TriClip G4 System

P230007 · ABBOTT MEDICAL · NPS · Apr 1, 2024 · Cardiovascular

Device Facts

• Record ID: P230007
• Device Name: TriClip G4 System
• Applicant: ABBOTT MEDICAL
• Product Code: NPS · Cardiovascular
• Decision Date: Apr 1, 2024
• Decision: APPR
• Device Class: Class 3
• Attributes: Therapeutic

Intended Use

The TriClip G4 System is indicated for improving quality of life and functional status in patients with symptomatic severe tricuspid regurgitation despite optimal medical therapy, who are at intermediate or greater risk for surgery and in whom transcatheter edge-to-edge valve repair is clinically appropriate and is expected to reduce tricuspid regurgitation severity to moderate or less, as determined by a multidisciplinary heart team.

Device Story

The TriClip G4 System is a transcatheter edge-to-edge repair device used to treat severe tricuspid regurgitation (TR) without open-heart surgery. The system includes a steerable guide catheter, delivery system, and implantable clips. Under general anesthesia with echocardiographic and fluoroscopic guidance, a physician accesses the right atrium via the femoral vein to position the delivery system. The implant grasps and coapts tricuspid valve leaflets, creating fixed approximation throughout the cardiac cycle. The device is available in four sizes (NT, XT, NTW, XTW) and allows for repositioning by opening, closing, and locking the clip. By reducing TR severity, the device aims to improve patient quality of life and functional status. It is intended for patients at intermediate or greater surgical risk who remain symptomatic despite optimal medical therapy.

Clinical Evidence

Evidence based on the TRILUMINATE Pivotal trial (NCT03904147), a prospective, multicenter, randomized controlled trial (N=350 randomized; N=100 single-arm). Primary endpoint (hierarchical composite of death/surgery, HF hospitalization, KCCQ improvement) met (p=0.0311). Device group showed superior KCCQ improvement (12.34 vs 0.61, p<0.0001) and TR reduction to moderate or less at 30 days (87.0% vs 5.4%, p<0.0001). 6MWD endpoint not met. Safety demonstrated by 98.3% freedom from MAEs at 30 days. Single-arm cohort met primary endpoint (46.2% survival with ≥10-point KCCQ improvement).

Technological Characteristics

System includes TriClip Steerable Guide Catheter, Delivery System, and implantable clips. Materials: nickel-titanium alloy, cobalt-chromium alloy. Implant available in four sizes (NT, XT, NTW, XTW). Principle: transcatheter edge-to-edge leaflet coaptation. Sterilization: ethylene oxide (EtO) for sterile components. MRI conditional at 1.5T and 3.0T. Bench testing per ISO 5910:2018; corrosion resistance per ASTM F2129.

Indications for Use

Indicated for symptomatic severe tricuspid regurgitation patients (age ≥18) despite optimal medical therapy, at intermediate or greater surgical risk, where transcatheter edge-to-edge repair is expected to reduce TR to moderate or less. Contraindicated in patients with procedural anticoagulation intolerance, hypersensitivity to nickel-titanium or cobalt-chromium alloys, or active endocarditis/infection.

Regulatory Classification

Identification

These devices are intended to repair heart valves by methods other than that used by the classified device (annuloplasty rings). They may be placed using non-traditional techniques or other novel methods for decreasing valvular regurgitation.

Reference Devices

• MitraClip G4 (commercially available)
• COAPT trial control group (NCT01626079)

Submission Summary (Full Text)

{0}

SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)

I. GENERAL INFORMATION

Device Generic Name: Tricuspid Valve Repair Device, Percutaneously Delivered

Device Trade Name: TriClip G4 System

Device Procode: NPS

Applicant Name and Address: Abbott Medical
177 County Road B East
St. Paul, MN 55117 USA

Date of Panel Recommendation: February 13, 2024

Premarket Approval Application (PMA) Number: P230007

Date of FDA Notice of Approval: April 1, 2024

Breakthrough Device: Granted breakthrough device status on November 19, 2020, because the device can provide for more effective treatment of an irreversibly debilitating disease; as well as represents a breakthrough technology, offers significant advantages over existing approved or cleared alternatives, and is in the best interest of patients.

II. INDICATIONS FOR USE

The TriClip G4 System is indicated for improving quality of life and functional status in patients with symptomatic severe tricuspid regurgitation despite optimal medical therapy, who are at intermediate or greater risk for surgery and in whom transcatheter edge-to-edge valve repair is clinically appropriate and is expected to reduce tricuspid regurgitation severity to moderate or less, as determined by a multidisciplinary heart team.

III. CONTRAINDICATIONS

The TriClip G4 System is contraindicated in patients with the following conditions:
- Intolerance, including allergy or untreatable hypersensitivity, to procedural anticoagulation.
- Untreatable hypersensitivity to implant components (nickel-titanium alloy, cobalt-chromium alloy).
- Active endocarditis or other active infections of the tricuspid valve.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{1}

PMA P230007: FDA Summary of Safety and Effectiveness Data
2 of 51

# IV. WARNINGS AND PRECAUTIONS

The warnings and precautions can be found in the TriClip G4 System labeling.

# V. DEVICE DESCRIPTION

The TriClip G4 System is designed to repair the native tricuspid valve without open heart surgery by grasping and bringing together (coapting) the tricuspid valve leaflets to reduce tricuspid regurgitation (TR). The device, as shown in Figure 1, is composed of the TriClip Steerable Guide Catheter (SGC), the TriClip G4 Delivery System (TDS), and Accessories.

Figure 1. TriClip G4 System and Stabilizer Accessory.

## TriClip Steerable Guide Catheter

The TriClip Steerable Guide Catheter's primary function is to access the right atrium, maneuver to the target location above the tricuspid valve and position the TriClip G4 Delivery System.

## TriClip G4 Delivery System

The TriClip G4 Delivery System (TDS) is used to deliver, position, and place the implant of the TriClip G4 System on the tricuspid valve leaflets. The TDS is comprised of the Delivery Catheter, the Steerable Sleeve, a handle, and the TriClip G4 Implant. The user interface on the TDS allows for the adjustment of the implant to the desired position for implantation, which are open, closed or inverted.

{2}

The Delivery Catheter controls the actuation and deployment of the TriClip G4 Implant. The Delivery Catheter is controlled using the Arm Positioner, Gripper Levers, Actuator Knob, and Lock Lever.

The Steerable Sleeve facilitates the navigation and positioning of the TriClip G4 Implant in the appropriate location above the tricuspid valve.

The Implant grasps and coapts the tricuspid valve leaflets resulting in fixed approximation of the leaflets throughout the cardiac cycle. It is available in four sizes (NT, XT, NTW, XTW) and can be locked, unlocked, and repeatedly opened and closed to allow for repositioning of the implant to the target location.

## Accessories

The Accessories are intended to support the TriClip G4 System during the procedure. The Accessories consist of the Stabilizer, the Support Plate, and the Lift. The Lift and the Support Plate are used outside of the sterile field to provide a stable platform during the procedure. The Stabilizer is used in the sterile field to support and position the TriClip Steerable Guide Catheter and the TriClip G4 Delivery System during the procedure.

## VI. ALTERNATIVE PRACTICES AND PROCEDURES

There are alternative TR treatments including medical therapy, surgery, and transcatheter tricuspid valve replacement. Each alternative has its own advantages and disadvantages. A patient should discuss these alternatives with his/her physician to select the treatment that best meets expectations and lifestyle.

## VII. MARKETING HISTORY

The TriClip G4 System is commercially available in the European Union, Argentina, Australia, Canada, Colombia, Costa Rica, Ecuador, Indonesia, Israel, Lebanon, Malaysia, Mexico, New Zealand, Russia, Saudi Arabia, Singapore, Taiwan, Thailand, Turkey, and United Kingdom. The device has not been withdrawn from marketing for any reason related to its safety or effectiveness.

## VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH

Below is a list of the potential adverse effects (e.g., complications) associated with the use of the TriClip G4 System.

- Allergic reactions or hypersensitivity to latex, contrast agent, anaesthesia, device materials and drug reactions to anticoagulation, or antiplatelet drugs
- Additional treatment or surgery due to device-related complications
- Bleeding

- Nausea or vomiting
- Pain
- Pericardial effusion
- Stroke/cerebrovascular accident (CVA) and transient ischemic attack (TIA)
- System organ failure:
- Cardio-respiratory arrest

PMA P230007: FDA Summary of Safety and Effectiveness Data
3 of 51

{3}

- Blood disorders (including coagulopathy, hemolysis, and heparin induced thrombocytopenia (HIT))
- Cardiac arrhythmias (including conduction disorders, atrial arrhythmias, ventricular arrhythmias)
- Cardiac ischemic conditions (including myocardial infarction, myocardial ischemia, unstable angina, and stable angina)
- Cardiac perforation
- Cardiac tamponade
- Chest pain
- Death
- Dyspnea
- Edema
- Embolization (device or components of the device)
- Endocarditis
- Fever or hyperthermia
- Fluoroscopy and transesophageal echocardiogram (TEE)-related complications:
- Skin injury or tissue changes due to exposure to ionizing radiation
- Esophageal irritation
- Esophageal perforation
- Gastrointestinal bleeding
- Hypotension or hypertension
- Infection including:
- Septicemia

- Worsening heart failure
- Pulmonary congestion
- Respiratory dysfunction, failure, or atelectasis
- Renal insufficiency or failure
- Cardiogenic or anaphylactic shock

- Thrombosis
- Tricuspid valve complications, which may complicate or prevent later surgical repair, including:
- Chordal entanglement or rupture
- Single leaflet device attachment (SLDA)
- Dislodgement of previously implanted devices
- Tissue damage
- Tricuspid valve stenosis
- Worsening, persistent or residual tricuspid regurgitation

- Vascular access complications which may require additional intervention, including:
- Wound dehiscence,
- Bleeding at the access site
- Arteriovenous fistula, pseudoaneurysm, aneurysm, dissection, perforation or rupture, vascular occlusion
- Embolism (air, thrombus)
- Peripheral nerve injury

- Venous thrombosis (including deep vein thrombosis) and thromboembolism (including pulmonary embolism)

For specific adverse events that occurred in the clinical study, please see Section X below.

