XPS™ with Steen Solution™ Perfusate

P180014 · Xvivo Perfusion, Inc. · PHO · Apr 26, 2019 · Gastroenterology, Urology

Device Facts

• Record ID: P180014
• Device Name: XPS™ with Steen Solution™ Perfusate
• Applicant: Xvivo Perfusion, Inc.
• Product Code: PHO · Gastroenterology, Urology
• Decision Date: Apr 26, 2019
• Decision: APPR
• Device Class: Class 3
• Attributes: Therapeutic

Indications for Use

The XVIVO Perfusion System (XPS™) with STEEN Solution™ Perfusate is indicated for use in flushing and temporary continuous normothermic machine perfusion of initially unacceptable excised donor lungs during which time the ex vivo function of the lungs can be reassessed for transplantation.

Device Story

The XPS™ system provides normothermic ex vivo lung perfusion (EVLP) to reassess initially unacceptable donor lungs. The system includes a perfusion cart, centrifugal pump, heater/cooler, ventilator, and gas monitor. Donor lungs are cannulated and placed in an organ chamber; STEEN Solution™ perfusate is circulated through the lungs while the Hamilton C2 ventilator provides pressure-controlled ventilation. The system monitors flow, pressure, temperature, and gas parameters (pH, PCO2, PO2) in real-time. Surgeons use these parameters and visual inspection to determine if lungs meet transplantability criteria (e.g., delta PO2 >= 350 mmHg, stable compliance). If acceptable, lungs are cooled and transported for transplantation. The device is used in transplant centers by surgeons and clinical staff. By enabling the assessment and potential optimization of marginal donor lungs, the device increases the available donor pool for patients on transplant waiting lists.

Clinical Evidence

Evidence from two prospective, non-randomized clinical trials (HELP and NOVEL/NOVEL Extension). NOVEL cohort (N=226) compared EVLP-treated marginal lungs (n=110) to standard-criteria lungs (n=116). Primary endpoints (1-year survival, 72-hour PGD Grade 3) did not meet pre-specified 12% non-inferiority margins. 1-year all-cause survival was 86% (EVLP) vs 94% (Control). 72-hour PGD Grade 3 rates were 16% (EVLP) vs 9% (Control) after adjudication. Post-hoc comparisons to UNOS registry and LTOG dataset showed comparable long-term survival and PGD rates. Study demonstrated increased utilization of marginal/DCD donor lungs.

Technological Characteristics

System includes XPS Perfusion Cart, CardioHelp centrifugal pump, Hirtz VarioTherm 550 heater/cooler, Hamilton C2 ventilator, and PGM gas monitor. Disposables include STEEN Solution (physiological salt solution with HSA and dextran 40), XVIVO Lung Cannula Set, Disposable Lung Circuit, and Organ Chamber. Materials are biocompatible per ISO 10993. Sterilization via aseptic filtration (STEEN) or ethylene oxide (disposables). Software is major level of concern. Connectivity includes touchscreen CPU for data stream transfer and motor control.

Indications for Use

Indicated for flushing and temporary continuous normothermic machine perfusion of initially unacceptable excised donor lungs to reassess function for transplantation in patients 18 years or older requiring single or bilateral lung transplant.

Regulatory Classification

Identification

Preservation of donor lungs for transplantation.

Reference Devices

• TransMedics Organ Care (OCS™) Lung System (P160013)
• CardioHelp System (K102726)
• C2 ventilator (K092148)
• MAQUET Rotaflow (K991864)

Submission Summary (Full Text)

{0}

SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)

I. GENERAL INFORMATION

Device Generic Name: Ex Vivo Lung Perfusion (EVLP)

Device Trade Name: XVIVO Perfusion System (XPS™) with STEEN Solution™ Perfusate

Device Procode: PHO

Applicant's Name/Address: XVIVO Perfusion, Inc.
3666 S. Inca Street
Englewood, CO 80110

Date of Panel Recommendation: None

Premarket Approval Application Number: P180014

Date of Notice of Approval: April 26, 2019

II. INDICATIONS FOR USE

The XVIVO Perfusion System (XPS™) with STEEN Solution™ Perfusate is indicated for use in flushing and temporary continuous normothermic machine perfusion of initially unacceptable excised donor lungs during which time the ex vivo function of the lungs can be reassessed for transplantation.

III. CONTRAINDICATIONS

There are no known contraindications.

