Highlights
-
•
CATALYST compares Amulet LAAO vs. NOACs in AF patients at elevated stroke risk.
-
•
Three co-primary endpoints assess stroke, systemic embolism, CV death, and bleeding.
-
•
Post-LAAO care includes short DAPT and TEE/CT imaging for contemporary practice.
-
•
Results may support LAAO as a first-line alternative to long-term NOACs.
ABSTRACT
Introduction
Both percutaneous left atrial appendage occlusion (LAAO) and nonvitamin K antagonist oral anticoagulants (NOACs) are noninferior to warfarin for stroke prevention in high-risk patients with atrial fibrillation (AF). However, there is limited data comparing LAAO with NOACs. The CATALYST trial compares a dual-seal LAAO device (Amplatzer™ Amulet™) to NOACs in AF patients indicated for thromboprophylaxis.
Method
CATALYST is a prospective, multicenter, randomized controlled, open-label trial with an adaptive statistical design. Up to 2,650 AF patients with CHA 2 DS 2 -VASc score ≥2 (men) or ≥3 (women) will be randomly assigned to LAAO or NOAC at 123 global sites. Patients randomized to NOACs take the appropriate labeled dose with compliance monitored at each visit, while LAAO patients receive dual antiplatelet therapy followed by aspirin monotherapy for ≥12 months postimplant. Patients are followed through 5 years, with postimplant cardiac imaging at 3- and 12-months. There are three co-primary endpoints: (1) ischemic stroke, systemic embolism, or cardiovascular death through 2 years, tested for noninferiority; (2) major or clinically relevant nonmajor bleeding through 2 years, tested for superiority; and (3) ischemic stroke or systemic embolism through 3 years, tested for noninferiority. The following secondary endpoints will be tested if the primary endpoints are met: (1) all-bleeding, tested for noninferiority; (2) followed by testing for superiority; (3) disabling or fatal strokes, tested for superiority; all through 2 years.
Conclusions
CATALYST is evaluating the safety and effectiveness of a dual seal LAAO device compared to NOACs in patients with AF at increased risk of stroke.
Clinical Trial Registration
URL https://clinicaltrials.gov ; Unique Identifier NCT04226547.
Background
Atrial fibrillation (AF) is the most common cardiac arrhythmia, and its prevalence in the United States (US) is projected to increase to 12.1 million people by 2030. AF is associated with heart failure, cognitive decline, and vascular dementia, resulting in hospitalization rates of 10% to 40% annually. AF is associated with a 5-fold increased risk of ischemic stroke, and AF-related ischemic strokes are nearly twice as often fatal compared to ischemic strokes of other etiologies. ,
Dose-adjusted vitamin K antagonists such as warfarin reduce the risk of ischemic stroke, but treatment challenges exist. These include a narrow therapeutic index mandating anticoagulation monitoring, delayed onset and offset of effect, an increased risk of bleeding—particularly intracerebral hemorrhage (ICH), and interactions with food and other drugs. , On the other hand, nonvitamin K antagonist oral anticoagulants (NOACs) offer a more rapid onset and offset of effect, fixed-dosing without the need for monitoring of anticoagulation intensity, more predictable pharmacokinetics, fewer contraindications, and a lower risk of ICH. Accordingly, in society guidelines, NOAC has a Class I recommendation, level of evidence (LoE) A and B-R, as preference over vitamin K antagonists for the prevention of ischemic stroke and thromboembolism in patients with AF at elevated risk for stroke. , NOACs reduce mortality and ICH compared with warfarin. But as a class, NOACs increase the risk of gastrointestinal bleeding, are associated with suboptimal long-term adherence, and are prohibitively costly in some geographies. Therefore, there still exists a need for an alternative approach to stroke prophylaxis. ,,
In AF patients, the LAA has been implicated as the most frequent source of cardioembolism—early data suggested it to be the source in >90%, but more recent data implicates the LAA in 99%. , Mechanical occlusion of the LAA by percutaneously implanting an LAAO device offers an alternative means to reduce the risk of ischemic stroke and thromboembolic events without the requirement for long-term OAC. The Amplatzer™ Amulet™ Left Atrial Appendage Occlusion (LAAO) device (Abbott, Plymouth, MN, USA) received CE Mark in 2013 and U.S. FDA approval in 2021 for the prevention of thromboembolism in patients with AF at increased risk for stroke, who have an appropriate rationale to seek a nonpharmacologic alternative to OAC. Society guidelines for the management of patients with AF recommend LAAO with a focus on patients with contraindications to long-term OAC (Class IIa, LoE B-NR or Class IIb, LoE B-R depending on patient characteristics in the U.S. and Class IIb, LoE C in Europe). ,
The investigator-initiated PRAGUE-17 trial randomized 402 AF patients to LAAO (61.3% with the Amulet device, remainder Watchman) or NOACs (mainly apixaban), finding LAAO to be noninferior to NOACs for the primary composite endpoint of stroke, transient ischemic attack, systemic embolism, cardiovascular death, clinically-relevant bleeding, or procedure-/device-related complications. However, nearly half the enrolled patients had prior bleeding, and the primary endpoint in PRAGUE-17 was a composite of both thromboembolic and bleeding events, thereby limiting utility of the results to inform clinical practice for NOAC-eligible patients. Thus, additional well-designed randomized trial data is required to establish the role of percutaneous LAAO as an alternative, front-line therapy for stroke prevention in AF patients.
