ANTIARRHYTHMIC DRUGS IN ATRIAL FIBRILLATION: STILL A CHALLENGE ROLE OF CLASS III ANTIARRHYTHMIC DRUGS REVISITED J.Y. LE HEUZEY Georges Pompidou Hospital René Descartes University, Paris

Bologna, February 17, 2017 Disclosure

Consultant / Conferences / Advisory Board fees from Astra Zeneca, Sanofi, BMS/Pfizer, Meda, Boehringer Ingelheim, Bayer, Novartis, Servier and Daiichi Sankyo. 1- 9- 2- 10- 3- 11- NIFERIDIL 4- BUDIODARONE 12- SEMATILIDE 5- 13- 6- 14- 7- 15- 8- 1- AMIODARONE 9- IBUTILIDE 2- AZIMILIDE 10- NIFEKALANT 3- BRETYLIUM 11- NIFERIDIL 4- BUDIODARONE 12- SEMATILIDE 5- BUNAFTINE 13- SOTALOL 6- CELIVARONE 14- TEDISAMIL 7- DOFETILIDE 15- VERNAKALANT 8- DRONEDARONE 1- AMIODARONE 9- IBUTILIDE 2- AZIMILIDE 10- NIFEKALANT 3- BRETYLIUM 11- NIFERIDIL 4- BUDIODARONE 12- SEMATILIDE 5- BUNAFTINE 13- SOTALOL 6- CELIVARONE 14- TEDISAMIL 7- DOFETILIDE 15- VERNAKALANT 8- DRONEDARONE . Pharmacological cardioversion : Amiodarone, Vernakalant

. maintenance (rhythm control strategy) : Amiodarone, Dronedarone, Sotalol

NEW ! Combination Dronedarone + Potassium currents blocked by vernakalant, and potency

VERNAKALANT

Current IC50 (µM) Ito I 5-30 0 mV to IKur IKur 3-13 IKr IKACh 10 I 7-21 IKACh Kr

IKs > 100 0 100 200 300 400 ms IK1 > 100

AAD: Anti-arrhythmic drug IC: Inhibitory concentration Fedida et al. J Card Electrophysiol 2005, p1227-1238 Effect on AERP vs VERP

. In an open-label electrophysiologic study in 19 patients, vernakalant significantly prolonged atrial, but not ventricular, effective refractory period – Low-dose vernakalant: 2 mg/kg over 10 min + 0.5 mg/kg/h for 35 min (total dose, ~2.25 mg/kg) – High-dose vernakalant: 4 mg/kg over 10 min + 1 mg/kg/h for 35 min (total dose, ~4.5 mg/kg)

Paced cycle length: 600 msec 50 (31 ± 14)a

Low Dose 40 High Dose

30 (14 ± 14)a

20 (4 ± 15)

(0 ± 10) 10

Change Change From Baseline, msec 0 aP<0.05 vs baseline. AERP VERP AERP=atrial effective refractory period; VERP=ventricular effective refractory period. Dorian P et al. J Cardiovasc Pharmacol. 2007;50:35–40. Adapted with permission from Dorian P et al. J Cardiovasc Pharmacol. 2007. 8 ACT Studies (Phase III) Key Inclusion and Exclusion Criteria