IV. WARNINGS AND PRECAUTIONS

The Warnings and Precautions can be found in the XPS™ with STEEN Solution™ Perfusate labeling (Instructions for Use).

V. DEVICE DESCRIPTION

A. Overview of the Device System

The XPS™ with STEEN Solution™ Perfusate consists of the XPS Perfusion Cart Hardware, fluid path and non-fluid path disposables, XPS Cart Software, and STEEN Solution™. The STEEN Solution™ is a clear, sterile, non-pyrogenic, non-toxic physiological salt solution containing human serum albumin (HSA) and dextran 40. This solution is an extracellular (low potassium) electrolyte solution with physiological colloid-osmotic pressure (COP) designed for use as a temporary continuous normothermic machine perfusion solution for ex vivo assessment of isolated lungs after removal from the donor.

PMA P180014: FDA Summary of Safety and Effectiveness Data

{1}

# B. Device System Component Description

XPS Perfusion Cart Hardware

The XPS Perfusion Cart is designed with the sub-assembly parts shown in Figure 1.

Figure 1 - XVIVO Perfusion System (XPS™)


CardioHelp XVIVO Centrifugal Pump

The CardioHelp XVIVO is a centrifugal pump with bubble, level, flow, temperature, and pressure sensors, and is identical in function to the $510(\mathrm{k})$ cleared CardioHelp System (K102726). It pumps the STEEN Solution™ into the lung(s) and monitors the temperatures and pressures going into and coming out from the ex vivo lung(s).

# Heater/Cooler

The Hirtz VarioTherm 550 Heater/Cooler Unit (HCU) provides water at a set water temperature that flows into the medical device heat exchange interface to create the normothermic environment during EVLP. The HCU pumps water to the Quadrox-iR heat exchange membrane to control the temperature of the STEEN Solution™ perfusate through conduction. The HCU water does not come in contact with the STEEN Solution™ or any other portion of the aseptic fluid path. It remains on the non-aseptic side of the heat exchange membrane of the Quadrox-iR.

PMA P180014: FDA Summary of Safety and Effectiveness Data

{2}

C2 Ventilator

The Hamilton C2 ventilator is an Intensive Care Unit (ICU) pressure-controlled ventilator used to ventilate the lungs during ex vivo perfusion and is identical to the 510(k) cleared C2 ventilator (K092148). It allows the user to pre-set pressure and volume limits according to the established EVLP ventilation protocols, preventing the C2 ventilator from over-ventilating and, therefore, damaging the ex vivo lung.

Perfusate Gas Monitor (PGM)

The PGM is an in-line trending monitor that measures the following critical gas parameters in the circulating STEEN Solution™: pH, PCO₂ (partial pressure of dissolved CO₂), and PO₂ (partial pressure of dissolved O₂). These parameters are displayed in real time to the operator. The PGM has no direct contact with the sterile fluid path. It uses fluorescent Light-Emitting Diodes (LED) light transmission through an in-line disposable device that contains pre-calibrated sensors.

STEEN Solution™ Peristaltic Pumps

The three (3) pumps aseptically move STEEN Solution™ to 1) fill the hard shell reservoir during priming; 2) remove it from the perfusion circuit into a connected drain bag; and 3) recycle it back to the reservoir from the XVIVO Organ Chamber™.

Monitors & Controls

- The AAEON medical grade touchscreen monitor has an integrated computer central processing unit (CPU) to connect to the XVIVO Perfusion Cart hardware for data stream transfer for display purposes. In addition, the computer Central Processing Unit (CPU) connects to the three (3) Allied Motion peristaltic pump motors to display data as well as control motor function (on/off, low/medium/high speed).
- The AAEON medical grade display-only monitor (no touch) provides data stream information to the surgeon/sterile side of the XVIVO Perfusion Cart.

Software

The XVIVO Perfusion Cart Software comprises the software system that resides on the AAEON Computer/Touchscreen Display and provides the following:

- Data stream displays from CardioHelp XVIVO, Hamilton C2 ventilator, and XVIVO PGM;
- Control and data display of Allied Motion peristaltic pump motors.

XVIVO Organ Chamber™ platform

The hinged table is attached to the sterile side of the XVIVO Perfusion Cart and locks in place in the horizontal position to provide a location to set the XVIVO

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 3

{3}

Organ Chamber™.