IX. SUMMARY OF NONCLINICAL STUDIES

A. Laboratory Studies

Nonclinical laboratory studies on the TriClip G4 System were performed in accordance with, but not limited to, ISO 5910:2018, Cardiovascular implants and extracorporeal systems - Cardiac valve repair devices, along with relevant FDA guidance documents.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{4}

# 1. Biocompatibility

Biocompatibility of the TriClip G4 System was assessed in accordance with ISO 10993-1, Biological Evaluation of Medical Devices - Part 1: Evaluation and testing within a risk management process, and the FDA Guidance for Industry and Food and Drug Administration Staff, Use of International Standard ISO 10993-1, Biological evaluation of medical devices - Part 1: Evaluation and testing within a risk management process. The required testing for each component was determined based on the nature and duration of body contact per ISO 10993-1. The test articles consisted of patient-contacting device components after exposure to all manufacturing processes, including sterilization.

Testing to support the TriClip G4 implant was leveraged from the commercially available MitraClip G4 implant testing. The TriClip G4 implant is identical to the MitraClip G4 implant in formulation, geometry, sterilization, and manufacturing processes. The biocompatibility test results to support the TriClip G4 implant are summarized in Table 1.

Table 1. Summary of TriClip G4 Clip Biocompatibility Assessment

| Biological Effect per ISO10993-1 | Test Method | Results |
| --- | --- | --- |
| Cytotoxicity | Colony assay | Non-cytotoxic |
| Sensitization | Guinea pig maximization | Non-sensitizing |
| Irritation/Intracutaneous Reactivity | Intracutaneous reactivity | Non-irritating |
| Pyrogenicity | Material-mediated rabbit pyrogen | Non-pyrogenic |
| Systemic Toxicity | Acute Systemic Toxicity | No biologically significant signs of systemic toxicity |
| | Subchronic Systemic Toxicity | No biologically significant signs of systemic toxicity |
| Implantation | Implant 13 Weeks - IM | Non-irritant |
| | Implant 4 Weeks - IM | Slight local irritant |
| | Implant 13 Weeks - IM | Local tissue inflammatory or tissue toxicity response (e.g., degenerative or necrotic changes) to the test article in all animals was significantly stronger with test articles than negative control |
| Implant Toxicity | Implant Toxicity 4 Weeks - IM | Non-irritant, and no patterns of systemic toxicity |
| | Implant Toxicity 13 Weeks - IM | Non-irritant, and no patterns of systemic toxicity |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{5}

PMA P230007: FDA Summary of Safety and Effectiveness Data
6 of 51

| Biological Effect per ISO10993-1 | Test Method | Results |
| --- | --- | --- |
| Hemocompatibility | Hemolysis – Direct Contact and Extract | Non-Hemolytic |
| | Complement Activation Assay – SC5b-9 | Equivalent to comparison and/or reference material |
| | Prothrombin Time | Clotting time greater than or not statistically significantly lower than the clotting time of the negative control and negative reference control |
| | Partial Thromboplastin Time | Clotting time greater than or not statistically significantly lower than the clotting time of the negative control and negative reference control |
| | Platelet & Leukocyte | The mean percentage value of the platelet cell counts was within 80 to 120% of the negative control |
| Genotoxicity | Ames – Bacterial Reverse Mutation | No biologically significant increases in reversion rates |
| | In vitro Mouse Lymphoma | Non-genotoxic, non-clastogenic |
| | Chromosomal Aberration | No biologically significant increases in aberrations rates |
| | In vivo Mouse micronucleus | Non-mutagenic |

Testing to support the TriClip G4 delivery system was leveraged from the commercially available MitraClip G4 delivery system testing. The TriClip G4 delivery system is similar to the MitraClip G4 delivery system in terms of formulation, geometry, sterilization, and manufacturing processes, and minor differences are not expected to impact the biocompatibility device profile. The biocompatibility test results to support the TriClip G4 delivery system are summarized in Table 2.

Table 2. Summary of TriClip G4 TDS (excluding Clip), TriClip G4 Delivery Catheter, and Steerable Guide Catheter Biocompatibility Assessment

| Biological Effect per ISO10993-1 | Test Method | Results |
| --- | --- | --- |
| Cytotoxicity | Cytotoxicity – MEM Elution Assay | Non-cytotoxic |
| | Cytotoxicity – Colony | Non-cytotoxic |

{6}

| Biological Effect per ISO10993-1 | Test Method | Results |
| --- | --- | --- |
| | Assay | |
| Sensitization | Sensitization – Guinea Pig Maximization | Non-sensitizing |
| Irritation | Irritation –Intracutaneous Reactivity | Non-irritant |
| Pyrogenicity | Material-Mediated Pyrogen | Non-pyrogenic |
| Systemic Toxicity | Acute Systemic Toxicity | No biologically significant signs of systemic toxicity |
| Hemocompatibility | Hemolysis – Direct Contact and Extract | Non-Hemolytic |
| | Complement Activation Assay – SC5b-9 | Equivalent to comparison and/or reference material |
| | Partial Thromboplastin Time | Clotting time greater than or not statistically significantly lower than the clotting time of the negative control and negative reference control |
| | Platelet & Leukocyte | The mean percentage value of the platelet cell counts was within 80 to 120% of the negative control |

## 2. Bench Testing

A summary of the bench testing results is provided in Table 3.

Table 3. Summary of TriClip G4 System Bench Testing

| Test | Purpose | Results |
| --- | --- | --- |
| Catheter dimensions | Quantitatively assess the dimensions (lengths, outer diameters, inner diameters) of the Steerable Guide Catheter (SGC), Dilator, Steerable Sleeve, Delivery Catheter and Clip | Steerable Guide Catheter, Dilator, Steerable Sleeve, Delivery catheter and Clip dimensions met all design requirements and acceptance criteria. |
| Curves and Steering Performance | Quantitatively assess curve positioning and steerability of the SGC and Sleeve curves | All curves met design requirements and acceptance criteria. |
| Clip Reliability | Confirm the Clip is able to open, lock and unlock, maintain the locked position, and raise and lower Grippers the required number of times under | Tested Clip Delivery Systems met all reliability requirements and acceptance criteria. |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{7}

| Test | Purpose | Results |
| --- | --- | --- |
| | simulated use conditions | |
| Delivery Catheter Stability | Quantitatively assess the stability of the Delivery Catheter during Clip positioning | The Delivery Catheter met all design requirements and acceptance criteria. |
| Clip Deployment | Quantitatively assess the forces on the system when the Clip is deployed from the catheter | All deployment forces met all design requirements and acceptance criteria. |
| Tensile Strengths | Quantitatively assess the tensile strength of the SGC, Dilator, Steerable Sleeve and Delivery Catheter bonds | All bonds met all tensile strength design requirements and acceptance criteria. |
| Torsional Strengths | Quantitatively assess the torsional strength of the SGC, Dilator, Steerable Sleeve and Delivery Catheter bonds | All bonds met all torsional strength design requirements and acceptance criteria. |
| Delivery Catheter Functionality | Quantitatively assess the forces delivered to the by the catheter during actuation of the Clip and deployment | The Delivery Catheter met all design requirements and acceptance criteria. |
| Hemostasis | Confirm the catheters do not leak | The catheters met all design requirements and acceptance criteria. |
| Clip Functionality | Quantitatively assess the forces required to operate the Clip | The Clip met all design requirements and acceptance criteria. |
| Particulate | Evaluate the size and count of particulate generated by the TriClip System | Size and count of particulate generated were within acceptable limits. |
| Clip Finite Element Analysis | Determine mechanical stress/strain during worst-case in vivo loading conditions of the tricuspid valve. Results are used to assess the fatigue life of the device | Worst-case tricuspid valve loading conditions for TriClip G4 were derived, and no implant structural component fractures were predicted at 600 million cycles of life under worst-case valvular loading conditions. |
| Clip Fatigue Resistance | Assess durability of the Clip with in vitro long-term benchtop testing | Worst-case fatigue loading conditions were conservatively applied during accelerated Clip durability testing. All tested Clips remained locked, resisted opening, and were free of any fractures throughout 15 years (600 million cycles) of applied worst-case cardiac loading cycles. |
| Clip Corrosion | Evaluate Clip corrosion resistance | Acceptable corrosion resistance was |

PMA P230007: FDA Summary of Safety and Effectiveness Data
8 of 51

{8}

| Test | Purpose | Results |
| --- | --- | --- |
| Resistance | | demonstrated per ASTM F2129 with breakdown potential (Eb) and breakdown gap (Eb-Er) test results that exceeded recommended acceptability thresholds from the literature. |
| MRI Compatibility Testing | Evaluate magnetic resonance imaging (MRI) safety and compatibility of the implant and ensure that the implant can be safely scanned at 1.5 Tesla and 3.0 Tesla field strengths | The TriClip G4 implants were determined to be Magnetic Resonance Conditional under the conditions listed in the device labeling. |

## 3. Sterilization

The TriClip G4 System (TriClip G4 Delivery System and TriClip Steerable Guide Catheter) and the Silicone Pad and Fasteners are sterilized via ethylene oxide (EtO) in accordance with EN ISO 11135-1:2014, Sterilization of health care products – Ethylene oxide – Requirements for development, validation and routine control of a sterilization process for medical devices. The validated EtO sterilization process demonstrated a minimum Sterility Assurance Level (SAL) of 10⁻⁶.

The TriClip G4 Stabilizer, Lift, and Support Plate are provided separately as non-sterile and must be cleaned, disinfected, and/or sterilized prior to each use.

## 4. Packaging and Shelf-Life

The TriClip G4 System is provided to the end user in two packages, one for the TriClip G4 Delivery System and one for the TriClip Steerable Guide Catheter. The TriClip G4 Delivery System and TriClip Steerable Guide Catheter are individually packaged. The Silicone Pad and Fasteners are single use components and are included in the TriClip SGC packaging.

The TriClip G4 Delivery System and TriClip Steerable Guide Catheter packaging systems both include a thermoformed internal tray and lid. The device is placed in the tray, and the tray/lid is individually pouched in a Tyvek/Nylon pouch and heat sealed. The sealed pouch is placed in a shelf carton. The Stabilizer, Lift, and Support Plate are provided separately.

The packaging validation for the sterile components of the TriClip G4 system was conducted per EN ISO 11607-1:2020, Packaging for terminally sterilized medical devices – Part 1: Requirements for materials, sterile barrier systems and packaging systems and EN ISO 11607-2:2020, Packaging for terminally sterilized medical devices – Part 2: Validation requirements for forming, sealing and assembly processes. The packaging validation demonstrated that the packaging system was able to maintain a sterile barrier after exposure to temperature, distribution conditioning, and accelerated aging.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{9}

The shelf life is 12 months for the TriClip G4 Delivery System and 18 months for the TriClip SGC as demonstrated by packaging integrity and product functional testing on aged samples.