Inclusion criteria Exclusion criteria . ACT-I, ACT-III and ACT-IV studies . Female patients pregnant or nursing – Age ≥ 18 years . ECG findings: – Uncorrected QT interval >0.440 – Sustained (> 3 hrs up to 45 days) AF seconds or QRS >0.140 seconds ACT-III included patients with AF – (<50 beats/min) or sick or atrial flutter sinus syndrome, unless controlled by – Hemodynamically stable (systolic a pacemaker BP >90 to <160; diastolic BP <95 mm . Cardiovascular disease: NYHA class IV Hg) failure, acute coronary syndrome, myocardial infarction, or cardiac surgery – Adequate anticoagulant therapy within 30 days before enrolment – Weight 45 to 136 kg . Other: – Patients stratified in – Failed direct-current cardioversion Short-duration AF (3 hours to 7 – IV antiarrhythmic therapy within 24 days) (primary endpoint) hrs Long-duration AF (7-45 days) – Reversible cause of AF – Previously failed electric conversion . ACT-II study – toxicity – End-stage disease, serious hepatic or – Sustained AF or atrial flutter for 3 to renal disease 72 hours occurring between 24 hours and 7 days after a cardiac – Uncorrected electrolyte imbalance surgery – History of Torsade de Pointes (ACT II) Roy D et al. Circulation. 2008;117:1518-25; Kowey PR et al. Circ Electrophysiol 2009;2:652-9. ACT Studies (Phase III) Primary Efficacy Endpoint

60%

51.7% 47.0% 51.2% 50.9% 50%

Vernakalant

40% Placebo

30% p<0.001 compared to placebo in ACT I, 20%

sinus rhythm (%) rhythm sinus 14.0% ACT II, and ACT III

10% Percent conversion of AF AF to of conversion Percent 4.0% 3.6%

0% ACT I ACT II* ACT III ACT IV

*ACT II enrolled patients with AF duration 3-72 hours. Roy D et al. Circulation. 2008;117:1518-25; Kowey PR et al. Circ Arrhythmia Electrophysiol 2009;2:652-9. ACT Studies (Phase III) Other Findings

. In those patients who converted to sinus rhythm: – Conversion was successful with the first dose in ~75% – 97.2% of patients remained in sinus rhythm at 24 hours (pooled analysis of ACT I and ACT III studies)

. The median time to conversion to sinus rhythm with vernakalant was 11 minutes (pooled analysis of ACT I and ACT III studies) . Vernakalant was not

successful in % conversion to sinus rhythm terminating primary atrial flutter Time (minutes) AVRO Study Design

R n=116 • Phase 3, multicenter, randomized, double- A blind, active-controlled trial N Vernakalant • Key inclusion criteria: D – Age 18 to 85 years O – Symptomatic AF of 3 to 48 hours M duration I – Eligible for cardioversion Amiodarone Z E n=116

. Primary endpoint: the proportion of patients in SR at 90 minutes after initiating therapy.

12 60 51.7 50 40 30 AVRO 20 10 5.2 0

Vernakalant (N=116) Amiodarone (N=116)

Vernakalant (N=116) 0.6 Amiodarone (N=116)

Proportion P<0.0001 (Log-rank test) of patients 0.4 Median time to conversion with in vernakalant responders conversion 0.2 was 11 minutes to SR

0.0 0 5 10 20 35 50 70 90 Time (Minutes) AVRO Safety Summary

. There were more adverse events in the vernakalant group than the amiodarone group . Most common side effects of vernakalant were dysgeusia (6.9%), sneezing (3.4%) and nausea (2.6%) . One death (pulmonary embolism) in vernakalant group occurred on Day 24 . Cardiac arrest in amiodarone group occurred at 37 minutes after start of first infusion . Ventricular arrhythmia, hypotension and bradycardia were rare and occurred at similar rates between vernakalant and amiodarone groups . There were no cases of TdP or sustained VT . More patients experienced AFL in the vernakalant group (8.6%) than the amiodarone group (0.9%) . There was no 1:1 conduction of AFL

. EMA labelling : rapid conversion of recent onset atrial fibrillation to sinus rhythm in adults - for non surgery patients : atrial fibrillation ≤ 7 days duration - for post cardiac surgery patients : atrial fibrillation ≤ 3 days duration.