## Gas Cylinders

The perfusion cart has two (2) gas cylinders, one containing medical grade (100%) oxygen for membrane oxygenation and the other containing a mixture of medical grade gases (6% O₂, 8% CO₂, 86% N₂) for membrane deoxygenation.

## Power Distribution & Subsystem

The power subsystem assembly provides power and backup power to critical hardware items in the XVIVO Perfusion Cart. The subassembly is made up of the following:

- The UPS (uninterruptable power supply) provides battery backup support to the AAEON display and touchscreen monitors and PGMs. The CardioHelp XVIVO and the Hamilton C2 ventilator have their own internal battery backups. The Hirtz Variotherm 550 heater/cooler consumes too much power to run on battery power, so in case of emergency power outage, this device is not supported.
- The Power Supply is the Synqor +24 Volt DC power supply and is capable of supplying up to 400W of power.
- The Isolation Transformer is the Powertronix and it is used to protect equipment from power spikes and to filter out electrical interference.

## Single Use Disposables

The XPS™ System interfaces with single-use disposable products, including STEEN Solution™, the XVIVO Lung Cannula Set™, the XVIVO Disposable Lung Circuit™, and the XVIVO Organ Chamber™. They are designed to interact safely and aseptically with the fluid path during EVLP.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 4

{4}



Figure 2 - XVIVO Perfusion Disposable Components


# C. Safety Elements

A number of safety elements are incorporated into the XPS with STEEN Solution Perfusate device, including:

- Audible and visual alarms indicating perfusate flow, device status, and connections to software and battery status;
- Battery for alarms in the event that both primary power and batteries fail;
- Keyed connectors for all cable, console, and disposable connections;
24 hour, 365 days per year technical support;
Detailed directions for use;
Device/system training.

All system components, with the exception of the single-use items, are intended for use on multiple ex vivo lungs. These components can be used for multiple ex vivo lungs, but only on one set of donor lungs at a time (the device can treat single and double lungs). The $\mathrm{XPS^{TM}}$ with STEEN Solution™ Perfusate device is intended for

PMA P180014: FDA Summary of Safety and Effectiveness Data

{5}

use in an aseptic setting to provide mechanical circulatory support during EVLP assessment.

## VI. ALTERNATIVE PRACTICES AND PROCEDURES

There are several other alternatives for the preservation of donor lungs. Each alternative has its own advantages and disadvantages. A patient should fully discuss these alternatives with his/her physician to select the method that best meets expectations and lifestyle. Standard of care procedures used in the preservation of standard-criteria or ideal donor lungs typically consist of the cold, static storage of the lungs in a hypothermic preservation solution prior to transplantation. Normothermic machine perfusion of standard-criteria double lungs is available with the TransMedics Organ Care (OCSTM) Lung System, which was approved in 2018 under P160013. Other options are not to transplant, which would mean the patient would remain on the transplant waiting list and would undergo mechanical ventilation and/or extracorporeal membrane oxygenation, if necessary. Cold, static storage is sometimes used for non-ideal lungs under practice of medicine at the discretion of the transplant surgeon.

The XVIVO XPS™ System and STEEN Solution Perfusate™ received approval as a humanitarian device under H120003 on August 12, 2014, for the normothermic flushing and assessment of previously unacceptable excised donor lungs.

## VII. MARKETING HISTORY

The XVIVO Perfusion System (XPS™) and STEEN Solution Perfusate™ has been marketed in the United States (U.S.) since 2014, having gained marking approval under H120003. The STEEN Solution Perfusate™ obtained CE marking in 2006 and became available for use with commercially available cardio-pulmonary by-pass circuit equipment. Australian Therapeutic Goods Administration (TGA) approval was obtained in 2009. Hundreds of EVLP transplants using the XTS™ System and STEEN Solution™ have been performed worldwide. STEEN Solution Perfusate™ received approval by Health Canada on November 6, 2012. Including the EVLP transplants performed in the clinical trial, Toronto General Hospital has transplanted over 100 patients with EVLP lungs. The XPS™ System, XVIVO Organ Chamber™, XVIVO Lung Cannula Set™, and XVIVO Disposable Lung Circuit™ obtained their CE marking in 2014. In addition, these devices are commercially available and marketed in Australia and Canada.

None of these devices have been withdrawn from marketing for any reasons related to the safety and effectiveness of these devices.