# B. Animal Studies

The TriClip G4 System underwent Good Laboratory Practice-compliant preclinical in vivo evaluations in a porcine model as summarized in Table 4. Preclinical testing of the TriClip G4 System was conducted through 90 days. Evaluation through 20 weeks was leveraged from previous studies of the MitraClip implant through at least 20 weeks, which demonstrated that the device is well encapsulated and endothelialized by 90 days. The TriClip implant is identical to the MitraClip implant in formulation, geometry, sterilization, and manufacturing processes.

Table 4: Summary of TriClip G4 Animal Studies

| Chronic 30-Day and 90-Day Study | |
| --- | --- |
| Sample size / animal model | 15 adult pigs (to achieve a sample size of at least N=6 animals at 30 days and N=6 at 90 days that reach the terminal time point) |
| Test articles | 15 MitraClip NT clips (delivered on the tricuspid valve repair system) |
| Technique | The animals in the 90-day group had a sedated follow up at the 30-day time point. At both the follow up and termination procedures echocardiography and fluoroscopy was performed. At the designated time point (either 30 or 90 days), the animals were euthanized and sent to necropsy for gross evaluation and tissue harvest. |
| Objective | The purpose of this study was to demonstrate the safety of the TriClip implant at 30 Day (±2 days) and 90 Day (±2 days) post implant. The objective was to demonstrate safety through systemic and histological evaluation. Endpoints included: - Overall animal health (moribundity) - Device deployment and hemodynamics - Tissue response to the device |
| Results | There was one procedural death, one early termination, and one early death. The pre-determined safety endpoints acceptance criteria were met for all 12 animals (both the 30-day and the 90-day test groups). |
| Conclusion | There were no observed clinically relevant adverse gross or histological changes in the myocardium or tricuspid valves attributed to the test article. Although there were two early deaths in the study (one death prior to the early termination and one early termination), no definite cause of death was determined and, in both cases, tissue responses to the clips were typical for the time points and appeared unrelated to the cause of death. |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{10}

PMA P230007: FDA Summary of Safety and Effectiveness Data
11 of 51

# X. SUMMARY OF PRIMARY CLINICAL STUDY

The applicant performed a clinical study, the TRILUMINATE Pivotal trial (NCT03904147), to establish a reasonable assurance of safety and effectiveness of the TriClip G4 System for patients with symptomatic severe TR despite optimal medical therapy (OMT). The study was conducted under IDE# G170118. Data from this clinical study were the basis for the PMA approval decision. A summary of the clinical study is presented below.

The TriClip System and TriClip G4 System (a next-generation system) were used in the TRILUMINATE Pivotal trial. Minor design changes were made to the TriClip G4 Delivery System compared to the TriClip Delivery System, but the same TriClip Steerable Guide Catheter was used with both generations. The TriClip G4 System added two additional clip sizes compared to the TriClip system, resulting in a total of four clip length and width options with similar designs and no difference in materials or principle of operation. The minor changes and additional implant sizes were not anticipated to impact TRILUMINATE Pivotal trial outcomes.

## A. Study Design

The TRILUMINATE Pivotal trial was a prospective, multicenter, randomized (1:1), controlled clinical trial designed to test the superiority of transcatheter tricuspid repair using the TriClip device plus OMT (device group) vs. OMT alone (control group) in subjects with severe symptomatic TR who were determined by the site's local heart team to be at intermediate or greater risk for mortality or morbidity with open heart surgery. In addition to the Randomized Cohort, the trial also included a Single-Arm Cohort. After being enrolled into the trial, subjects were assigned to a cohort based on the following criteria:

- Randomized Cohort: High likelihood that the TriClip could reduce TR to moderate or less (i.e., less than or equal to grade 2).
- Single-Arm Cohort: High likelihood that the TriClip could reduce TR by at least 1 grade but a low likelihood that TR will be reduced to moderate or less.

This determination was based on multiple considerations, including but not limited to:

- Baseline TR severity
- The presence of cardiovascular implantable electronic device (CIED) leads across the tricuspid valve
- The coaptation gap width

A Cardiac Computed Tomography/Magnetic Resonance Imaging (CT/MRI) imaging sub-study (referred to as imaging sub-study) was conducted for a maximum of 100 subjects to provide insights into cardiac reverse remodeling and quantitative measurements to assess TR severity and the effect of TR changes on clinical endpoints.

The TRILUMINATE Pivotal trial utilized: an independent Eligibility Committee (EC), which confirmed that the subject met enrollment criteria, assessed anatomic suitability for the

{11}

TriClip device, and assigned eligible patients to the Randomized or Single-Arm Cohort; an Echocardiography Core Laboratory (ECL), which reviewed screening echocardiography images to confirm patient eligibility and assessed TR severity, right ventricular measurements, and other measures at baseline and follow-up; a Clinical Events Committee (CEC), which adjudicated all adverse events per pre-established definitions; and a Data Monitoring Committee (DMC), which monitored the safety of subjects throughout trial. The study was unblinded except for the research staff administering Kansas City Cardiomyopathy Questionnaire (KCCQ), 6-minute walk test, SF-36, and New York Heart Association (NYHA) functional classification assessments.

The trial was to enroll up to 550 patients in the Randomized Cohort and up to 200 patients in the Single-Arm Cohort. Up to 3 roll-in patients per implanter could be enrolled at sites with implanters who did not have prior or recent experience using the TriClip device.

## 1. Clinical Inclusion and Exclusion Criteria

Enrollment in the TRILUMINATE Pivotal trial was limited to patients who met the following inclusion criteria:

- In the judgment of the site local heart team, subject has been adequately treated per applicable standards (including medical management) and stable for at least 30 days as follows:
- Optimized medical therapy for TR treatment (e.g., with diuretics)
- Medical and/or device therapy for mitral regurgitation, atrial fibrillation, coronary artery disease and heart failure

The EC confirmed that the subject has been adequately treated medically.

- Subject was symptomatic with severe TR despite optimal medical ± device treatment. TR severity was determined by the assessment of a qualifying transthoracic echocardiogram (TTE) and confirmed by the ECL. The ECL also requested a transoesophageal echocardiogram (TEE) to confirm TR etiology. Note: If any cardiac procedure(s) occurred after eligibility was determined, TR severity was re-assessed 30 days after the cardiac procedure(s).
- The site heart team cardiac surgeon concurred that the patient was at intermediate or greater estimated risk for mortality or morbidity with tricuspid valve surgery.
- NYHA Functional Class II, III or ambulatory class IV
- In the judgment of the TriClip implanting Investigator, femoral vein access was feasible and could accommodate a 25 Fr catheter.
- Age ≥18 years at time of consent.
- Subject provided written informed consent prior to any trial related procedure.

Patients were not permitted to be enrolled in the TRILUMINATE Pivotal trial if they met any of the following exclusion criteria:

- Systolic pulmonary artery pressure (sPAP) &gt;70 mmHg or fixed pre-capillary pulmonary hypertension as assessed by right heart catheterization (RHC).
- Severe uncontrolled hypertension: Systolic blood pressure (SBP) ≥180 mmHg and/or diastolic blood pressure DBP) ≥110 mm Hg.
- Prior tricuspid valve procedure that would interfere with placement of the TriClip device.

PMA P230007: FDA Summary of Safety and Effectiveness Data
12 of 51

{12}

- Indication for left-sided heart intervention (e.g., for severe aortic stenosis, severe mitral regurgitation) or pulmonary valve correction in the prior 60 days. Note: Patients with concomitant mitral and tricuspid valve disease had option of undergoing MR treatment and waiting 60 days prior to being reassessed for the trial.
- Pacemaker or ICD leads that would prevent appropriate placement of the TriClip device.
- Tricuspid valve stenosis, defined as a tricuspid valve orifice of ≤1.0 cm² and/or a mean gradient ≥5 mmHg as measured by the ECL.
- Left ventricular ejection fraction (LVEF) ≤20%
- Tricuspid valve leaflet anatomy which may preclude clip implantation, proper device positioning on the leaflets or sufficient reduction in TR. This may include:
- Evidence of calcification in the grasping area.
- Presence of a severe coaptation defect (&gt;2 cm) of the tricuspid leaflets.
- Severe leaflet defect(s) preventing proper device placement.
- Ebstein Anomaly - Identified by having a normal annulus position while the valve leaflets are attached to the right ventricular walls and interventricular septum.
- Tricuspid valve anatomy not evaluable by TTE and TEE.
- Active endocarditis or active rheumatic heart disease or leaflets degenerated from rheumatic disease (i.e., noncompliant, perforated).
- MI or known unstable angina within prior 30 days.
- Percutaneous coronary intervention within prior 30 days.
- Hemodynamic instability defined as systolic blood pressure (SBP) &lt;90 mmHg with or without afterload reduction, cardiogenic shock or the need for inotropic support or intra-aortic balloon pump or other hemodynamic support device.
- Cerebrovascular accident (CVA) within the prior 90 days.
- Chronic dialysis.
- Bleeding disorders or hypercoagulable state.
- Active peptic ulcer or active gastrointestinal (GI) bleeding.
- Contraindication, allergy, or hypersensitivity to dual antiplatelet and anticoagulant therapy.

Note: Contraindication to either antiplatelet or anticoagulant therapy (individually not both therapies) was not an exclusion criterion.

- Ongoing infection requiring antibiotic therapy (if temporary illness, patients could enroll 30 days after discontinuation of antibiotics with no active infection).
- Known allergy or hypersensitivity to device materials.
- Evidence of intracardiac, inferior vena cava (IVC), or femoral venous mass, thrombus, or vegetation.
- Life expectancy of less than 12 months.
- Subject currently participating in another clinical trial that had not yet reached its primary endpoint.
- Subject currently participating in another clinical investigation for valvular heart disease(s).
- Pregnant or nursing subjects and those who planned pregnancy during the clinical investigation follow-up period. Female subjects of child-bearing potential were required to have a negative pregnancy test done within 7 days of the baseline visit per site standard

PMA P230007: FDA Summary of Safety and Effectiveness Data
13 of 51

{13}

test. Female patients of childbearing potential instructed to use safe contraception (e.g., intrauterine devices, hormonal contraceptives: contraceptive pills, implants, transdermal patches hormonal vaginal devices, injections with prolonged release). It was accepted, in certain cases, to include subjects having a sterilized regular partner or subjects using a double barrier contraceptive method.

- Presence of other anatomic or comorbid conditions, or other medical, social, or psychological conditions that, in the investigator's opinion, could limit the subject'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.