. 3 mg/Kg over 10 minute period

. Caution +++ in patients with structural heart disease

16

Cardioversion of recent onset of atrial fibrillation

Recent onset AF

Yes Haemodynamic No

instability? ESC 2016 GUIDELINES

Elective

Patient choice Urgent

Pharmacological cardioversion

Coronary artery Severe HFrEF, disease, moderate No relevant significant HFrEF or structural aortic stenosis HFmrEF/HFpEF, heart disease abnormal LVH

Electrical Intravenous Intravenous Intravenous Pill in the pocket cardioversion (IB) Amiodarone (IA) Vernakalant (IIbB) (IA) Flecainide (IIaB) Amiodarone (IA) Ibutilide (IIaB) (IIaB) Propafenone (IA) Vernakalant (IA) www.escardio.org/guidelines European Heart Journal - doi:10.1093/eurheartj/ehw 210 BENZOFURANES

Amiodarone Budiodarone Celivarone Dronedarone

DAFNE 2003 ADONIS 2007 EURIDIS 2007 ERATO 2008 ANDROMEDA 2008 ATHENA 2008 DIONYSOS 2010 PALLAS 2011 Dronedarone Showed a Significant Reduction in First AF Recurrence in Combined Analysis Amiodarone

EURIDIS/ADONIS

0.8

Dronedarone 0.7 0.6 25% 0.5 reduction in relative 0.4 risk

0.3 0.2 Hazard ratio, 0.75 (95% CI, 0.65 to 0.87)

Cumulative Incidence Cumulative p<0.001 0.1 0 0 60 120 180 240 300 360 Time (days) Placebo + standard therapy Dronedarone 400mg bid + standard therapy

Singh BN, et al. N Engl J Med. 2007;357:987-99. 18 Cumulative Incidence of All-cause Mortality

0.8 Placebo Dronedarone 800mg n=317 n=310 0.7 Number of patients who died 12 25

Hazard Ratio 2.13 0.6 95% CI [1.07; 4.25] Log rank p value 0.03 0.5

0.4 Placebo 0.3 DR 400mg bid

0.2 Cumulative Incidence Cumulative 0.1 Time (days) 0.0 0 30 60 90 120 150 180 210

Patients at risk: Placebo 317 256 181 103 50 18 6 1 DR 400mg bid 310 257 174 104 59 22 5 1

Køber L, et al. N Engl J Med. 2008;358:2678-87. 19 Dronedarone Significantly Decreased Risk of CV Hospitalisation or Death by 24%

50 Placebo on top of standard therapy* 24% reduction

DR 400mg bid on top of standard therapy* in relative risk 40

30

20

HR=0.76

10 p<0.001 Cumulative (%) Incidence Cumulative

0 Months 0 6 12 18 24 30

Patients at risk: Placebo 2327 1858 1625 1072 385 3 DR 400mg bid 2301 1963 1776 1177 403 2 * Standard therapy may have included rate control agents (beta-blockers, and/or Ca-antagonist and/or digoxin) and/or anti-thrombotic therapy (Vit. K antagonists and /or and other antiplatelets therapy) and/or other cardiovascular agents such as ACEIs/ARBs and statins. Mean follow-up 21 ±5 months. Hohnloser SH et al. N Engl J Med 2009;360:668-78. 20 DIONYSOS Primary Endpoint: More AF Events But Less Early Discontinuation With Dronedarone

1.0

0.8 184 (73.9%) 43 endpoints 0.6 42 endpoints 141 (55.3%)

0.4

0.2 RRR (95%CI) = 1.589 (1.275;1.98)

Cumulative incidence Cumulative p-value <0.001 0.0 0 3 6 9 12 15 Months Dronedarone Patients at risk Amiodarone 249 99 84 40 12 0

255 146 126 61 13 0

Dronedarone Amiodarone (n=249) (n=255) Number of patients with endpoint 184 (73.9%) 141 (55.3%) ECG documented AF endpoint 158 (63.5%) 107 (42.0%) Documented AF after conversion 91 (36.5%) 62 (24.3%) Unsucessful electrical cardioversion 29 (11.6%) 16 (6.3%) No spontaneous conversion and no electrical 38 (15.3%) 29 (11.4%) cardioversion on day 10 to day 28 Premature study drug discontinuation 26 (10.4%) 34 (13.3%) Lack of efficacy 1 (0.4%) 0 Intolerance 25 (10.0%) 34 (13.3%) Le Heuzey J.Y. et al., J. Cardiovasc. Electrophysiol. 2010; 21 : 597 - 605 Stroke, systemic embolism, myocardial infarction or cardiovascular death