Several device changes have been implemented since the approval of the H120003 in August of 2014. These are summarized in the table below.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 6

{6}

Table 1 – Device Modifications Since 2014

| Device Modification Description | HDE Supplement | FDA Decision and Date |
| --- | --- | --- |
| Device Modification (Software Changes) | H120003/S001 | Approval; 11/10/2014 |
| Device Modification (Hardware Changes) | H120003/S002 | Approval; 11/10/2014 |
| STEEN Manufacturing Site Change | H120003/S003 | Approval; 6/2/2015 |
| Labeling Change (Revision K of Operator’s Manual) | H120003/S005 | Approval; 11/16/15 |
| Filter Change | H120003/S006 | Approval; 11/13/15 |
| Software Modification (Version 3.3 of XPS™ System) | H120003/S007 | Approval; 1/7/2016 |
| Software Modification (Version 4.0 of XPS™ System) | H120003/S008 | Approval; 11/16/2016 |
| Change in Sterilization Site | H120003/S009 | Acknowledgement; 12/12/18 |

As can be seen, these changes were submitted to FDA for review and received approval.

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

The XPS™ with STEEN Solution™ Perfusate device is indicated for use only on previously unacceptable excised donor lungs in an ex vivo setting. There is no direct patient contact when this device is used as labeled; however, the device has a direct contact with the lungs that are subsequently transplanted into the recipients. The donor lung quality and optimization after preservation have direct effects on allograft function and survival.

The potential for contamination and mechanical trauma, due to the manipulation and cannulation of the lung airway and vascular structures, may lead to complications after transplantation.

Patients receiving a lung treated with the XPS™ System with STEEN Solution™ Perfusate device may experience adverse events including those experienced with any lung transplant.

Below is a list of the potential adverse effects associated with the use of the device.

- Death
- Renal failure or dysfunction
- Respiratory dysfunction or failure
- Respiratory infection
- Sepsis

PMA P180014: FDA Summary of Safety and Effectiveness Data

{7}

- Primary graft dysfunction
- Acute or chronic rejection
Cardiac arrhythmias
- Bronchiolitis Obliterans Syndrome (BOS)
- Bronchial stenosis/Dehiscence

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

## IX. SUMMARY OF NONCLINICAL STUDIES

### A. Laboratory Studies

The bench testing consisted of performance, safety and reliability testing.

#### 1. Biocompatibility

Biocompatibility testing of the XPS™ disposables and STEEN Solution™ was performed in accordance with the FDA Blue Book Memorandum #G95-1 and Biological Evaluation of Medical Devices Guidance - International Standard ISO 10993-1, and in accordance with United States Pharmacopoeia – XXIII. The specific tests included: cytotoxicity, sensitization, intracutaneous irritation, systemic toxicity, hemocompatibility, endotoxin, and sub-chronic toxicity.

#### a. STEEN Solution™

Biocompatibility testing according to ISO 10993, Part 1, was performed on STEEN Solution™. The results showing it is a biocompatible product are provided in Table 2.

Table 2 - STEEN Solution™ Biocompatibility Matrix

| Experimental Study | Results |
| --- | --- |
| Cytotoxicity study using the ISO agarose overlay method, liquid-macroscopic and microscopic evaluation of mouse fibroblast cell culture. | STEEN Solution™ showed no evidence of causing cell lysis or toxicity and conforms to the relevant sections of ISO 10993: Biological evaluation of medical devices part 5: Test for cytotoxicity In Vitro Method. |
| ISO modified intracutaneous study of the rabbit modified for a chemical solution. | STEEN Solution™, injected intracutaneously into rabbits, showed no evidence of causing significant irritation and conforms to relevant sections of ISO 10993: Biological evaluation of medical devices part 10: Tests for Irritation and Sensitization- modified for chemical solutions. |

