
53 (0.8) 0.9 (0.6–1.1)
RBCs or whole blood‡
Non-major bleeding events 878 (12.9) 15.4 (14.4–16.5)
*Events per 100 patient-years. #Numbers are for major mucosal and gastrointestinal bleeding events. ‡Representing major bleeding.
Patients could experience multiple bleeding events in different categories. RBC=Red blood cell
Camm AJ, et al. Eur Heart J. 2016;37:1145–53
 Risk Factors for Major Bleeding
XANTUS Population
Risk factors Hazard ratio (95% CI) p-value

Age (1 unit increase) 1.05 (1.02–1.07) <0.001

HF at baseline: yes vs no 1.97 (1.36–2.86) <0.001

Vascular disease: yes vs no 1.91 (1.32–2.77) <0.001

XANTUS
Concomitant AP, NSAIDs 1.80 (1.24–2.61) 0.002 Major Bleed
or paracetamol: yes vs no
Alcohol: 2.37 (1.24–4.53) 0.009
Population
heavy vs abstinent/mild
Uncontrolled 1.79 (1.05–3.05) 0.034
hypertension: yes vs no
Alcohol: medium vs 0.95 (0.62–1.46) 0.819
abstinent/mild
0.2 1 5
Modifiable risk factors Non-modifiable
risk factors
Modifiable and No risk factors
non-modifiable risk factors
AP=Antiplatelet; HF=Heart failure; NSAID=Nonsteroidal anti-inflammatory drug; RF=Risk factor

Kirchhof P, et al. Presented at the European Society of Cardiology (ESC) congress,

Rome, Italy, 27–31 August 2016
 XANTUS
Modelled Effect of Modifiable Risk Factors
XANTUS: Predicted Probability of Major Bleeding
Events According to Multivariable Model
16.2%
0.17
Predicted probability of major bleeding

0.16
0.15
0.14
0.13
0.12
0.11
0.10
0.09 8.2%
0.08
0.07
0.06
0.05
0.04 3.8%
0.03 2.1%
0.02
0.01 1.7%
0.00
0 50 100 150 200 250 300 350 400
Time to major bleeding event (days)
Kirchhof P, et al. Presented at the European Society of Cardiology (ESC) congress,
Rome, Italy, 27–31 August 2016
 Adjudicated Causes of Death
Number of patients
(N=118*), n (%)
Cardiovascular 49 (41.5)
Cardiac decompensation, heart failure 24 (20.3)
Sudden or unwitnessed death 14 (11.9)
MI 6 (5.1)
Non-haemorrhagic stroke 4 (3.4)
Dysrhythmia 1 (0.8)
Cancer 23 (19.5)
Other 16 (13.6)
Bleeding 12 (10.2)
Extracranial haemorrhage 5 (4.2)
Intracranial bleeding 7 (5.9)
Infectious disease 10 (8.5)
Unexplained 9 (7.6)

*Multiple reasons were recorded for the cause of treatment-emergent adjudicated death of some patients
Camm AJ, et al. Eur Heart J. 2016;37:1145–53
Outcomes According to Dosing (20/15 mg od)
Major bleeding, all-cause death, and thromboembolic events occurred at
higher incidence rates for the 15 mg od versus the 20 mg od dose

6,0
Events per 100 patient-years

15 mg dose
5,0
20 mg dose
Incidence rate

4,0

3,0

2,0

1,0
2,3 1,6 3,1 1,8 3,7 1,4
0,0
Stroke/SE/TIA/MI Major bleeding Mortality
28/1,410 79/5,336 39/1,410 86/5,336 46/1,410 69/5,336

Dosing decisions may have been based on other clinical considerations

besides impaired renal function
OD=Once daily
Camm AJ, et al. Eur Heart J. 2016;37:1145–53
 XANTUS: Out of Label Dosing
Associated with Less Favourable Outcomes
Composite of major bleeding, death and stroke/systemic embolism
2,5
p=0.393 p=0.012
p=0.004
2,0
HR* (95% CI) compared
with on label

1,5

1,0

0,5

0,0
On label Off label Over-dosed Under-dosed
(N=3,608) (N=856)# with 20 mg with 15 mg
(N=232) (N=583)
*Data not adjusted for baseline characteristics; #includes 41 patients who took other off-label doses