First Co-primary Dronedarone vs placebo Dronedarone Placebo Outcome HR and 95% CI

43 (2.7%) 19 (1.2%) 2.29 (1.34 – 3.94) p=0.002 0.05

Dronedarone 0.04 Placebo

0.03

0.02

Cumulative Incidence 0.01

Days 0.00 0 30 60 90 120 150 180 Number at risk :

Dronedarone 1619 1421 930 353

Placebo 1617 1445 908 377 Baseline Patient Characteristics

Placebo Dronedarone All patients n=2327 n=2301 n=4628

Age (mean ±SD, years) 71.7 ±9.0 71.6 ±8.9 72 ±9.0 <65yr 442 (19.0%) 431 (18.7%) 873 (18.9%) 65 to 75yr 907 (39.0%) 923 (40.1%) 1830 (39.5%) ≥75yr 978 (42.0%) 947 (41.2%) 1925 (41.6%) Female gender 1038 (44.6%) 1131 (49.2%) 2169 (46.9%) AF/AFL at baseline 586 (25.2%) 569 (24.7%) 1155 (25.0%) Structural heart disease 1402 (60.9%) 1330 (58.3%) 2732 (59.6%) 1996 (85.8%) 1999 (86.9%) 3995 (86.3%) Coronary heart disease 737 (31.7%) 668 (29.0%) 1405 (30.4%) Valvular heart disease 380 (16.3%) 379 (16.5%) 759 (16.4%) Non-ischemic cardiomyopathy 131 (5.6%) 123 (5.3%) 254 (5.5%) History of CHF NYHA II/III 515 (22.1%) 464 (20.2%) 979 (21.2%) LVEF <0.45 285/2281 (12.5%) 255/2263 (11.3%) 540/4544 (11.9%) LVEF <0.35 87/2281 (3.8%) 92/2263 (4.1%) 179/4544 (3.9%) Lone atrial fibrillation 139 (6.0%) 140 (6.1%) 279 (6.0%) Pacemaker 243 (10.4%) 214 (9.3%) 457 (9.9%)

Hohnloser SH, et al. J Cardiovasc Electrophysiol 2008;19:69-73. 23 Baseline Characteristics

Dronedarone Placebo N=1619 N=1617

Age years mean (SD) 75.0 (5.9) 75.0 (5.9)

Duration of permanent AF > 2 years 1119 (69.1%) 1124 (69.5%)

Coronary artery disease 661 (40.8%) 666 (41.2%)

Peripheral arterial disease 187 (11.6%) 213 (13.2%)

Prior Stroke or TIA 436 (26.9%) 458 (28.3%)

History of heart failure 1139 (70.4% ) 1117 (69.1%)

Left ventricular ejection fraction ≤ 40% 345 (21.3%) 335 (20.7%)

Baseline use of a Beta-blocker 1201 (74%) 1201 (74%)

Baseline use of Vitamin K antagonist 1359 (84%) 1363 (84%) 24 Dronedarone : do not cross the red line !

Remodeling Ventricular dilation MI HF

End-stage Atherosclerosis microvascular and LVH heart disease

Risk factors (diabetes, Atrial fibrillation Death hypertension)

ESC Eur.2010 Heart GUIDELINES J. 2010, 31 : 2369 - 469 28 Initiation of long term rhythm control therapy in symptomatic patients with atrial fibrillation

Initiation of long term rhythm control therapy to improve symptoms in AF

No or minimal signs Coronary artery disease, for structural heart significant valvular heart Heart failure disease disease, abnormal LVH