The Clinical Trial of Atrial Fibrillation Patients Comparing Left Atrial Appendage Occlusion Therapy to Nonvitamin K Antagonist Oral Anticoagulants ( CATALYST ) Trial (NCT04226547) will compare the safety and effectiveness of the Amulet LAAO device (device group) to NOAC therapy (control group) in patients with AF at elevated risk of ischemic stroke without contraindications to anticoagulation. The results of the CATALYST Trial will inform whether LAAO is a reasonable first-line alternative to long-term NOAC therapy for patients with AF at increased risk of stroke.
Methods
Trial Population
The trial will randomize up to 2,650 patients at 123 sites in the U.S., Europe, Japan, Hong Kong, Australia, and Canada. Physicians without prior LAAO experience with the Amulet device may attempt implant in up to 3 patients during a roll-in phase to gain familiarity with the device and delivery sheath. The inclusion and exclusion criteria and follow-up protocol are identical for roll-in and randomized patients.
Patients aged 18 years or older or at the age of local legal consent with documented AF and CHA 2 DS 2 -VASc score ≥3 (women) or ≥2 (men) are recruited (earlier versions of the trial protocol recruited all patients with CHA 2 DS 2 -VASc score of ≥3). Inclusion and exclusion criteria are listed in Table 1 . Eligible patients must be able to comply with the medication regimens in the two treatment groups. The main exclusion criteria include a need for long-term OAC for conditions other than AF, prior atrial septal defect (ASD)/ patent foramen ovale (PFO) occlusion or repair, stroke or transient ischemic attack (TIA) within 90 days of randomization, myocardial infarction within 90 days of randomization, or a left ventricular ejection fraction ≤30%. History of bleeding was not strictly exclusionary, but patients needed to be able to comply with long-term NOAC therapy in the Investigator’s opinion.
Table 1
Inclusion and key exclusion criteria.
| Inclusion criteria | |
|---|---|
| 1 | Documented paroxysmal, persistent, or permanent nonvalvular AF (documentation must include an electrocardiogram, Holter, or event recorder). |
| 2 | At high risk of stroke or systemic embolism, defined as a CHA 2 DS 2 -VASc score of ≥ 3 for women and ≥ 2 for men. |
| 3 | Able to comply with the required NOAC medication regimen if randomized to the Control Group. |
| 4 | Able to comply with the required medication regimen postdevice implant if patient is randomized to the Device Group or patient is a Roll-in. |
| 5 | Able to understand, and is willing to provide, written informed consent to participate in the trial, prior to any clinical investigation related procedure or assessment. |
| 6 | 18 years of age or older, or the age of legal consent. |
| 7 | Able and willing to return for required follow-up visits and assessments. |
| Key exclusion criteria | |
| 1 | Requires long-term OAC therapy for a condition other than AF. |
| 2 | Known contraindication to, or allergic to, aspirin, clopidogrel, or OAC medication use. |
| 3 | Indicated for P2Y12 platelet inhibitor for > 1 year postrandomization. |
| 4 | Has undergone atrial septal defect (ASD) repair or has an ASD closure device. |
| 5 | Has undergone patent foramen ovale (PFO) repair or has a PFO closure device. |
| 6 | Is implanted with a mechanical valve prosthesis. |
| 7 | Is implanted with an inferior vena cava filter. |
| 8 | History of rheumatic or congenital mitral valve heart disease. |
| 9 | Customary contraindications for TEE/TOE (e.g., presence of esophageal varices, esophageal stricture, or history of esophageal cancer). |
| 10 | Experienced stroke or transient ischemic attack (TIA) within 90 days prior to randomization or implant procedure (as applicable). |
| 11 | Underwent catheter ablation for AF or atrial flutter within 60 days prior to randomization. |
| 12 | Experienced myocardial infarction within 90 days prior to randomization. |