Amarenco P, et al. Presented at ESOC, 10–12 May 2016, Barcelona, Spain (Abstract ESOC6-1438)
 XANTUS: Inappropriate Dosing
Baseline Demographics
On-label Inappropriate Overdosed Under-dosed
dosing dosing with 20 mg with 15 mg
(n=3,608) (n=856) (n=232) (n=583)

Age years, mean (SD) 70.5 (9.92) 76.6 (8.71) 76.3 (7.98) 76.7 (8.87)

Weight, kg, mean (SD) 83.9 (17.3) 78.1 (17.1) 75.3 (18.0) 79.2 (16.8)

Existing comorbidities, n (%)

Hypertension 2,712 (75.2) 677 ( 79.1) 184 (79.3) 465 (79.8)

Diabetes mellitus 706 (19.6) 205 ( 23.9) 53 (22.8) 141 (24.2)

Prior stroke/SE/TIA 668 (18.5) 207 ( 24.2) 56 (24.1) 141 (24.2)

Prior myocardial infarction 359 (10.0) 101 (11.8) 35 (15.1) 62 (10.6)

Congestive heart failure 670 (18.6) 229 (26.8) 69 (29.7) 156 (26.8)

CHA2DS2-VASc score, mean (SD) 3.3 (1.7) 4.2 (1.6) 4.4 (1.6) 4.1 (1.5)

HAS-BLED score, mean (SD) 2.0 (1.0) 2.3 (1.1) 2.4 (1.0) 2.3 (1.1)

Concomitant AP/NSAID use, n (%) 683 (18.9) 165 (19.3) 38 (16.4) 113 (19.4)
Amarenco P, et al. Presented at ESOC, 10–12 May 2016, Barcelona, Spain (Abstract ESOC6-1438);
Data on file
Stroke and Systemic Emboli plus Major Bleeding
Hazard Ratios Before and After Adjustment for
Baseline Factors
Patients Incidence
Hazard ratio (95% CI) vs on-
with events, rate (95% p-value
label
n (%) CI), %/year
On-label
157 (4.4) 4.8 (4.1–5.7) 1.00
dosing
57 (6.7) 7.5 (5.7–9.8) 1.55 (1.15–2.10) 0.004
Inappropriate
dosing
1.06 (0.77– 1.45)* 0.719

Over-dosed 13 (5.6) 6.2 (3.3–10.6) 1.28 (0.73–2.25) 0.393

with 20 mg
0.85 (0.48–1.51)* 0.574

Under-dosed 39 (6.7) 7.6 (5.4–10.4) 1.57 (1.10–2.22) 0.012

with 15 mg
1.10 (0.77–1.58)* 0.589

0.2 1 5
Favours Favours
*Adjusted for baseline factors inappropriate on-label
Amarenco P, et al. Presented at ESOC, 10–12 May 2016, dosing dosing
Barcelona, Spain (Abstract ESOC6-1438)
 Comparison of Main Outcomes
XANTUS versus ROCKET AF
ROCKET AF XANTUS
4,0
3,6 CHADS2 Prior stroke*
3,5
Event per 100 patient-years

ROCKET AF 3.5 55%

3,0 XANTUS 2.0 19%
Incidence rate

2,5
2,1 2,0
2,0 1,9 1,9
1,7 1,7
1,5

1,0 0,9
0,8 0,7
0,5 0,4
0,5

0,0
Stroke/SE Major Death All strokes ICH Major GI
bleeding bleeding
Not intended for direct comparison
*Includes prior stroke, SE or TIA
Camm AJ, et al. Eur Heart J. 2016;37:1145–53
 ROCKET-AF and XANTUS
Similar Populations
Major Bleeds Death Stroke/SE
CHADS2 N
(%/year) (%/year) (%/year)
RO*1 XA5 RO2 XA5 RO3 XA5 RO4 XA5

0 n.a. 703 n.a. 0.5 n.a. 0.2 n.a. 0

1 n.a. 2,061 n.a. 1.1 n.a. 1.2 n.a. 0.5

2 922 2,035 3.4 2.7 1.03 1.9 1.2 0.6

3 3,025 1,111 3.7 2.9 1.85 2.9 1.2 1.1

4 2,073 618 3.7 2.8 1.89 2.6 2.2 2.1

5 918 222 3.2 5.4 2.86 7.3 2.6 3.0

6 122 34 5 0 2.44 3.1 3.7 0

* Refers to the safety analysis set Not intended for direct comparison
1. Patel MR, et al. N Engl J Med 2011;365:883–91; 2. Goodman SG, et al. J Am Coll Cardiol. 2014 63:891–900;
3. Data on file; 4. JnJ Briefing Book; 5. Camm AJ, et al. Eur Heart J 2016;37:1145–53
 Incidence Rates of Adjudicated
Outcomes and MAIC Rate Ratios
Weighting for CHADS2 score/gender
XANTUS
XANTUS
XANTUS (excluding ROCKET
post- MAIC rate
Outcome pre-matching CHADS2 0 and 1) AF#
matching* ratio
(N=6784) pre-matching (N=7061)
(N=2492)
(N=4020)