Patient choice Patient choice Patient choice

Dronedarone (IA) Catheter Dronedarone (IA) Catheter Amiodarone Catheter Flecainide (IA) ablation Sotalol (IA) Ablation (IA) Ablation Propafenone (IA) (IIaB) Amiodarone (IA) (IIaB) (IIaB) Sotalol (IA)

ESC 2016 GUIDELINES www.escardio.org/guidelines European Heart Journal - doi:10.1093/eurheartj/ehw 210

SAFE - T

B.N. Singh et al., NEJM 2005 ; 352 : 1861 - 72

ANTIARRHYTHMICS FOR MAINTAINING SINUS RHYTHM AFTERFig CARDIOVERSIONure 5. Atrial fibrillation recu rOFrence ATRIAL FIBRILLATION : RECURRENCE

Lafuente-Lafuente C. et al., Cochrane Database Syst. Rev. 2012; 5 : CD005049

Substantial heterogeneity (I2 = 76%, P = 0.02) between studies was detect ed for dofetilide. Moderate, non-significant inconsis- with flecainide, five with dofetilide and propafenone, eight with tency (I2 = 52%, P = 0.15) appeared between the two studies and sotal ol, 10 with dronedar one and and on metoprolol. In both cases, dofetilideand metoprolol, all st ud- 17 with azimilide(the95% CI varied between 2 and 60). iesshowed thesame direct ion of theeffect (i.e. to a reduct ion of In direct comparisons between antiarrhythmics (Analysis 4.5), atrial fibrillation recurrences) and the heterogeneity seemsprob- amiodaronereduced recurrencesof atrial fibrillation significantly ably caused by the differences in the charact eristics of recr uited more than combined cl ass I drugs (OR 0.36, 95% CI 0.26 to patients. 0.50, P < 0.00001), more than dronedarone (OR 0.45, 95% CI Pooled recurrence ratesof atrial fibrillation at oneyear werehigh: 0.31 to 0.63, P < 0.00001, results based in one single trial, 504 69% to 84% in controlsnot receiving antiarrhythmic treatment, patients), and morethan sotalol (OR 0.43, 95% CI 0.33 to 0.56, reduced to 43% to 67% in patientstreated with antiarrhythmics. P < 0.00001). No other significant difference appeared in com- The corresponding average NNT for one year, to avoid one re- parisonsbetween antiarrhythmics. currence of atrial fibrillation, were three with amiodarone, four Resultsfor atrial fibrillation recurrence wereunchanged in sensi-

Antiarrhythmics for maintaining sinus rhythm after cardioversion of atrial fibrillation (Review) 14 Copyright © 2012 The Cochrane Collaboration. Published by John W iley & Sons, Ltd. ATMA - SIDE EFFECTS (I) DYSFUNCTION

Hypothyroidism Hyperthyroidism

A (%) P (%) A (%) P (%) EMIAT 7.1 1.3 2.3 1.1

CAMIAT 15.7 1.7 0.7 0.7

GEMICA 0.1 0 0 0

CHFSTAT 9.5 2.1 2.1 0.1

PAT 0 0 3.6 0.3

CAMIAT P 0 0 0 0

All studies 7.0 1.1 1.4 0.5 (2580 vs 2545)

(181 vs 27) (37 vs 13)

ATMA investigators, Lancet 1997; 350 : 1417-24 ATMA - SIDE EFFECTS (II)

Peripheral Lung infiltrate Bradycardia Liver function neuropathy A (%) P (%) A (%) P (%) A (%) P (%) A (%) P (%)

EMIAT 0.1 0 0.3 0.3 1.5 0 0.8 0.3

CAMIAT 1.0 0.2 3.1 1.1 1.3 0.8 1.0 0.3

GEMICA 0 0 0.2 0 3.5 2.3 0 0.2

CHFSTAT 1.5 0.9 4.8 0.9 0.9 0 2.4 0.3

PAT 0 0 0.3 0 7.2 0.3 0.7 0.6

CAMIAT P 0 0 6.3 0 0 3.4 8.3 3.4

All studies 0.5 0.2 1.6 0.5 2.4 0.8 1.0 0.4 (2580 vs 2545)

