First-line Cryoablation for Early Treatment of Persistent Atrial Fibrillation

The primary goal is to evaluate if early pulmonary vein isolation (PVI) performed with the Arctic Front cryoballoon as first-line therapy is superior to antiarrhythmic drugs (AAD) in preventing atrial arrhythmia recurrences in patients with persistent atrial fibrillation (AF).

The main secondary goal is to evaluate the impact of early invasive intervention on health related quality of life (HRQOL) and symptoms, and on safety in comparison to primary AAD therapy, using generic and disease-specific HRQOL questionnaires and also assess Quality Adjusted Life Years (QALYs) score and EHRA classification of symptoms.

The third goal is to assess the impact of an early intervention on cardiovascular health care use and its relation to AF/AT burden and to assess treatment burden and cost-effectiveness compared to AAD.

Study Design: Multicentre, prospective 1:1 randomized open blinded for evaluation of end point (PROBE) controlled parallel-group superiority trial comparing first-line pulmonary vein isolation (PVI) using the cryoballoon (Arctic Front AdvanceR, Medtronic) and first-line antiarrhythmic drug (AAD) therapy. Patients will be randomized 1:1 to first-line PVI using the cryoballoon or to first-line AAD therapy. A 1:1 block randomization immediately after implantation of an implantable cardiac monitor (ICM ) before run-in.

After fulfilling inclusion criteria and no exclusion criteria, an ICM (Reveal Linq™, Medtronic, Inc., Minneapolis, MN) will be injected subcutaneously for continuous rhythm monitoring during a 2 months "run-in" period prior initiation of allocated treatment, and then during 3 years of follow up. A 6 minutes detection cut-off for AF will be used. Patients and physicians will be blinded to the ICM result. Three month blanking period after treatment will not be included in the analysis. SAEs detected via ICM will be managed independently from the blinding of AF. If the ICM has to be explanted the patient should continue in the study with alternative monitoring tool as defined.

The trial will be performed at 4-5 university centres in Sweden, at 1-2 centres in United Kingdom (UK) and at 4-5 centres in Europe, all with experience in Cryoballoon AF ablation. Each centre will register at least 20 -30 patients for a total of 220 patients.

AAD therapy prescribed by the local investigators and collected by the subjects at a pharmacy as in normal clinical practice. They should be initiated with either dronedarone or flecainide as first options. Dronedarone:- 400 mg twice daily or Flecainide:- (50-)100 (-200) mg twice daily or slow release (100-)200 mg once daily. The 3-month "blanking period" will allow for drug titration and optimization. A change to the other of the two first option AADs guided by lack of efficacy defined by AF related symptoms or intolerable side effects. Thereafter, other AADs may be tested in order; propafenone and sotalol as long as the medication changes is within the 3 months blanking period: Propafenone:- 150 mg three times daily increasing to 300 mg twice daily, if necessary max 300 mg three times daily. Sotalol:- 80 mg twice daily up to 160 mg twice daily. Once the blanking period has ended, a change in medication will be defined as clinical endpoint, i.e. treatment failure.

Surveillance of proarrhythmia risk by analysing ECG for prolonged corrected QT interval (QTc), widened QRS-complex, and prolonged PR interval at each follow-up. ECG monitoring on days 1-3 in patients prescribed flecainide, propafenone, or sotalol is recommended. Cardioversion should be performed as soon as possible with adequate protection against thromboembolism if at risk.

Physicians advised to keep patients in the same treatment arm during follow-up. If symptoms worsen despite AAD therapy due to limited efficacy at adequate dosages, the patient may, if requested, undergo AF ablation with PVI following preferably a minimum of 12 months AAD therapy.

Catheter ablation with PVI should follow clinical routine. PVI alone will be performed using a 28 mm cryoballoon with an Achieve circular mapping (ACM) catheter for recording of pulmonary vein potentials. Cryoballoon application for 4 minutes in each vein guided by disappearance of PV potentials or by reduction of temperature to at least minus 40 C degrees within the first 120 seconds if no PV potentials can be visualized. After ablation of all pulmonary veins, entrance block during sinus rhythm should be assessed with PV entrance block documented from each PV using a circular mapping catheter.

Biomarkers s100b, troponin-T (TnT) and NT-pro-BNP (brain natriuretic peptide) will be obtained from peripheral venous blood in the lab on the day of the procedure (baseline) before placing the catheters in the heart and again after all ablation applications after the transseptal catheter has been withdrawn. All three markers are thereafter sampled a third time from peripheral blood after 6-18 hours, and NT-pro-BNP at 3, 12, 24, 36 months follow up.

Follow-up Re-ablations may be performed at earliest 3 months after the prior AF ablation procedure if symptoms persists or recurs provided an AF/AT is ECG documented. PVI alone will be performed using a cryoballoon with a circular mapping catheter for assessing entrance conduction block. Patients may be offered a 3rd AF ablation. If patient declines a repeat intervention despite symptomatic recurrence of AF, the patient may be offered AAD, and will be defined as failed ablation unless a beta-blocking agent alone is prescribed. If all PVs are isolated at a repeat ablation procedure, other lesions may be performed except for posterior wall isolation unless pulsed field ablation is used, but treatment will be defined as failed PVI.