Major bleeding 2.10 (1.75–2.50) 2.89 (2.36–3.50) 3.10 (2.44–3.94) 3.60 (3.26–3.97)‡ 0.86 (0.67–1.12)

Stroke/non-
0.83 (0.62–1.10) 1.13 (0.81–1.54) 1.54 (1.09–2.19) 1.70 (1.47–1.96) 0.91 (0.62–1.32)
CNS SE
MI 0.44 (0.29–0.64) 0.52 (0.32–0.82) 0.75 (0.46–1.22) 0.91 (0.75–1.11) 0.82 (0.49–1.39)

Death 1.93 (1.60–2.31) 2.62 (2.12–3.20) 3.22 (2.53–4.09) 1.87 (1.63–2.14) 1.72 (1.31–2.27)

Vascular
1.00 (0.76–1.28) 1.43 (1.07–1.88) 1.83 (1.33–2.51) 1.53 (1.32–1.78) 1.19 (0.84–1.70)
death§
XANTUS Programme: >11,000 Patients
Receiving Rivaroxaban Globally
XANTUS-EL
XANTUS XANAP
Middle East, Eastern Europe,
Europe, Israel and Canada East Asia
Africa and Latin America

XANTUS
XANTUS-EL
XANAP

XANTUS pooled is the largest pre-planned, prospective, observational analysis of a single

NOAC, rivaroxaban, used for stroke prevention in patients with AF
The analysis uses combined data from three multicentre non-interventional studies enrolling >11,000
patients from 47 countries worldwide

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
Rationale for the XANTUS Programme
Differences in Clinical Characteristics of
Patients with AF Worldwide

Asia
Higher overall
Latin America stroke risk in Asian
AF-related stroke patients compared
and associated with Caucasian
morbidity is populations4
increasing, partly
due to poor control
of key risk factors1
Middle East Africa
Patients are ~10 years younger Patients have higher rates of
than in Western countries but heart failure and left
the incidence of diabetes and ventricular systolic dysfunction
obesity is higher2,3 than Western patients2

1. Massaro AR, Lip GYH, Arq Bras Cardiol 2016:doi: 10.5935/abc.20160116

2. Healey JS et al, Lancet 2017:388:1161–1169
3. Zubaid M et al, Circ Cardiovasc Qual Outcomes 2011:4:477–482
4. Sabir I et al, Nat Rev Cardiol 2014;11:290–303
 XANTUS Programme Study Design
 XANTUS, XANTUS-EL and XANAP study protocols were aligned
 Prospective, single-arm, non-interventional studies
 Major outcomes adjudicated by an independent Central
Adjudication Committee
N=11,121
Population: Data collection at initial
Rivaroxaban; visit, hospital discharge
Adult patients with a diagnosis treatment duration (if applicable) and quarterly*
of non-valvular AF newly and dose at
starting rivaroxaban treatment physician’s
for prevention of stroke discretion
or non-CNS SE
Final visit:
1 year #

Primary outcomes: major bleeding (ISTH definition), all-cause mortality, any other adverse events

*Exact referral dates for follow-up visits were not defined (every 3 months recommended); #for rivaroxaban discontinuation
after ≤1 year, the observation period ended 30 days after the last dose
Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 XANTUS Global Pooled Analysis

Screened Patients excluded

N=16,187 N=5065

Reasons patient not included in study:*

1 patient did not receive
any rivaroxaban
Primary analysis population:
defined as all patients who took
Safety population
at least one dose of rivaroxaban
• Patient decision (n=1599, 9.9%)
Enrolled • Administrative reason (n=468, 2.9%)
N=11,122 • Availability of drug (n=140, 0.9%)
• Medical guidelines (n=426, 2.6%)
• Price of drug (n=612, 3.8%)
• Medical reasons (n=502, 3.1%)
• Internal hospital guidelines (n=37, 0.2%)
• Type of health insurance (n=205, 1.3%)
N=11,121 • Other (n=1716, 10.6%)