(12 vs 4) (42 vs 12) (44 vs 19) (26 vs 9)

ATMA investigators, Lancet 1997; 350 : 1417-24 Le Heuzey J.Y. et al., J.Cardiovasc. Electrophysiol. 2010; 21 : 597 - 605 ANTIARRHYTHMICS FOR MAINTAINING SINUS RHYTHM AFTER FigCARDIOVERSIONure 3. Overall mortality OF ATRIAL FIBRILLATION : OVERALL MORTALITY

Lafuente-Lafuente C. et al., Cochrane Database Syst. Rev. 2012; 5 : CD005049

Quinidine, compared with controls, showed anon-significant but cl ear trend to increasemortality (OR 2.26, 95% CI 0.93 to 5.45, . P = 0.07). Thistrend becamesignificant if missing patientswere Sotalol also showed a significant increase in associ ated mortality counted as deaths (OR 2.29 95% CI 1.05 to 5.01, P = 0.04), compared with controls (OR 2.47, 95% CI 1.21 to 5.05, P = and when all drugsof cl assIA (quinidineand ) were 0.01). Thisincreasewasconfirmed In all sensitivity analysis, ei ther combined (OR 2.39, 95% CI 1.03 to 5.59, P = 0.04) (Analysis counting missing patients as deaths (OR 2.14, 95% CI 1.40 to 1.5). Thecorresponding NNH for combined cl assIA drugswas 3.25, P=0.0004), poolinghigh-qualitytrialsonly(OR2.78, 95% 109 patientstreated for oneyear to haveoneexcessdeath, witha CI 1.00 to 7.69, P = 0.05) or pooling trialswith more than 200 wide95% CI of 34 to 4895 patients. patientsonly (OR1.97, 95% CI 1.03 to 3.75, P=0.04) (Analysis However, sensitivity analysisof studieson quinidine and cl assIA 1.11 an d Analysis1.12). Thecorresponding NNH for sotalol was drugs, select ively pooling trialswith adequate allocation conceal- 166 patientstreated for 1 year to have1 excessdeath, the95% CI ment or those incl uding more than 200 patients, left only two being also wide: 61 to 1159 patients. st udies (PA FAC 2004; SO PAT 2004) in which no difference in A strongbut non-significant trendtoincreased mortalityappeared mortality compared with controlswasapparent. Thesetwo trials also with azimilide compared with controls (OR 2.18, 95% CI employed a lower dose of quinidine (320 to 480 mg/day) than 0.98, 4.89, P=0.06). Thistrendpersistedin all sensitivityan al ysis. other studies(800to1800mg/day), andcombinedquinidinewith Very little data on mortality was found on cl ass IC drugs. We

Antiarrhythmics for maintaining sinus rhythm after cardioversion of atrial fibrillation (Review) 10 Copyright © 2012 The Cochrane Collaboration. Published by John W iley & Sons, Ltd. Freemantle N. et al. Europace 2011;13 : 328 - 45

AMIODARONE AND THE RISK OF CANCER

Su V.Y.-F. et al. Cancer, 2013 ; 119 : 1699 – 705

CONCLUSIONS

1- For pharmacological cardioversion, Vernakalant is quickly (11 minutes) efficient in more than 50% of patients 2- For sinus rhythm maintenance Sotalol and Amiodarone can be used : Amiodarone is the most effective but their safety remains questionable, especially long term safety for Amiodarone 3- The efficacy of Dronedarone alone remains modest, the drug is contra-indicated in case of heart failure and permanent atrial fibrillation 4- The combination of Dronedarone and Ranolazine can have a potential interest but future developments are needed before an use in routine CLASS III ANTIARRHYTHMIC DRUGS

CLASS III + NON ANTIARRHYTHMIC

…arone …ilide …kalant