2. Follow-up Schedule

All subjects were required to have a treatment visit within 14 days of randomization (within 14 days of the baseline visit for Single-Arm Cohort subjects). At this visit, device group patients underwent the TriClip procedure, and control group patients were seen by a heart failure specialist and underwent a physical exam, including vital signs, cardiac health status, and evaluation of heart failure medications.

The follow-up time points were at 30 days, 6 months, and 12 months from the date of the treatment visit and will continue annually through 5 years. The device group patients were also assessed at hospital discharge.

Baseline and follow-up visit assessments included physical assessments, medical history laboratory tests, imaging studies, and health status surveys. Adverse events and complications were recorded at all visits.

3. Statistical Analysis Populations

The analysis populations for the TRILUMINATE Pivotal trial are shown in Table 5.

PMA P230007: FDA Summary of Safety and Effectiveness Data
14 of 51

{14}

Table 5: Statistical Analysis Populations

| Analysis Population | Definition |
| --- | --- |
| Randomized Cohort | |
| Intent-to-Treat (ITT) | All patients randomized in the trial. |
| As-Treated (AT) | ITT patients grouped by treatment received.* |
| Per Protocol (PP) | ITT patients who received assigned randomized treatment according to protocol and followed all major study requirements. |
| Attempted Procedure (AP) | Patients randomized to the device group with an attempted TriClip procedure (i.e., femoral vein puncture performed). |
| Single-Arm Cohort | |
| Attempted Procedure (AP) | Patients with an attempted TriClip procedure (i.e., femoral vein puncture performed). |

*Patients randomized to the device group who died or had heart failure hospitalization prior to the TriClip procedure were considered to be in the Control group regardless of randomization. Patients randomized to the device group who died or had heart failure hospitalization after (but not prior to) a TriClip procedure are considered to be in the device group regardless of randomization. Patients who did not experience death or heart failure hospitalization at any time during follow-up were assigned to the group that constituted &gt;50% of their follow-up duration.

## 4. Randomized Cohort Clinical Endpoints

### Primary Endpoint

The primary safety and effectiveness endpoint was a hierarchical composite of the following components at 12 months:

1. Time to all-cause death or tricuspid valve surgery
2. Number of heart failure (HF) hospitalizations
3. An improvement of ≥15 points in KCCQ score from baseline

The hypothesis for the primary endpoint was as follows:

H₀: None of the components are different between the Treatment and Control group
H₁: At least one component is different between the Treatment and Control group

The alternative hypothesis that the device group was superior to the control group in at least one component of the primary endpoint was tested using the Finkelstein-Schoenfeld methodology (Finkelstein et al. 1999)¹ at a two-sided significance level of 5%.

A sample size of 350 randomized patients was simulated to provide approximately 83% power to reject the null hypothesis at a two-sided significance level of 5%.

The 350 randomized ITT patients was defined as the Primary Analysis Population.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{15}

As a supplementary analysis, the win-ratio method (Pocock et al. 2012) $^{2}$ was used to evaluate the treatment effect of the composite endpoint. In this analysis, each pair of patients from the device group and the control group were compared in the order of the hierarchy defined above, and the win ratio was defined as the number of winners divided by the number of losers in the device group.

An adaptive design with sample size re-estimation was planned when the first 150 randomized patients completed the 12-month follow-up visit. At that time, an independent statistician was unblinded to the interim data and calculated the conditional power for the primary endpoint. The interim analysis concluded that the original 350-patient sample size would provide adequate power to assess the primary endpoint.

## Secondary Endpoints

Four powered secondary safety and effectiveness endpoints were assessed hierarchically at 12 months (see Table 6)

Table 6. Ordered List of Secondary Endpoints for Hierarchical Testing (Randomized Cohort)

| Order | Secondary Endpoint | Null and Alternative Hypotheses | Analysis Population | Significance Level |
| --- | --- | --- | --- | --- |
| 1 | Freedom from MAEs at 30 days post-procedure | H0: PD(MAEs) ≤ 90% H1: PD(MAEs) > 90% | AP | 2.5% (one-sided) |
| 2 | Change in KCCQ score at 12 months over baseline | H0: μD(ΔKCCQ) - μC(ΔKCCQ) = 0 H1: μD(ΔKCCQ) - μC(ΔKCCQ) ≠ 0 | ITT | 5% (two-sided) |
| 3 | TR reduction to moderate or less at 30-day visit | H0: PD(TR ≤ 2) - PC(TR ≤ 2) = 0 H1: PD(TR ≤ 2) - PC(TR ≤ 2) ≠ 0 | ITT | 5% (two-sided) |
| 4 | Change in 6MWD at 12 months over baseline | H0: μD(Δ6MWD) - μC(Δ6MWD) = 0 H1: μD(Δ6MWD) - μC(Δ6MWD) ≠ 0 | ITT | 5% (two-sided) |

MAEs: major adverse events, including cardiovascular mortality, new onset renal failure, endocarditis requiring surgery, and non-elective cardiovascular surgery for TriClip device-related adverse events post-index procedure; KCCQ: Kansas City Cardiomyopathy Questionnaire; 6MWD: 6-minute walk distance; AP: attempted procedure; ITT: intent-to-treat; $H_0$ : null hypothesis; $H_1$ : alternative hypothesis; $P_{\mathrm{D}}(\mathrm{MAEs})$ : proportion of TriClip patients free from MAEs; $\mu_{\mathrm{D}}(\Delta \mathrm{KCCQ})$ and $\mu_{\mathrm{C}}(\Delta \mathrm{KCCQ})$ : mean KCCQ score change in TriClip and control patients; $P_{\mathrm{D}}$ TR≤2) and $P_{\mathrm{C}}(\mathrm{TR} \leq 2)$ : proportion of TriClip and control

PMA P230007: FDA Summary of Safety and Effectiveness Data

{16}

patients with
\leq
moderate TR;
\mu_{\mathrm{D}}(\Delta 6\mathrm{MWD})
: mean 6MWD change in TriClip and control patients.

# Additional Outcomes

Additional outcomes assessed for the Randomized Cohort included the following:

- Technical success at exit from procedure room: alive with successful access, delivery and retrieval of the device delivery system, and deployment and correct positioning of a clip, and no need for additional unplanned or emergency surgery or re-intervention related to the device or access procedure
- Device success at 30-days post-procedure: alive with original intended clip(s) in place, and no additional surgical or interventional procedures related to access or device since completion of the original procedure, and intended performance of the clip(s) i.e., $\geq 1$ grade improvement in TR severity, no embolization, single leaflet device attachment, absence of para-device complications)
- Procedural success at 30-days post-procedure: device success, and no device- or procedure-related serious adverse event
- Echocardiographic parameters of tricuspid valve and cardiac function
Clinical and functional parameters

# 5. Single-Arm Cohort Clinical Endpoints

# Primary Endpoint

The primary safety and effectiveness endpoint was survival at 12 months plus a KCCQ score improvement of $\geq 10$ points compared to baseline, tested in the AP population.

The null $(H_0)$ and alternative $(H_1)$ hypotheses for the primary endpoint were as follows:

$H_0: P_D \leq 30\%$

$H_{1}:P_{D} &gt; 30\%$

where $30\%$ was a performance goal based on the expected TriClip patient survival rate and the KCCQ result observed in the COAPT trial control group (NCT01626079; Stone et al. 2018) $^3$ . A sample size of 100 patients was estimated to provide $90\%$ power to reject the null hypothesis at a one-sided significance level of $2.5\%$ .

# Secondary Endpoints

Five powered secondary safety and effectiveness endpoints were assessed hierarchically at 12 months (see Table 7).

PMA P230007: FDA Summary of Safety and Effectiveness Data

{17}

Table 7. Ordered List of Secondary Endpoints for Hierarchical Testing – Single-Arm Cohort.

| Order | Secondary Endpoint | Null and Alternative Hypotheses | Analysis Population | Significance Level |
| --- | --- | --- | --- | --- |
| 1 | TR reduction by at least one grade at 30 days post-procedure | H0: PD(ΔTR ≥ 1) ≤ 50% H1: PD(ΔTR ≥ 1) > 50% | AP | 2.5% (one-sided) |
| 2 | Freedom from MAEs at 30 days post-procedure | H0: PD(MAEs) ≤ 80% H1: PD(MAEs) > 80% | AP | 2.5% (one-sided) |
| 3 | Change in 6MWD at 12 months over baseline | H0: μD(Δ6MWD) ≤ 0 H1: μD(Δ6MWD) > 0 | AP | 2.5% (one-sided) |
| 4 | Freedom from all-cause mortality and tricuspid valve surgery at 12 months | H0: PD(Survival) ≤ 65% H1: PD(Survival) > 65% | AP | 2.5% (one-sided) |
| 5 | Recurrent HF hospitalizations at 12 months | H0: λD(PRE) ≤ λD(POST) H1: λD(PRE) > λD(POST) | AP | 2.5% (one-sided) |

TR: tricuspid regurgitation; MAEs: major adverse events, including cardiovascular mortality, new onset renal failure, endocarditis requiring surgery, and non-elective cardiovascular surgery for TriClip device-related adverse events post-index procedure; 6MWD: 6-minute walk distance; AP: attempted procedure; HF: heart failure; $H_0$ : null hypothesis; $H_1$ : alternative hypothesis; $P_{\mathrm{D}} \Delta \mathrm{TR} \geq 1$ : proportion of TriClip patients with TR reduction by at least 1 grade; $P_{\mathrm{D}}(\mathrm{MAEs})$ : probability of freedom from any MAE; $\mu_{\mathrm{D}} \Delta 6\mathrm{MWD}$ : mean 6MWD change; $\lambda_{\mathrm{D}}(\mathrm{PRE})$ and $\lambda_{\mathrm{D}}(\mathrm{POST})$ : annualized event rates for recurrent HF hospitalizations within 12 months pre- and post-procedure.

# B. Accountability of PMA Cohort

The database for this PMA included data collected through April 24, 2023. A total of 936 eligible patients were enrolled between August 21, 2019 and June 29, 2022 at 68 investigational sites in the US, Canada, and Europe. Of these patients, 901 were approved by the Eligibility Committee and were randomized or had an attempted procedure, including 141 in the Roll-in Cohort, 572 in the Randomized Cohort, and 188 in the Single-Arm Cohort. Patient accountability is shown in Figure 2. As planned, the primary endpoint analysis was performed on the first 350 patients (296 in the US, 38 in Canada, and 16 in Europe) in the Randomized Cohort and the first 100 patients with an attempted procedure in the Single-Arm Cohort.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{18}


Figure 2. Patient Accountability.