Western Europe, Middle East,

Eastern Europe East Asia Latin America
Canada, Israel Africa
N=2577 N=2233 N=334
N=5287 N=690
(23.2%) (20.1%) (3.0%)
(47.5%) (6.2%)

*Some patients could have > 1 reason for exclusion

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 Baseline Demographics and Clinical Characteristics
XANTUS Pooled Safety Population
Rivaroxaban Rivaroxaban
(N=11,121) (N=11,121)
Age, years, mean ± SD 70.5±10.5 Prior anticoagulation therapy, n (%) 8048 (72.4)
Male, n (%) 6345 (57.1)
First available CrCl, ml/min, n (%)
Weight, kg, mean ± SD 80.0±17.8
≤80 4752 (42.7)
BMI, kg/m2, mean ± SD 28.0±5.2
>80 2320 (20.9)
>30, n (%) 2523 (22.7)
CHADS2 score, mean ± SD 2.0±1.3 Missing 4049 (36.4)

CHA2DS2-VASc score, mean ± SD 3.5±1.7 Co-morbidities, n (%)

HAS-BLED score, mean ± SD 2.0±1.1 Hypertension 8476 (76.2)
AF, n (%)
Diabetes mellitus 2484 (22.3)
First diagnosed 2049 (18.4)
Prior stroke/non-CNS SE/TIA 2372 (21.3)
Paroxysmal 4147 (37.3)
Congestive heart failure 2359 (21.2)
Persistent 1798 (16.2)
Permanent 3084 (27.7) Prior MI 994 (8.9)

Missing 43 (0.4) Hospitalization at baseline, n (%) 2072 (18.6)

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
Baseline Demographics and Clinical Characteristics
Main Differences in XANTUS by Region
Western Europe/Canada/Israel
and Middle East/Africa
Lowest stroke risk scores
Eastern Europe
Highest prevalence of hypertension,
heart failure and prior MI

East Asia
Highest prevalence of
prior stroke/TIA/non-
CNS SE

Middle East, Africa and

Latin America
Lowest bleeding risk scores Eastern Europe and
Middle East/Africa
Youngest patients

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 XANTUS Programme
Rivaroxaban Treatment Profile
XANTUS Pooled1
1,8% 20 mg 15 mg Other dose

XANTUS2 XANAP3 XANTUS-EL4

0,5% 6,4% 0,8%
25,1%

20,8%
18,9%
73,1% 49,8%
43,8%
78,7% 80,3%

● XANTUS pooled1: treatment duration (mean ± SD) was 324.5±117.80 days,

with a median of 366 days*
● XANTUS pooled1: treatment persistence at 1 year was 77.4%
*Interquartile range 330–379 days
1. Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691; 2. Camm
AJ et al, Eur Heart J 2016;37:1145–1153; 3. Kim YH et al, presented at APHRS 2016, poster number
2–116; 4. Turpie AGG et al, presented at Cardioalex 2017, Day 1, Hall A, Session 2
 XANTUS Pooled
Key Overall Outcomes of Safety Population

1,9
1,7 (1.6–2.2*)
Incidence rate (events/

2,0
(1.5–2.0*)
100 patient-years)

1,5
1,0
(0.8–1.2*)
1,0

0,5

0,0
Major bleeding Stroke/SE All-cause death

*Represents 95% confidence interval

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 Adjudicated Treatment-Emergent
Bleeding Events in Safety Population
Incidence proportion, Incidence rate, events/
n (%) 100 patient-years (95% CI)
Major bleeding 172 (1.5) 1.7 (1.5–2.0)
Fatal* 17 (0.2) 0.2 (0.1–0.3)
Critical organ bleeding 62 (0.6) 0.6 (0.5–0.8)
ICH 42 (0.4) 0.4 (0.3–0.6)
Mucosal bleeding 80 (0.7) 0.8 (0.6–1.0)
Gastrointestinal bleeding 71 (0.6) 0.7 (0.6–0.9)
Haemoglobin decrease in ≥2
58 (0.5) 0.6 (0.4–0.8)
g/dl
Transfusion in ≥2 units of
73 (0.7) 0.7 (0.6–0.9)
packed RBCs or whole blood
Non-major bleeding events 1195 (10.7) 12.8 (12.1–13.5)

Rates of fatal bleeding and ICH were low in patients treated with rivaroxaban#

*Fatal bleeding using narrow definitions (the patient experienced a treatment-emergent major bleeding event and died within 30 days of the major bleeding event and the adjudicated
primary cause of death was either ICH or extracranial bleeding; #Analyses based on 11,121 patients in the safety population