| 13 | New York Heart Association class IV congestive heart failure. |
| 14 | Left ventricular ejection fraction ≤ 30% (per most recent assessment). |
| 15 | Symptomatic carotid disease (defined as >50% lumen diameter narrowing on CTA, MRA, or TCD with symptoms of ipsilateral transient or visual TIA evidenced by amaurosis fugax, ipsilateral hemispheric TIAs or ipsilateral stroke); if patient has a history of carotid stent or endarterectomy the patient is eligible if there is <50% lumen diameter narrowing. |
| 16 | Has known intracranial atherosclerosis and/or intracranial small vessel disease (defined as 6 points on the Fazekas Scale). |
| 17 | Reversible cause of AF (i.e., secondary to thyroid disorders, acute alcohol intoxication, trauma, recent major surgical procedures). |
| 18 | History of idiopathic or recurrent venous thromboembolism. |
| 19 | LAA is obliterated or surgically ligated. |
| 20 | Thrombocytopenia (defined as <50,000 platelets per microliter (<50 × 109/L) or anemia (defined as hemoglobin <10 g/dL) requiring transfusions. |
| 21 | Hypersensitivity to any portion of the device material or individual components of the Amulet LAA occluder device (e.g., nickel allergy). |
| 22 | Actively enrolled in, or plans to enroll in, a concurrent clinical study in which the active treatment arm may confound the results of this trial. |
| 23 | Active endocarditis or other infection producing bacteremia. |
| 24 | Severe renal failure (estimated glomerular filtration rate <30 ml/min/1.73m2), but not on dialysis. |
| 25 | Life expectancy is less than 2 years in the opinion of the Investigator. |
| 26 | Presence of other anatomic or comorbid conditions, or other medical, social, or psychological conditions that, in the Investigator’s opinion, could limit the patient’s ability to participate in the clinical investigation or to comply with follow up requirements, or impact the scientific soundness of the clinical investigation results. |
Randomization and Treatments
After the roll-in phase, patients are randomized 1:1 to either attempt LAAO or receive a NOAC medication ( Figure 1 ). All patients are required to initiate treatment (i.e. , undergo an LAAO attempt or receive a first dose of NOAC) within 14 days of randomization. Only patients undergoing an LAAO attempt (randomized to the device group or designated as a roll-in patient) require prescreening cardiac imaging by either transesophageal echocardiography (TEE) or contrast-enhanced computerized tomography (CT) for assessments listed in Table 2 , prior to introduction of the delivery sheath to the vasculature. If an intracardiac thrombus is observed, patients should be anticoagulated to resolve the thrombus and undergo repeated imaging after 30 to 45 days. If thrombus persists, or if any of the other cardiac imaging exclusion criteria are met, patients are deemed incompatible with LAAO and will not be treated percutaneously; however, these patients should remain in the trial and receive OAC at the discretion of the treating physician. Similarly, device group patients declining an implant attempt (e.g., procedural risk concerns) or having a failed attempt should remain in the trial and receive OAC. No cardiac imaging is required for patients randomized to the NOAC group.
Patient Flow Chart . Up to 3 patients may have undergone LAAO as part of a roll-in phase for physicians without prior experience with the Amulet device, with roll-in status designated prior to the procedure. Beyond the roll-in phase, patients were randomized 1:1 to either LAAO (device group) or NOAC medication (control group) with cardiac imaging only required in the device group. Both groups were to be treated within 14 days of randomization and then followed through a 5-year visit. Device group patients not undergoing an implant attempt or having a failed implant attempt were to be treated with OAC, preferably a NOAC.