At the time of database lock, of the Randomized Cohort patients eligible for the for the 1-year visit, 100% in the device group and 99% in the control group completed the visit (Table 8, Figure 3.) In the single-arm cohort, 96% of eligible patients completed the 1-year visit (Figure 4).

PMA P230007: FDA Summary of Safety and Effectiveness Data
19 of 51

{19}

Table 8. Visit Compliance.

| | Device Group | | | Control Group | | | Single-Arm Cohort | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Visit | Expected Visits | Actual Visits | Compliance1 | Expected Visits | Actual Visits | Compliance1 | Expected Visits | Actual Visits | Compliance1 |
| Baseline | 175 | 175 | N/A | 175 | 175 | N/A | 100 | 100 | N/A |
| Index Procedure or Treatment Visit | 172 | 172 | 100% | 174 | 174 | 100% | 100 | 100 | 100% |
| Discharge (Device group only) | 172 | 172 | 100% | - | - | - | 100 | 100 | 100% |
| 30-Day Visit | 170 | 168 | 99% | 172 | 162 | 94% | 97 | 96 | 99% |
| 6-Month Visit | 157 | 155 | 99% | 158 | 155 | 98% | 90 | 89 | 99% |
| 12-Month Visit | 152 | 152 | 100% | 152 | 150 | 99% | 84 | 81 | 96% |
| Overall Follow-up2 | 823 | 819 | 99% | 656 | 641 | 98% | 471 | 466 | 99% |

1Compliance calculated as Actual/Expected, where Expected excludes subject withdrawal.
2Overall follow-up includes discharge through 12-month visit (excludes baseline visit).

PMA P230007: FDA Summary of Safety and Effectiveness Data

{20}

Figure 3. Disposition of Subjects - Randomized Cohort.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{21}

Figure 4. Disposition of Subjects - Single Arm Cohort.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{22}

# C. Study Population Demographics and Baseline Characteristics

Patient demographics and baseline characteristics for the primary analysis population of the Randomized Cohort and Single-Arm Cohort are shown in Table 9.

In both the randomized and single-arm cohorts, the majority of patients were Caucasian and just over half were female. Over $90\%$ of Randomized Cohort patients had functional TR and atrial fibrillation, and most patients were in NYHA functional class II/III with an average KCCQ score in the mid-50s. Torrential TR was present in approximately half of the patients in both the device and control groups. Medication use at baseline was similar between the two randomized groups. In all, demographics and baseline characteristics were similar between Randomized Cohort Device and Control groups.

Compared to the Randomized Cohort, a higher proportion of Single-Arm Cohort patients had torrential TR (74.0% vs. 50.9%), CIED-related TR (5.1% vs. 0%), a pacemaker or defibrillator (35.0% vs. 16.0%), and larger coaptation gaps (7.4 ± 2.7 vs. 5.5 ± 1.8 mm). Some baseline covariate differences were expected between the Randomized and Single-Arm cohorts as TR severity and complex tricuspid anatomy were considered when assigning patients to each cohort.

Table 9. Demographics and Baseline Characteristics - Primary Analysis Population.

| Demographics and Baseline Characteristics | Summary Statistic* | | |
| --- | --- | --- | --- |
| | Randomized Cohort (N=350) | | Single-Arm Cohort (N=100) |
| | Device (N=175) | Control (N=175) | |
| Demographics | | | |
| Age | 78.0 ± 7.4 (175) | 77.8 ± 7.2 (175) | 80.4 ± 6.2 (100) |
| Sex | | | |
| Male | 44.0% (77/175) | 46.3% (81/175) | 47.0% (47/100) |
| Female | 56.0% (98/175) | 53.7% (94/175) | 53.0% (53/100) |
| Race | | | |
| Caucasian | 85.1% (149/175) | 81.7% (143/175) | 87.0% (87/100) |
| Black/African American | 4.0% (7/175) | 5.7% (10/175) | 7.0% (7/100) |
| Asian | 4.0% (7/175) | 4.0% (7/175) | 3.0% (3/100) |
| American Indian/Alaska Native | 0.6% (1/175) | 0.0% (0/175) | 0.0% (0/100) |
| Native Hawaiian/Pacific Islander | 0.0% (0/175) | 0.0% (0/175) | 0.0% (0/100) |
| Declined or unable to disclose | 6.3% (11/175) | 8.6% (15/175) | 3.0% (3/100) |
| Ethnicity | | | |
| Hispanic or Latino | 2.9% (5/175) | 5.1% (9/175) | 4.0% (4/100) |
| Not Hispanic or Latino | 93.1% (163/175) | 87.4% (153/175) | 94.0% (94/100) |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{23}

PMA P230007: FDA Summary of Safety and Effectiveness Data
24 of 51

| Declined/unknown | 4.0% (7/175) | 7.4% (13/175) | 2.0% (2/100) |
| --- | --- | --- | --- |
| Body mass index (BMI, kg/m²) | 27.0 ± 5.8 (175) | 26.9 ± 5.2 (175) | 26.3 ± 5.3 (100) |
| Medical history | | | |
| Atrial fibrillation | 87.4% (153/175) | 93.1% (163/175) | 93.0% (93/100) |
| Chronic obstructive pulmonary disease | 10.9% (19/175) | 13.7% (24/175) | 22.0% (22/100) |
| CRT/CRT-D/ICD/permanent pacemaker | 16.0% (28/175) | 13.7% (24/175) | 35.0% (35/100) |
| Dyslipidemia | 66.9% (117/175) | 52.6% (92/175) | 64.0% (64/100) |
| Hypertension | 81.1% (142/175) | 80.6% (141/175) | 83.0% (83/100) |
| Liver disease | 6.3% (11/175) | 9.1% (16/175) | 3.0% (3/100) |
| Renal disease | 35.4% (62/175) | 35.4% (62/175) | 36.0% (36/100) |
| Peripheral vascular disease | 9.1% (16/175) | 10.3% (18/175) | 11.0% (11/100) |
| Prior aortic valve intervention | 15.4% (27/175) | 15.4% (27/175) | 11.0% (11/100) |
| Prior mitral valve intervention | 25.7% (45/175) | 24.0% (42/175) | 36.0% (36/100) |
| Echocardiography measurements | | | |
| TR severity | | | |
| Trace | 0.0% (0/173) | 0.0% (0/165) | 0.0% (0/96) |
| Mild | 0.0% (0/173) | 0.0% (0/165) | 0.0% (0/96) |
| Moderate | 2.3% (4/173) | 1.2% (2/165) | 0.0% (0/96) |
| Severe grade 3 (severe) | 25.4% (44/173) | 29.7% (49/165) | 9.4% (9/96) |
| Severe grade 4 (massive) | 21.4% (37/173) | 18.2% (30/165) | 16.7% (16/96) |
| Severe grade 5 (torrential) | 50.9% (88/173) | 50.9% (84/165) | 74.0% (71/96) |
| TR etiology | | | |
| Functional | 94.8% (165/174) | 92.9% (158/170) | 85.9% (85/99) |
| Degenerative | 2.3% (4/174) | 1.2% (2/170) | 5.1% (5/99) |
| Mixed | 2.9% (5/174) | 5.9% (10/170) | 4.0% (4/99) |
| Pacer-related | 0.0% (0/174) | 0.0% (0/170) | 5.1% (5/99) |
| Coaptation gap (mm) | 5.5 ± 1.8 (137) | 5.2 ± 1.7 (142) | 7.4 ± 2.7 (75) |
| Health status | | | |
| KCCQ overall summary score | 56.0 ± 23.4 (175) | 54.1 ± 24.2 (174) | 54.5 ± 22.6 (99) |
| 6MWD (m) | 240.5 ± 117.1 (164) | 253.6 ± 129.1 (169) | 237.7 ± 120.4 (97) |
| NYHA functional class | | | |
| Class I | 0.0% (0/175) | 0.0% (0/175) | 0.0% (0/100) |

{24}

| Class II | 40.6% (71/175) | 44.6% (78/175) | 41.0% (41/100) |
| --- | --- | --- | --- |
| Class III | 57.1% (100/175) | 52.0% (91/175) | 53.0% (53/100) |
| Class IV | 2.3% (4/175) | 3.4% (6/175) | 6.0% (6/100) |
| Medication use | | | |
| Beta-blockers | 72.6% (127/175) | 73.1% (128/175) | 74.0% (74/100) |
| ACE-I or ARBs | 42.3% (74/175) | 45.1% (79/175) | 41.0% (41/100) |
| Vasodilators | 10.9% (19/175) | 12.0% (21/175) | 12.0% (12/100) |
| Diuretics | 97.1% (170/175) | 98.9% (173/175) | 98.0% (98/100) |

CRT: cardiac resynchronization therapy; CRT-D: cardiac resynchronization therapy defibrillator; ICD: implantable cardioverter-defibrillator; KCCQ: Kansas City Cardiomyopathy Questionnaire; 6MWD: 6-minute walk distance; NYHA: New York Heart Association; ACE-I: angiotensin-converting enzyme 1; ARBs: angiotensin receptor blockers.
*Continuous measures – Mean ± standard deviation (total no.); Categorical measures - % (no./total no.)

# D. Safety and Effectiveness Results

The analysis of safety and effectiveness were primarily based on the Randomized Cohort of 350 patients available for the 12-month evaluation. The key safety outcomes for this study included mortality, need for tricuspid valve surgery, and major adverse event rates as assessed by the primary and secondary endpoints. All CEC-adjudicated adverse effects are reported in Table 13 for the Randomized Cohort and Table 23 for the Single-Arm Cohort. The key effectiveness outcomes included number of heart-failure hospitalizations, improvement in KCCQ score, TR reduction and 6-minute walk distance (6MWD) as assessed by the primary and secondary endpoints and summarized below.

# 1. Primary Endpoint - Randomized Cohort

The Randomized Cohort primary safety and effectiveness endpoint analysis results are shown in Table 10. The Finkelstein-Schoenfeld test statistic result was 2.16 with a 2-sided p-value of 0.0311, which is less than the pre-specified two-sided significance level of 0.05. Thus, the primary endpoint was met indicating the device group was superior to the control group.

Table 10. Primary Analysis Result – Randomized Cohort ITT Population.

| Primary Endpoint | Test Statistic | p-Value (2-sided) | Significance Level (2-sided) | Result |
| --- | --- | --- | --- | --- |
| Finkelstein-Schoenfeld analysis | 2.16 | 0.0311 | 0.05 | Superiority endpoint met |

The supplemental win ratio analysis is shown in Figure 5. The win ratio of the device group vs. the control group was 1.44 (95% confidence interval of 1.03 - 2.08). The number of wins in the device group and control group were similar for death or TV surgery, and there were slightly more wins in the control group for heart failure

PMA P230007: FDA Summary of Safety and Effectiveness Data

{25}

hospitalization (6% in the device group vs 10% in the control group). The primary endpoint success was driven by KCCQ score improvement of at least 15 points, which had 21% wins in the device group and 7% wins in the control group.