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 Adjudicated Treatment-Emergent
Thromboembolic Events in Safety Population

Incidence proportion, Incidence rate, events/

n (%) 100 patient-years (95% CI)
Thromboembolic events
179 (1.6) 1.8 (1.6–2.1)
(stroke, TIA, non-CNS SE and MI)
Stroke/non-CNS SE 98 (0.9) 1.0 (0.8–1.2)
Stroke 87 (0.8) 0.9 (0.7–1.1)
Primary ischaemic 64 (0.6) 0.6 (0.5–0.8)
Primary haemorrhagic* 20 (0.2) 0.2 (0.1–0.3)
Non-CNS SE 11 (0.1) 0.1 (0.1–0.2)
TIA 41 (0.4) 0.4 (0.3–0.6)
MI 42 (0.4) 0.4 (0.3–0.6)

Rates of stroke/non-CNS SE were low in patients treated with rivaroxaban#

*Haemorrhagic strokes and haemorrhagic transformations of ischaemic stroke were reported as both stroke and major bleeding (multiple reasons for major bleedings were possible);
#Analyses based on 11,121 patients in the safety population

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 Key Outcomes in Safety Population
Stratified by CHA2DS2-VASc Score
5
Major bleeding
Incidence rate (events/100

4 Stroke/SE
All-cause death
patient years)

0
0 1 2 3 4 5 6–9 Unknown Overall
(n=301) (n=1045) (n=2026) (n=2527) (n=2305) (n=1446) (n=1468) (n=3) (n=11,121)
CHA2DS2-VASc score
In general, rates of major outcomes increased
with higher CHA2DS2-VASc scores*
*Analyses based on 11,121 patients in the safety population

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 Cumulative Rates for Treatment-
Emergent Major Bleeding, Stroke/non-
CNS SE and Death in Safety Population
0.030 Death
Cumulative event rate

Major bleeding
0.025 Stroke/non-CNS SE

0.020

0.015

0.010

0.005

0.000
0 30 60 90 120 150 180 210 240 270 300 330 360 390 420
Time to event (days)
Number of patients at risk
Major bleeding 11121 10720 10394 10138 9823 9618 9439 9239 9091 8917 8703 8313 6734 1862 844

Stroke/SE* 11121 10729 10404 10155 9842 9637 9456 9257 9108 8939 8724 8332 6748 1864 843

Death 11121 10726 10403 10153 9847 9648 9471 9272 9125 8961 8751 8361 6772 1871 845

Over 96% of the XANTUS pooled safety population# did not experience any of the events of
treatment-emergent major bleeding, stroke/non-CNS SE or all-cause death
*Non-CNS SE; #Safety population n=11,121

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 The XANTUS Program
XANTUS, XANAP, XANTUS EL
ROCKET AF1 XANTUS2 XANAP3 XANTUS EL4
(7,111) (6,784) (2,273) (2,064)

Age (mean) 73 (median) 71.5 70.5 67

Weight (mean) 83 66 83
CrCl < 50 (%) 20.7 9.4 16.3 12.7
CHF (%) 63 19 20 31
Hypertension (%) 90 75 74 84
Diabetes (%) 40 20 27 27
Prior stroke/SE/TIA (%) 55 19 33 16
Mean CHADS2 3.5 2.0 2.3 2.05
Mean CHA2DS2-VASc 3.4 3.7 3.6
Mean Has-Bled 3 2.0 2.1 1.6
Use of 15 mg dose (%) 20.7 21 44 19
Use of 10 mg dose (%) N/A 0.4 6 0.8
1. Patel MR, et al. N Engl J Med. 2011;365:883–91; 2. Camm AJ, et al. Eur Heart J. 2016;37:1145–53; 3. Kim Y-H, et
al. Presented at APHRS, Seoul, Korea, 12–15 October 2016; 4. Data on file
 The XANTUS Program
XANTUS, XANAP, XANTUS EL
ROCKET AF1 XANTUS2 XANAP3 XANTUS EL
Events (per year)
(7,111) (6,784) (2,273) (2,064)
Major bleeding 3.6 2.1 1.5 0.9
Fatal bleeding 0.2 0.2 0.2 0.1
Critical organ bleeding 0.8 0.7 0.8 0.3
Intracranial haemorrhage 0.5 0.4 0.7 0.16
Haemoglobin drop 2.8 0.9 0.1 0.2
Transfusions 1.6 0.9 0.6 0.5
Stroke/systemic embolism 1.7 0.8 1.9 0.7
Stroke 1.7. 0.7 1.7 0.6
Ischaemic stroke 1.3 0.5% (IP) 0.9 % (IP) 0.5% (IP)
Haemorrhagic stroke 0.3 0.2% (IP) 0.4 % (IP) 0.05% (IP)
Unknown type 0.1 0 0.2 % 0.1% (IP)
Systemic embolism < 0.1 0.1 0.1 0.1
Myocardial infarction 0.9 0.4 0.5 0.3
Serious adverse events 18 15.6 8.3
All cause death 1.9 1.9 2.0 1.7
1 year persistence 80% 66% 82%
IP=Incidence proportion
1. Patel MR, et al. N Engl J Med. 2011;365:883–91; 2. Camm AJ, et al. Eur Heart J. 2016;37:1145–53; 3. Kim Y-H, et
al. Presented at APHRS, Seoul, Korea, 12–15 October 2016; 4. Data on file
 XANTUS Pooled
Key Differences in Outcomes by Region