The study duration is 3 years with 18 months' enrolment period. Patients will be followed at three, six, nine, 12, 18, 24, 30 and 36 months after the ablation procedure or start of AAD therapy. A nurse telephone visit will be scheduled at 6, 9, 18 and 30 months for recording of clinical events, medication and adverse events. All endpoints will be evaluated at last visit before cross-over.

A clinical event is any event related to a cardiovascular disease or AF/AT; such as acquisition of a new comorbidity, new risk factor that results in increased bleeding risk, new cardiovascular intervention such as cardioversion, angiography, bypass surgery etc but which is not defined as an AE.

All "Serious Adverse Events" (SAE) will be documented on the "Adverse events Form" AND "Serious adverse events Form" and reported to an independent Clinical Events Committee. All deaths events will be documented on "Death report forms".All-cause death will be classified as either Non-cardiovascular (including unknown, excluding sudden death or Cardiovascular death (Cardiac (sudden (including arrhythmic, myocardial infarction) vs non-sudden) or Vascular) Statistical analysis Safety - All randomized patients will be included in the safety analysis. Only observed observations are used in the safety analysis.

Modified Intention to treat (mITT) - All randomized patients who receive treatment. The main analysis will be performed on the mITT-population.

Per protocol (PP) - All randomized patients completing the study treatment period of 12 months without any major protocol violation (for example ineligibility, early withdrawals, poor compliance). The PP population will be defined at clean file.

Missing Data; The Primary variable will be imputed using a "worst case imputation" for the Logistic Regression if the magnitude of the missing data is small, say, approximately 5%. The Statistical Analysis Plan will include an outline for the multiple imputation strategy to be used for the situation with a larger amount of missing primary endpoints.

The hypothesis generating analyses of Secondary Variables will use a simplistic Last Observation approach in case the amount of missing data is limited, say, less than 5%. The Statistical analysis Plan (SAP) will include an outline for the multiple imputation strategy to be used for the situation with a larger amount of missing secondary endpoints.

Subgroups and explorative analyses; Explorative analyses may be performed to investigate relationships between treatments and endpoints, and for predictive factors for AF recurrences.

Statistical methods; The Primary variable, freedom from atrial tachyarrhythmia at least 12 months with 90-day initial blanking period, will be analyzed using a Logistic Regression adjusted for the following additional covariates (baseline values): Coronary artery disease, Hypertension and left atrial volume index. Graphical methods such as Kaplan-Meier plots will be used to visualize the treatment effect in the full study population: "Freedom of tachyarrhythmia up to 12 months with 90-day blanking period".

The main analysis population will be a modified Intention to Treat population: all treated patients as randomized. The number of randomized subjects who are not treated are assumed to be at most a handful. The Per Protocol population will be defined at the Clean File Meeting and will be used for sensitivity analysis for the analysis of the primary variable using the same methods as for the main analysis.

The set of secondary variables will not be adjusted for multiplicity. All findings among secondary variables in main population viewed as exploratory and hypothesis generating. The statistical Methods for each of the Secondary Endpoints will be described in detail in the Statistical Analysis Plan. All continuous variables will be presented per treatment group using descriptive statistics by mean, standard deviation (SD), max and min values, in addition medians, 25th and 75th percentiles will be presented when suitable. The mean difference between treatment groups will be presented with 95% Confidence Intervals (CIs).

Sample Calculation: Study randomized 1:1: Proportion of freedom from atrial tachy-arrhythmia before 12 months with 90-day blanking period estimated to reach 55% in Ablation group based on previous cryoballoon AF ablation study of persistent AF patients monitored by ICM versus 30% in Drug arm, based on 25% difference in freedom from AF between treatment groups in a first line treatment study of paroxysmal AF patients. The 40% freedom from AF postablation in persist AF patients judged too pessimistic for a first line trial. Three prior AAD trials evaluating drug efficacy in persistent AF patients post cardioversion, using intermittent recordings, reported freedom from AF ranging between 30 and 42 %. Recognising that ICM would detect more AF episodes, the freedom from AF post cardioversion on AAD at 12 months is therefore estimated to be in the lower range, i.e. 30 %, in this trial.

A 2 minutes AF detection cut-off has a lower positive predictive value versus a 6 min cut-off episode duration, which will decrease number of total episode count and false positive detection without neglecting relevant clinical information, keeping AF-burden unaffected. It therefore seems rational to use 6 minutes as AF episode cut-off, which is anticipated to at most increase the freedom from AF by 5% in both treatment groups but without affecting the relative treatment differences in the 2 groups.

To have a 90% chance (i.e. power=90%) at the 5% significance level of detecting an improvement in the primary outcome from 30% in the control group to 55% in the ablation group, a total of 156 patients are needed. Adjusting for an assumed cross over rate of approximately 10% in the drug arm and 5% in the Ablation arm the total sample size will be 220.

Treatment allocation sequence generated in Statistical Analysis Software (SAS) or R by the Biostatistics section at Uppsala Clinical Research Centre (UCR) by using permuted blocks and 1:1 allocation, stratification by center and type of persistent AF. Patients randomized using the Interactive Web Response System (IWRS) function in Viedoc.