Figure 5. Win Ratio Analysis of the Randomized Cohort Primary Endpoint - ITT Population. TV: tricuspid valve; HFH: heart failure hospitalization; KCCQ-OS: Kansas City Cardiomyopathy Questionnaire Overall Summary Score.

# 2. Secondary Endpoint - Randomized Cohort

The results of the powered secondary safety and effectiveness endpoints are shown in Table 11. The endpoints of freedom from major adverse events (MAEs) at 30 days post-procedure, change in KCCQ score at 12 months vs. baseline, and TR reduction to moderate or less at 30 days were met. There was a numerically smaller reduction in 6MWD at 12 months in the device group vs. the control group (-8.12 vs. -25.17 meters), but the difference was not statistically significant, and standard deviations were large. Therefore, the 6MWD endpoint was not met.

Table 11. Summary of Powered Secondary Endpoint Results – Randomized Cohort ITT Population (Paired).

| Order | Secondary Endpoint | Summary Statistics | | p-Value | Result |
| --- | --- | --- | --- | --- | --- |
| | | Device Arm | Control Arm | | |
| 1 | Freedom from MAEs at 30 days post-procedure | 98.3% [96.3%, 100%]* | - | < 0.0001 | Endpoint met |
| 2 | Change in KCCQ score at 12 months over baseline | 12.34 (1.75)† | 0.61 (1.75)† | < 0.0001 | Endpoint met |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{26}

PMA P230007: FDA Summary of Safety and Effectiveness Data
27 of 51

| 3 | TR reduction to moderate or less at 30-day visit | 87.0% (141/162)‡ | 5.4% (8/147)‡ | <0.0001 | Endpoint met |
| --- | --- | --- | --- | --- | --- |
| 4 | Change in 6MWD at 12 months over baseline† (meters) | -8.12 (10.50)† | -25.17 (10.31)† | 0.2482 | Endpoint not met |

MAEs: major adverse events, including cardiovascular mortality, new onset renal failure, endocarditis requiring surgery, and non-elective cardiovascular surgery for TriClip device-related adverse events post-index procedure; TR: tricuspid regurgitation; KCCQ: Kansas City Cardiomyopathy Questionnaire; 6MWD: 6-minute walk distance.
*Kaplan-Meier estimate [95% confidence interval]
†Least square means (standard error) from analysis of covariance (ANCOVA) model
‡% (no./total no.)
†A KCCQ overall score of 0 and a 6MWD of 0 meter were imputed for subjects who had a heart failure related cardiovascular death or tricuspid valve surgery prior to 12 months.

The individual MAE component rates at 30 days are shown in Table 12. Of the MAEs, one case of new onset renal failure was adjudicated as procedure-related but not device-related. A second new onset renal failure case and the one cardiovascular mortality were adjudicated as neither procedure nor device related.

Table 12. Results of Individual MAE Components at 30 Days – Randomized Cohort AP Population.

| MAE Component | Event Rate* |
| --- | --- |
| Cardiovascular mortality | 0.6% (1/172) |
| New onset renal failure | 1.2% (2/172) |
| Endocarditis requiring surgery | 0% (0/172) |
| Non-elective cardiovascular surgery for TriClip device-related adverse events post index procedure | 0% (0/172) |

*% (no./total no.)

3. Adverse Events – Randomized Cohort

CEC-adjudicated adverse events through 12 months (unless otherwise noted) are shown in Table 13 for the Randomized Cohort. Rates of HF hospitalizations, cardiovascular mortality, and tricuspid valve reintervention at 12 months as well as major bleeding and new onset renal failure at 30 days were numerically higher in the device group vs. the control group.

{27}

Table 13. CEC-Adjudicated Adverse Events through 12 Months – Randomized Cohort ITT Population.

| Event | Summary Statistics | |
| --- | --- | --- |
| | Device Arm (N=175) | Control Arm (N=175) |
| All-cause mortality | 8.6% (15, 15, 0, 0, 1)* | 7.4% (13, 13, 0)† |
| Cardiovascular (VARC II definition) | 6.3% (11, 11, 0, 0, 0) | 4.6% (8, 8, 0) |
| Heart failure-related | 4.0% (7, 7, 0, 0, 0) | 2.9% (5, 5, 0) |
| Non-heart failure-related | 2.3% (4, 4, 0, 0, 0) | 1.7% (3, 3, 0) |
| Non-cardiovascular (VARC II definition) | 2.3% (4, 4, 0, 0, 1) | 2.9% (5, 5, 0) |
| Hospitalization | 36.0% (111, 63, 2, 7, 2) | 34.3% (100, 60, 0) |
| Heart failure hospitalization | 14.9% (35, 26, 1, 2, 0) | 11.4% (8, 20, 0) |
| Other cardiovascular hospitalization | 9.1% (17, 16, 1, 5, 0) | 9.1% (21, 16, 0) |
| Non-cardiovascular hospitalization | 21.7% (59, 38, 0, 0, 2) | 21.1% (51, 37, 0) |
| Tricuspid valve surgery | 1.7% (3, 3, 2, 2, 0) | 3.4% (6, 6, 0) |
| Tricuspid valve intervention‡ | 2.3% (4, 4, 3, 4, 0) | 1.7% (3, 3, 0) |
| Major bleeding (≥BARC 3a)† | 5.7% (10, 10, 0, 3, 0) | 1.7% (3, 3, 0) |
| New onset renal failure† | 2.3% (4, 4, 0, 1, 0) | 0.6% (1, 1, 0) |
| Transient ischemic attack (TIA) | 0.6% (1, 1, 0, 0, 0) | 0.0% (0, 0, 0) |
| Stroke (VARC II definition) | 1.7% (3, 3, 0, 0, 0) | 1.7% (4, 3, 0) |
| Myocardial infarction (VARC II definition)† | 0.0% (0, 0, 0, 0, 0) | 0.0% (0, 0, 0) |
| Endocarditis requiring surgery† | 0.0% (0, 0, 0, 0, 0) | 0.0% (0, 0, 0) |
| Non-elective cardiovascular surgery for TriClip-related adverse event post index procedure† | 0.0% (0, 0, 0, 0, 0) | 0.0% (0, 0, 0) |
| Cardiogenic shock | 0.0% (0, 0, 0, 0, 0) | 0.6% (1, 1, 0) |

*Event rate (no. of events, no. of subjects, no. of device-related events, number of procedure-related events, number of COVID-19-related events); event rate = no./total no. Number of COVID-19-related events includes related or possibly related events; this excludes events with unknown relatedness.
†Event rate (no. of events, no. of subjects, number of COVID-19-related events).
Tricuspid valve intervention includes reintervention for device group and first intervention for control group.
†Per the study CEC charter, myocardial infarction, bleeding, new onset renal failure, endocarditis requiring surgery, and non-elective cardiovascular surgery for TriClip-related adverse event post index

PMA P230007: FDA Summary of Safety and Effectiveness Data

{28}

procedure were adjudicated up to 30 days post treatment visit for the device and control groups.

VARC: Valve Academic Research Consortium; BARC: Bleeding Academic Research Consortium; TIA: transient ischemic attack.

Table 14 provides site-reported procedure- or device-related serious adverse events from the treatment visit through 1 year. Procedure- or device-related serious adverse events occurred in 3.5% (6/172) of subjects.

Table 14. Listing of Site-reported Procedure/Device Related Serious Adverse Events from treatment visit through 1 Year (Primary Analysis Population) (Attempted Procedure Population, n=172)

| Subject | Description |
| --- | --- |
| 1 | Access site bleeding |
| | Hypotension with tachycardia secondary to acute blood loss |
| 2 | Access site complication |
| 3 | Access site complication – thrombin injection for pseudoaneurysm |
| | Access site complication – surgical repair of pseudoaneurysm |
| 4 | TV surgery due to unsuccessful TriClip procedure |
| 5 | Re-intervention due to SLDA |
| 6 | Heart failure due to volume overload |

## 4. Other Randomized Cohort Observations

**Procedural Endpoints:**

Technical success was achieved in 98.8% of TriClip subjects, device success in 88.9%, and procedural success in 87.0% (see Table 15).

Table 4. Results of Procedural Endpoints – Randomized Cohort AP Population.

| Endpoints | Results |
| --- | --- |
| Technical success (at exit from procedure room) | 98.8% (170/172) |
| Device success (at 30 days post-procedure) | 88.9% (144/162) |
| Procedural success (at 30 days post-procedure) | 87.0% (141/162) |

Technical success was not achieved in 2 subjects due to inability to successfully deploy the TriClip device. Device success could not be evaluated in 10 subjects due to missing TR grade assessment. In addition, device success was not achieved in 18 subjects due to single leaflet device attachment (n=11), no reduction in TR (n=3), surgery/intervention within 30 days post procedure (n=3), and death within 30 days post procedure (n=1). Procedural success was not achieved in the same 18 subjects in whom device success was not achieved and in 3 additional subjects who experienced a

PMA P230007: FDA Summary of Safety and Effectiveness Data
29 of 51

{29}

device- or procedure-related site-reported serious adverse event: single leaflet device attachment (n=1; not confirmed by the ECL), ruptured chordae (n=1), and access site complication (n=1).

## Procedural Data:

The TriClip procedure was performed under general anesthesia with echocardiographic (TEE) and fluoroscopic guidance. Procedural data for the Randomized Cohort AP Population is shown in Table 16. The TriClip was successfully implanted in 170 of the 172 (98.8%) subjects with an attempted procedure in the Randomized Cohort, with approximately 85% of subjects receiving two or three TriClip devices.

Table 16. Procedural Data – AP Population

| Procedural Data | Randomized Cohort (Device Group) (N=172) |
| --- | --- |
| Number of clips implanted | 2.2 ± 0.7 (172) |
| 0 clips | 1.2% (2/172) |
| 1 clip | 10.5% (18/172) |
| 2 clips | 61.0% (105/172) |
| 3 clips | 24.4% (42/172) |
| 4 clips | 2.9% (5/172) |
| Device used | |
| TriClip (first-generation) | 47.1% (81/172) |
| TriClip G4 | 52.9% (91/172) |
| Total procedure time (min) | 151.0 ± 71.7 (171) |
| Device time (min) | 89.7 ± 66.4 (168) |
| Fluoroscopy exposure (min) | 31.9 ± 23.5 (171) |

*Continuous measures – Mean ± standard deviation (total no.); Categorical measures - % (no./total no.)