Table 2
Cardiac imaging assessment.
| 1 | Intracardiac thrombus. |
| 2 | Existing circumferential pericardial effusion >2 mm. |
| 3 | Mitral valve stenosis, defined as a mitral valve area of <1.5 cm 2. |
| 4 | Inter-atrial communication (PFO/ASD with right to left shunt, if TEE/TOE performed) with an atrial septal aneurysm (excursion >15 mm or length ≥15 mm; excursion defined as maximal protrusion of the atrial septal aneurysm beyond the plane of the atrial septum) or large shunt (i.e., during Valsalva maneuver ≥20 bubbles of right-to-left shunt within 3 cardiac cycles after contrast media enters the right atrium, if TEE/TOE performed). |
| 5 | Complex atheroma of the descending aorta and/or aortic arch, defined as plaque >4 mm in thickness, plaque ulceration, and/or mobile lesions. |
| 6 | Cardiac tumor. |
| 7 | LAA anatomy that cannot accommodate the Amulet LAA occluder, as per manufacturer’s IFU. |
| 8 | Placement of the device would damage or interfere with the function of any intracardiac or intravascular structure. |
Physicians are instructed to perform LAAO per the Amulet device Instructions for Use, guided by fluoroscopy and TEE; intracardiac echocardiography (ICE) guidance (instead of TEE) is reserved for physicians experienced with implanting the device with ICE. The degree of sealing is assessed by echocardiography color Doppler at the end of the procedure. An implant attempt is deemed unsuccessful when the patient leaves the procedure room without the device implanted within the LAA. TTE is performed prior to hospital discharge to assess for pericardial effusion or device embolization.
Patients successfully implanted with the LAAO device are discharged on dual antiplatelet therapy (DAPT), consisting of aspirin (75–100 mg daily) and clopidogrel (75 mg daily; Figure 2 ). Clopidogrel is stopped 6 weeks after implantation (earlier versions of the protocol required DAPT for 3 months as initial data was gained on the shortened 6-week DAPT regimen) and remain on aspirin monotherapy through at least 12 months postimplantation, when aspirin may be stopped at physician discretion. If at any time during follow-up, peri-device flow >5 mm or device-related thrombus is identified, patients are to be treated with an anticoagulant until peri-device flow resolves to ≤5 mm or the thrombus resolves, respectively. Closing of peri-device flow (e.g., with vascular plugs or coils) may be performed at physician discretion and is not recommended or prohibited by the trial protocol.
Device Group Antithrombotic Medication Regimen . Patients with successful LAAO were to receive dual antiplatelet therapy with aspirin and clopidogrel for 6 weeks. If peri-device leak was ≤5mm at the end of the procedure, patients were to transition to aspirin alone at 6 weeks through at least 12 months postimplant. OAC was to be prescribed if a device group patient did not receive the device, a peri-device leak >5mm was observed at any time, or a device-related thrombus formed.
Patients randomized to the NOAC arm are prescribed any commercially-available NOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) at a dose according to local country labeling as adjusted for specified patient characteristics (age, body weight, renal function, and/or concomitant therapy).
Follow-up visits are scheduled at 3, 6, 12, 18, 24, and 30 months, and then annually from 3 through 5 years. Patients implanted with the LAAO device require a TEE at the 3-month visit and cardiac imaging with either TEE or cardiac CT at the 12-month visit. The Questionnaire for Verifying Stroke-Free Status (QVSFS) is administered at baseline and at each follow-up visit, and the EQ-5D-5L quality of life questionnaire is administered at baseline, 3-, 6-, and 24-month visits. Any “yes” answer on the QVSFS requires further evaluation by the site Investigator. Neurological assessments using the Modified Rankin Scale (mRS) and National Institutes of Health Stroke Scale (NIHSS) are performed at baseline, when a patient is diagnosed with a stroke or TIA, and at 90-days following stroke (mRS only). An assessment of antithrombotic medications is performed at every visit for all patients. A medication log tracks dose, frequency, start-, and stop-dates for each medication from the date of enrollment until completion of follow-up.
Endpoints
The trial will evaluate three co-primary endpoints: (1) the composite of ischemic stroke, systemic embolism, or cardiovascular mortality through 2 years, tested for noninferiority; (2) the composite of major bleeding or clinically relevant nonmajor bleeding (CRNMB), excluding procedure-related events, through 2 years, tested for superiority; and (3) the composite of ischemic stroke or systemic embolism through 3 years, tested for noninferiority. If these 3 co-primary endpoints are met, then the following secondary endpoints will be evaluated: (1) the composite of major bleeding or CRNMB through 2 years, tested for noninferiority; (2) followed by testing for superiority through 2 years; and (3) disabling or fatal strokes, tested for superiority through 2 years.