Proportion of patients free from major bleeding or

stroke/non-CNS SE consistent between regions (95.7–97.4%)

Major bleeding Stroke/non-CNS Mortality

Highest in Western SE Highest in the
Europe/Canada/Isra Highest in East Asia Middle East/Africa
el vs other regions vs other regions and Latin America
(2.3 vs 0.7–1.6 (1.8 vs 0–1.0 vs other regions
events/ events/ (2.7 vs 1.5–2.0
100 patient-years) 100 patient-years) events/
100 patient-years)
 Rivaroxaban Is Highly Effective and
Provides a Beneficial Safety Profile
in the Real World
Pooled Baseline XANTUS Pooled* Pooled
Effectiveness* CHADS2,
mean
2.0 Safety*
4 4

(events/100 patient-years)
(events/100 patient-years)

Age, years,

On-treatment event rate

On-treatment event rate

70.5
mean
3 3
Heart failure 21%

1,9
2 Hypertension 76% 2 1,7

1,0 Diabetes 22%

1 1 0,7
0,4
Prior stroke# 21%
0 0
Stroke/SE Death Prior MI 9% Major bleeding ICH GI bleeding

*Results based on safety population n=11,121; #includes prior stroke, SE or TIA

Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
 Strengths and Limitations
The XANTUS Programme
 Strengths
 Large sample size
 Prospective design
 Independent endpoint adjudication, to minimize
reporting bias in the study
 Limitations
 Single-arm, open-label study
 Selection bias
 Limited influence on data completeness in
observational setting
 Outcomes not adjusted for baseline risk factors
 XANTUS Summary
 The XANTUS pooled study is a unique, pre-planned, prospective,
observational analysis of a single NOAC, rivaroxaban, for stroke
prevention n AF1
 Includes combined data from 3 studies (11,121 patients) receiving
rivaroxaban across 47 countries
 Key outcomes centrally adjudicated by an independent committee to
minimize bias
 Over 96% of patients did not experience any of the events of major
bleeding, stroke/non-CNS SE or all-cause death
 Results were consistent across different patient populations worldwide1
 The XANTUS international study programme offers a unique RWE
dataset
 Results were consistent with the phase III ROCKET AF study2

1. Kirchhof P et al, presented at the European Society of Cardiology 2017, abstract 86691
2. Patel MR et al, N Engl J Med 2011:365:883–891
 General Conclusions
 RCT data are indispensable (the gold standard) for drug approval
 Real world data exist, should be evaluated, and may strengthen,
extend or challenge data derived from RCTs
 Real world data obviously enlarge the safety information relating
to a particular drug or strategy
 Effectiveness can be gauged to only a limited extent
 Use of the drug in clinical practice can only be assessed by the
analysis of real world data: indication, dose, adherence,
persistence, physician management, etc.
 Comprehensive understanding of the value of a specific therapy
should be assessed using RCT and real world evidence,
appropriately weighted for its likely accuracy
Thank you for your attention
Phase
III/IV
RCTs

Data
Sources
Pro- Retro
spective spective
Registries Adminis-
including Real World
trative Data-
PMSS Evidence
bases
Thanks for your participation

Remember to complete the

Questionnaire obtaining a score ≥ 8
points out of 10

For downloading the material of the

Conference on PDF format, return to the
course visual and look for the section
“Slides Set”