## TR Severity:

TR severity for the Randomized Cohort ITT Population is shown in Figure 6. In the device group, the proportion of subjects with greater than moderate TR was 97% at baseline, which decreased to 13% at 30 days and 12% at 12 months. In the control group, TR severity was greater than moderate in 99% of subjects at baseline and remained greater than moderate in 95% of subjects at 30 days and 92% at 12 months.

PMA P230007: FDA Summary of Safety and Effectiveness Data
30 of 51

{30}

Note: Values $\leq 1\%$ are not labeled.
Figure 6. TR Severity Visit - Randomized Cohort ITT Population (Unpaired).

# KCCQ Score:

KCCQ scores and score changes through 12 months are shown in Figure 7 for the Randomized Cohort ITT Population. On average, the KCCQ score increased by 15.2 points in the device group vs. 4.8 points in the control group through 12 months.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{31}

A KCCQ Score

B KCCQ Score Change
Figure 7. KCCQ Score by Visit - Randomized Cohort ITT Population (Unpaired). The error bars are standard deviations.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{32}

Association between KCCQ Score and TR:

Post hoc analyses were performed to investigate the associations between KCCQ score changes and TR severity and between KCCQ score changes and TR severity changes at 12 months. These analyses were conducted to provide evidence that the KCCQ score improvement observed in the study was not solely the result of a placebo effect. The associations are shown in Figure 8. Lower TR severity and greater TR severity reductions were generally associated with greater KCCQ score improvements.

Figure 8. Association between KCCQ Score and TR at 12 Months. The error bars are standard deviations.


# SF-36 Score:

SF-36 scores through 12 months are shown in Figure 9 for the Randomized Cohort ITT Population. The mean physical component score increased by about 5 points through 12 months compared to the baseline in the device group, while remaining mostly unchanged from baseline through 12 months in the control group. A similar trend was seen in the mental component score. In some studies, SF-36 score changes similar to the changes observed in the device group have been interpreted as clinically significant.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{33}

A Physical Component

B Mental Component
Figure 9. SF-36 Score by Visit – Randomized Cohort ITT Population (Unpaired). The error bars are standard deviations.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{34}

# NYHA Functional Class:

The results for NYHA classification by visit are shown in Figure 10 for the Randomized Cohort ITT Population. At baseline, $59\%$ of subjects in the device group and $55\%$ in the control group were in NYHA III/IV. At 12 months, fewer device subjects were in NYHA III/IV than the control subjects (16% vs. 40%).

Figure 10. NYHA Functional Class by Visit - Randomized Cohort ITT Population (Unpaired).

# Echocardiographic Parameters:

PISA EROA, PISA regurgitant volume, and vena contracta width all showed substantial decreases from baseline to 12 months in the device group and were minimally changed in the control group (Table 17). There were no notable changes in cardiac size or function in either treatment group at 12 months. Right atrial volume, which would be expected to decrease as a result of reduced TR due to reverse remodeling, showed an unexpected small increase in the device group.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{35}

Table 17. Results of Echocardiographic Endpoints – Randomized Cohort ITT Population (Paired Analysis).

| Echocardiographic Endpoint Change from Baseline to 12 Months | Device Arm (N=175) | Control Arm (N=175) | Difference [95% CI]* |
| --- | --- | --- | --- |
| ΔTricuspid annulus diameter (end-diastole, apical 4Ch, cm) | | | |
| Mean ± SD (n) | -0.09 ± 0.64 (140) | -0.11 ± 0.74 (135) | 0.02 [-0.14, 0.19] |
| Median (Q1, Q3) | -0.10 (-0.50, 0.30) | -0.17 (-0.50, 0.30) | |
| Range (min, max) | (-1.46, 1.39) | (-3.90, 2.02) | |
| ΔPISA EROA (cm²) | | | |
| Mean ± SD (n) | -0.44 ± 0.33 (115) | -0.04 ± 0.31 (127) | -0.40 [-0.48, -0.32] |
| Median (Q1, Q3) | -0.42 (-0.56, -0.26) | 0.00 (-0.16, 0.12) | |
| Range (min, max) | (-2.33, 0.25) | (-1.25, 0.80) | |
| ΔPISA regurgitant volume calculation (mL) | | | |
| Mean ± SD (n) | -33.84 ± 20.48 (115) | -1.99 ± 23.56 (127) | -31.85 [-37.43, -26.28] |
| Median (Q1, Q3) | -33.20 (-44.90, -21.40) | -1.30 (-12.40, 10.21) | |
| Range (min, max) | (-105.20, 12.11) | (-115.90, 67.80) | |
| ΔVena contracta width (SL, 4Ch view, cm) | | | |
| Mean ± SD (n) | -0.52 ± 0.48 (139) | 0.03 ± 0.44 (136) | -0.54 [-0.65, -0.43] |
| Median (Q1, Q3) | -0.48 (-0.77, -0.26) | 0.00 (-0.30, 0.32) | |
| Range (min, max) | (-3.00, 0.97) | (-1.10, 1.40) | |
| ΔRV end diastolic diameter – mid (4Ch, cm) | | | |
| Mean ± SD (n) | -0.18 ± 0.73 (140) | -0.02 ± 0.85 (134) | -0.17 [-0.36, 0.02] |
| Median (Q1, Q3) | -0.20 (-0.60, 0.20) | 0.10 (-0.50, 0.50) | |
| Range (min, max) | (-1.90, 2.80) | (-2.20, 2.90) | |
| ΔRV end diastolic diameter – base (4Ch, cm) | | | |
| Mean ± SD (n) | -0.21 ± 0.71 (142) | -0.12 ± 0.76 (134) | -0.09 [-0.26, 0.08] |
| Median (Q1, Q3) | -0.15 (-0.70, 0.20) | -0.10 (-0.60, 0.40) | |
| Range (min, max) | (-2.40, 2.70) | (-2.00, 1.90) | |
| ΔRight atrial volume (single plane Simpson’s, mL) | | | |
| Mean ± SD (n) | 7.78 ± 55.92 (140) | -2.13 ± 54.14 (136) | 9.91 [-3.13, 22.95] |
| Median (Q1, Q3) | 8.17 (-22.48, 28.25) | -4.35 (-29.90, 21.90) | |
| Range (min, max) | (-122.03, 276.20) | (-154.44, 181.20) | |
| ΔRV fractional area change (%) | | | |
| Mean ± SD (n) | -0.73 ± 8.16 (133) | -0.52 ± 7.38 (125) | -0.21 |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{36}

| Median (Q1, Q3) | -0.50 (-6.40, 3.90) | -1.00 (-5.80, 3.90) | [-2.12, 1.69] |
| --- | --- | --- | --- |
| Range (min, max) | (-27.90, 21.22) | (-18.70, 23.00) | |
| ΔLV end diastolic volume (mL) | | | |
| Mean ± SD (n) | 3.91 ± 25.02 (129) | -4.80 ± 23.49 (114) | 8.70[2.57, 14.84] |
| Median (Q1, Q3) | 3.30 (-12.90, 16.30) | -4.98 (-16.80, 9.70) | |
| Range (min, max) | (-70.30, 94.50) | (-83.20, 52.70) | |
| ΔLV end systolic volume (mL) | | | |
| Mean ± SD (n) | 2.31 ± 15.28 (129) | -2.93 ± 12.52 (114) | 5.24[1.72, 8.75] |
| Median (Q1, Q3) | 0.82 (-4.80, 8.80) | -2.95 (-9.50, 4.20) | |
| Range (min, max) | (-37.00, 85.50) | (-65.34, 23.80) | |
| ΔRV TAPSE (cm) | | | |
| Mean ± SD (n) | -0.13 ± 0.45 (141) | 0.00 ± 0.48 (132) | -0.13[-0.24, -0.02] |
| Median (Q1, Q3) | -0.10 (-0.43, 0.10) | 0.01 (-0.20, 0.30) | |
| Range (min, max) | (-1.40, 1.00) | (-2.27, 1.00) | |
| ΔCardiac output (L/min) | | | |
| Mean ± SD (n) | -0.05 ± 1.89 (136) | 0.03 ± 1.40 (131) | -0.07[-0.47, 0.33] |
| Median (Q1, Q3) | -0.14 (-0.98, 0.63) | -0.04 (-0.88, 0.86) | |
| Range (min, max) | (-4.98, 14.95) | (-3.42, 4.10) | |
| ΔLVOT Doppler stroke volume (mL) | | | |
| Mean ± SD (n) | -1.58 ± 17.62 (138) | -1.93 ± 16.48 (133) | 0.35[-3.73, 4.43] |
| Median (Q1, Q3) | -2.04 (-11.00, 7.80) | -1.50 (-11.73, 4.40) | |
| Range (min, max) | (-49.50, 65.00) | (-40.60, 51.70) | |
| ΔInferior vena cava diameter (cm) | | | |
| Mean ± SD (n) | -0.09 ± 0.56 (135) | -0.01 ± 0.56 (136) | -0.08[-0.21, 0.05] |
| Median (Q1, Q3) | -0.04 (-0.48, 0.34) | 0.00 (-0.34, 0.32) | |
| Range (min, max) | (-1.80, 1.16) | (-1.90, 1.80) | |
| ΔTricuspid valve diastolic mean gradient (CW, mmHg) | | | |
| Mean ± SD (n) | 1.15 ± 1.28 (136) | 0.07 ± 0.58 (126) | 1.08[0.84, 1.32] |
| Median (Q1, Q3) | 0.86 (0.32, 1.89) | 0.02 (-0.31, 0.43) | |
| Range (min, max) | (-2.80, 7.32) | (-1.11, 1.60) | |

PISA: proximal isovelocity surface area (a method for estimating regurgitant volume); EROA: effective regurgitant orifice area; RV: right ventricular; LV: left ventricular; TAPSE: tricuspid annular plane systolic excursion (a measure of the RV apex to-base shortening and RV systolic function); LVOT: left ventricular outflow tract.
SD: standard deviation; CI: confidence interval. The CIs were calculated without multiplicity adjustment. The adjusted CIs could be wider than presented here.
*By normal approximation.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{37}

# 5. 1-Year Outcomes for All Available Subjects in the Randomized Cohort

The trial used an adaptive design with sample size re-estimation for the Randomized Cohort. The pre-specified sample size re-estimation occurred once the first 150 randomized subjects completed 12-month follow-up, while the trial was still enrolling. The trial continued to randomize subjects until the sample size re-estimation analysis was completed and indicated that no further subject enrollment was necessary, by which point 572 subjects were randomized at 68 sites. Fifty-six subjects (29 Device, 26 Control) were pending 12-month follow-up visits at the time of data analysis.