Ischemic stroke is defined as the sudden onset of signs or symptoms consistent with a focal or multifocal vascular territory in the brain, spinal cord, or retina that persists ≥24 hrs or until death in the absence of another apparent cause. When symptoms last <24 hrs, pathology or neuroimaging must confirm infarction in the corresponding vascular territory. Systemic embolism is defined as acute vascular insufficiency or arterial occlusion in an extremity or any organ outside the central nervous system associated with clinical, imaging, surgical, or autopsy evidence of arterial embolic occlusion in the absence of another likely mechanism (e.g., trauma, atherosclerosis, or instrumentation). In patients with prior peripheral artery disease, angiographic, surgical, or autopsy evidence of abrupt arterial occlusion is required. Cardiovascular mortality includes death due to a proximate cardiac cause or vascular disease (pulmonary or systemic embolism, aortic aneurysm, stroke), procedure-related death, nontraumatic sudden or unwitnessed death, or death of unknown cause.
Major bleeding follows the International Society of Thrombosis and Haemostasis (ISTH) definition of symptomatic hemorrhagic events involving a critical anatomical site or organ (e.g., intracranial, intraspinal, intraocular, retroperitoneal, pericardial, or intramuscular with compartment syndrome), or causing a fall in hemoglobin of ≥20 g/L or leading to transfusion of ≥2 units of whole blood or erythrocytes, or resulting in fatality. CRNMB is defined as a sign or symptom of hemorrhage not meeting the definition of major bleeding but resulting in hospitalization, medical intervention (inclusive of a healthcare provider instructing a patient to hold antithrombotic medications for ≥5 days or use of a NOAC reversal agent), or surgical intervention.
Disabling stroke is defined as stroke resulting in a 90-day poststroke modified Rankin score (mRS) ≥2, with an increase of ≥1 point compared to the prestroke baseline. Fatal stroke is defined as death attributable to stroke.
Peri-procedural safety descriptive endpoints include: (1) serious device- or procedure-related complications requiring either an invasive surgical or percutaneous intervention through 7 days postprocedure, and (2) serious device- or procedure-related complications through 90 days postprocedure irrespective of the need for invasive surgical or percutaneous intervention. Implant success will be descriptively reported as device implant success (LAAO device deployed and implanted within the LAA), technical success (occlusion of the LAA, no device-related complications through 7 days, and no end-of-procedure peri-device leak >5 mm), and procedural success (technical success with no procedure-related complications through 7 days, except for uncomplicated minor device embolization).
Statistical Methods
The trial is designed to evaluate the effectiveness and safety of LAAO compared to NOAC therapy for the three co-primary endpoints—criteria for which must all be satisfied. The primary analysis will be conducted on an intention-to-treat basis where treatment groups are analyzed as assigned at randomization, regardless of the final actual treatment(s) received (e.g., unattempted or unsuccessful LAAO or OAC use due to leak or thrombus are still analyzed in the device group). The randomization date will be considered Day 0 when referring to a specific day number. All available follow-up data after randomization will be used for the primary analysis. Patients without an observed outcome event who exit the study before completion of planned follow-up will be censored at the time of last information available.
Each co-primary endpoint will be analyzed using a Bayesian piecewise exponential time-to-event model with treatment group-specific hazard rates estimated within each time interval. The primary estimand from the model is the risk difference at 2 years between treatment groups for primary endpoints 1 and 2 and the risk difference at 3 years for primary endpoint 3. The success criteria for each endpoint requires that the 97.5% upper credible interval bound of the risk difference is less than the prespecified endpoint-specific margin (summarized in Table 3 ).
Table 3
Co-primary endpoints and corresponding null and alternative hypotheses.
| Co-Primary Endpoint | Endpoint 1: Ischemic stroke, systemic embolism, or CV mortality through 2 years (noninferiority test) | Endpoint 2: Major bleeding or CRNMB events through 2 years, excluding procedure-related events (superiority test) | Endpoint 3: Ischemic stroke or systemic embolism through 3 years (noninferiority test) |
|---|---|---|---|
| Null hypothesis | p 1 D − p 1 C ≥ 2.7 % | p 2 D − p 2 C ≥ 0 | p 3 D − p 3 C ≥ 2.6 % |
| Alternative hypothesis | p 1 D − p 1 C < 2.7 % | p 2 D − p 2 C < 0 | p 3 D − p 3 C < 2.6 % |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