The win ratio analysis result for all available randomized subjects was 1.53 (Figure 11), which is slightly greater than the win-ratio result for the Primary Analysis Population (1.44, Figure 4). The number of device wins and control wins for death or TV surgery continued to be similar between treatment groups for all available randomized subjects. While there were more control group wins for heart failure hospitalizations in the win ratio analysis for the Primary Analysis Population, the number of device group wins and control group wins for heart failure hospitalizations were similar for all available subjects. The win ratio continued to be driven by KCCQ score improvement, and the data for all available subjects support the Primary Analysis Population analysis.

Figure 11. Win Ratio Analysis for All Available Subjects – Randomized Cohort ITT Population. HFH: heart failure hospitalization; KCCQ-OS: Kansas City Cardiomyopathy Questionnaire overall summary score; CI: confidence interval. The CI was calculated without multiplicity adjustment. The adjusted CI could be wider than presented here.

PMA P230007: FDA Summary of Safety and Effectiveness Data

{38}

Components of the primary endpoint for the Primary Analysis Population and Full Randomized Cohort are provided in Table 18.

Table 18. Primary Endpoint Components – Primary Analysis Population and Full Randomized Cohort

| Component | Primary Analysis Population (N=350) | | Full Randomized Cohort (N=572) | |
| --- | --- | --- | --- | --- |
| | Device (N=175) | Control (N=175) | Device (N=285) | Control (N=287) |
| All-cause mortality or TV surgery at 12 months, Kaplan-Meier (%)1 | 9.4% | 10.6% | 9.9% | 9.7% |
| Rate of heart failure hospitalizations, per patient-year2 | 0.22 | 0.17 | 0.17 | 0.19 |
| Proportion with KCCQ-OS improvement ≥15 points at 12 months | 50% | 26% | 50% | 26% |

$^{1}$ Kaplan-Meier estimate with Log-rank test
$^{2}$ Normal approximation for differences in Binomial proportions

Secondary endpoints were consistent in the Primary Analysis Population and Full Randomized cohorts as summarized in Table 19. Device subjects experienced a larger improvement in 6MWD than Control subjects in the Full Randomized Cohort than in the Primary Analysis Population.

Table 19. Secondary Endpoints – Primary Analysis Population and Full Randomized Cohort

| Secondary Endpoints | Primary Analysis Population (N=350) | | Full Randomized Cohort (N=572) | |
| --- | --- | --- | --- | --- |
| | Device (N=175) | Control (N=175) | Device (N=285) | Control (N=287) |
| Freedom from MAE at 30 days | 98.3% | - | 98.9% | - |
| Moderate or less TR at 30 days | 87.0% | 5.4% | 88.9% | 5.3% |
| Change from Baseline to 12 months | | | | |
| KCCQ-OS (imputed\(^a\), ANCOVA), Mean ± SE | 12.3 ± 1.8 | 0.6 ± 1.8 | 11.5 ± 1.6 | -0.5 ± 1.6 |
| Between-group difference, Mean [95% CI]\(^b\) | 11.7 [6.9, 16.6] | | 11.9 [7.4, 16.4] | |
| KCCQ-OS (complete-case paired), Mean ± SD | 15.2 ± 22.3 | 4.8 ± 18.3 | 15.2 ± 22.8 | 4.2 ± 18.9 |
| Between-group difference, Mean [95% CI] | 10.4 [5.7, 15.1] | | 11.0 [6.9, 15.2] | |
| 6MWD (imputed\(^a\), ANCOVA), Mean ± SE | -8.1 ± 10.5 | -25.2 ± 10.3 | -5.0 ± 8.7 | -29.8 ± 8.4 |
| Between-group difference, Mean [95% CI] | 17.1 [-12.0, 46.1] | | 24.8 [1.1, 48.6] | |

PMA P230007: FDA Summary of Safety and Effectiveness Data

{39}

| 6MWD (complete-case paired), Mean ± SD | 11.5 ± 111.4 | -8.7 ± 109.7 | 15.1 ± 103.4 | -12.1 ± 102.0 |
| --- | --- | --- | --- | --- |
| Between-group difference, Mean [95% CI] | 20.3 [-7.2, 47.7] | | 27.2 [5.5, 48.9] | |

a Subjects who experienced HF-related death or had TV surgery prior to 12-month visit were assigned 12-month KCCQ-OS or 6MWD of 0. Subjects who were unable to exercise due to cardiac reasons were also assigned a 6MWD of 0 meters at 12-month follow-up. Subjects who experienced hospitalization related to COVID-19 had their follow-up information following the COVID-19 related hospitalization excluded. bCI: confidence interval. The CI was calculated without multiplicity adjustment. The adjusted CI could be wider than presented here.

CEC-adjudicated adverse event rates through 12 months (unless otherwise noted) were also consistent in the Primary Analysis Population and Full Randomized cohorts. Event rates for the Full Randomized Cohort are summarized in Table 20.

Table 20. Selected CEC-Adjudicated Adverse Events through 12 Months – Full Randomized Cohort ITT Population.

| Event | Summary Statistics | |
| --- | --- | --- |
| | Device Arm (N=285)* | Control Arm (N=287)† |
| All-cause mortality | 8.1% (23, 23, 0, 0, 1) | 7.0% (20, 20, 0) |
| Cardiovascular (VARC II definition) | 5.3% (15, 15, 0, 0, 0) | 3.8% (11, 11, 0) |
| Heart failure-related | 3.9% (11, 11, 0, 0, 0) | 2.8% (8, 8, 0) |
| Non-heart failure-related | 1.4% (4, 4, 0, 0, 0) | 1.0% (3, 3, 0) |
| Non-cardiovascular (VARC II definition) | 2.8% (8, 8, 0, 0, 1) | 3.1% (9, 9, 0) |
| Hospitalization | 33.7% (161, 96, 2, 7, 9) | 31.0% (155, 89, 0) |
| Heart failure hospitalization | 11.2% (44, 32, 1, 2, 0) | 11.8% (48, 34, 0) |
| Other cardiovascular hospitalization | 7.7% (23, 22, 1, 5, 0) | 7.0% (25, 20, 0) |
| Non-cardiovascular hospitalization | 22.8% (94, 65, 0, 0, 9) | 19.5% (82, 56, 0) |
| Tricuspid valve surgery | 1.8% (5, 5, 2, 2, 0) | 2.4% (7, 7, 0) |
| Tricuspid valve intervention‡ | 2.5% (7, 7, 5, 7, 0) | 1.0% (3, 3, 0) |
| Major bleeding (≥BARC 3a)l | 3.2% (9, 9, 0, 3, 0) | 1.7% (5, 5, 0) |
| New onset renal failurel | 0.7% (2, 2, 0, 1, 0) | 0.3% (1, 1, 0) |

*Event rate (no. of events, no. of subjects, no. of device-related events, number of procedure-related events, number of COVID-19-related events); event rate = no./total no. Number of COVID-19-related events includes related or possibly related events; this excludes events with unknown relatedness.

PMA P230007: FDA Summary of Safety and Effectiveness Data
40 of 51

{40}

†Event rate (no. of events, no. of subjects, number of COVID-19-related events).
Tricuspid valve intervention includes reintervention for device group and first intervention for control group.
Per the study CEC charter bleeding and new onset renal failure were adjudicated up to 30 days post treatment visit for the device and control groups.
VARC: Valve Academic Research Consortium; BARC: Bleeding Academic Research Consortium

## 6. Single-Arm Cohort Results

### Primary Endpoint:

There were 100 subjects with an attempted TriClip procedure in the Single-Arm Cohort. The primary analysis was performed on 91 subjects, which excluded subjects who withdrew (n=1), died or were hospitalized due to COVID-19 (n=2), or missed the 12-month visit or did not complete the 12-month KCCQ assessment (n=6). The results of the primary analysis are shown in Table 19. Fifteen (15) subjects died prior to 12 months, 34 had a KCCQ score improvement of &lt;10 points, and 42 survived with a KCCQ score improvement of ≥10 points at 12 months. The proportion of subjects who survived and experienced at least a 10-point improvement in KCCQ score at 12 months from baseline was 46.2%, with a lower 98.75% confidence limit of 34.3%, which exceeded the performance goal of 30%. Thus, the primary endpoint was met.

Table 215. Primary Analysis Results – Single-Arm Cohort.

| Primary Endpoint | Rate | Lower 98.75% Confidence Limit | Performance Goal | P-value | Result |
| --- | --- | --- | --- | --- | --- |
| Survival with ≥10 point improvement vs. baseline in KCCQ score at 12 months | 46.2% (42/91) | 34.3% | 30% | 0.008 | Endpoint Met |

### Secondary Endpoint:

The results of the powered secondary endpoints for the Single-Arm Cohort are summarized in Table 22. TR reduction by at least one grade at 30 days post-procedure occurred in 98.9% of subjects, and freedom from MAEs at 30 days post-procedure occurred in 100% of subjects; these endpoints were met. However, the improvement in 6MWD at 12 months from baseline (13.7±92.7) did not meet the performance goal, so the endpoint was not met. As a result, the subsequent endpoints in the pre-defined hierarchy (freedom from all-cause mortality or tricuspid valve surgery and recurrent HF hospitalizations at 12 months post-procedure) were not hypothesis-tested. Descriptively, the annualized HF hospitalization rates pre- and post-TriClip procedure were generally similar.

PMA P230007: FDA Summary of Safety and Effectiveness Data
41 of 51

{41}

Table 226. Summary of Powered Secondary Endpoints – Single-Arm Cohort AP Population.

| Order | Secondary Endpoint | Summary Statistics | p-Value | Result |
| --- | --- | --- | --- | --- |
| 1 | TR reduction by at least one grade at 30 days post-procedure | 98.9% (87/88)* | < 0.0001 | Endpoint met |
| 2 | Freedom from MAEs at 30 days post-procedure | 100% (99/99)* | <0.0001 | Endpoint met |
| 3 | Change in 6MWD at 12 months from baseline (m) | 13.7±92.7 (71)†
95% CI: [-8.3, 35.6] | 0.1090 | Endpoint not met |
| 4 | Freedom from all-cause mortality and tricuspid valve surgery at 12 months | 83.7% (3.7%)‡ | - | Not tested |
| 5 | Recurrent H…