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 8

{8}

| Experimental Study | Results |
| --- | --- |
| Acute systemic toxicity study following IV dose range finding/ limit dose study in the mouse. | STEEN SolutionTM showed no evidence of mortality or significant systemic toxicity and conforms to relevant sections of ISO 10993: Biological evaluation of medical devices part 11: Tests for Systemic Toxicity. |
| Murine local lymph node assay by topically dosing the dorsum of the mouse ear. | STEEN SolutionTM was not considered to be sensitizing to the mouse and conforms to relevant sections of ISO 10993: Biological evaluation of medical devices part 10: Tests for Irritation and Sensitization- modified for chemical solutions. |
| In vitro hemolysis study (modified ASTM-direct contact method) of diluted rabbit blood. | STEEN Solution was nonhemolytic and conforms to relevant sections of ISO 10993: Biological evaluation of medical devices part 4: Selection of Tests for Interactions with Blood. |
| White blood cell in vitro morphology study of anticoagulated whole canine blood. | STEEN SolutionTM did not have an effect upon white blood cell morphology and conforms to relevant sections of ISO 10993: Biological evaluation of medical devices part 4: Selection of Tests for Interactions with Blood. |
| In Vitro Lee-White clotting time study of canine blood. | STEEN Solution appeared to have no effect on clotting time according to the study and conforms to relevant sections of ISO 10993: Biological evaluation of medical devices part 4: Selection of Tests for Interactions with Blood. |

b. XVIVO Organ Chamber™ XVIVO Lung Cannula Set™, XVIVO Disposable Lung Circuit™, and XVIVO Disposable PGM Sensors™

Each of these devices was extracted and tested under Good Laboratory Practices (GLP) conditions in accordance with ISO 10993 standards, showing all materials are biocompatible, as listed in the Table 3, below.

Table 3 - XVIVO Disposables Plastics Biocompatibility Matrix

| Subject | Standard/Method | Pass/Fail |
| --- | --- | --- |
| Cytotoxicity | MEM Elution, ISO 10993-5 | Pass |
| Sensitization | Murine Local Lymph Node Assay (LLNA), ISO 10993-12 | Pass |
| ISO Intracutaneous Reactivity/ Toxicity | Albino rabbits, intracutaneous injections, ISO 10993-10 | Pass |
| Systemic Toxicity | Material Mediated Pyrogen, ISO 10993-11 | Pass |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{9}

| Subject | Standard/Method | Pass/Fail |
| --- | --- | --- |
| Sub-chronic Toxicity | Systemic Injection, ISO 10993-11 | Pass |
| Hemocompatibility | ASTM Hemolysis, ISO 10993-12, ISO 10993-4 | Pass |
| Endotoxin | LAL Test, USP <85> ANSI/AAMI ST72:2002 | Pass |

2. Sterilization Validation

a. STEEN Solution™

The STEEN Solution™ is provided sterile to the user. The device is sterilized via aseptic filtration using a 0.20μm filter into sterile Nalgene bottles. The sterilization method was validated to ensure successful sterilization to a Sterility Assurance Level (SAL) of 10⁻³ in accordance with USP 32 &lt;71&gt; Sterility Tests (method for Membrane Filtration).

b. XVIVO Organ Chamber™, XVIVO Lung Cannula Set™, and XVIVO Disposable Lung Circuit™

These components are also provided sterile to the user. These devices were extracted and tested under GLP conditions in accordance with the American National Standards Institute, Inc. (ANSI) standard ANSI/AAMI/ISO 11135 (Medical Devices - Validation and Routine Control of Ethylene Oxide Sterilization). All tests passed and the products were sterilized by the validated SAL 10⁻⁶ ethylene oxide sterilization cycle.

3. Hemolysis Testing

STEEN Solution™ is an acellular (no red blood cells) perfusate and is used without adding blood to the perfusion circuit, minimizing any risk of hemolysis. The centrifugal pump head used during EVLP (MAQUET Rotaflow) has previously been shown (K991864) to minimize hemolysis and is comparable to other centrifugal pump devices marketed for use with blood products.

4. Software Verification and Validation

Software on-board the XPS™ Perfusion Cart was verified and validated in accordance with the FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices. The software is of major level of concern, since the viability of the donor lungs will be impacted, should device/software malfunctions occur. The software information and validation included a description of the software architecture, a Product Requirements Document, Design Specification document, Verification Protocol, Validation Protocol, Hazard Analysis and Risk Assessment Summary, Traceability Matrix,

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 10

{10}

Revision Level History, list of unresolved anomalies (bugs), compatibility testing between the various components of the XPS™ System, testing of alarms and alarm conditions and overall validation of device functionality.

# 5. Shelf Life Studies

A combination of real-time aging (STEEN Solution™, XVIVO Lung Disposable Circuit™) and accelerated aging studies (XVIVO Organ Chamber™, XVIVO Lung Cannula Set™) were performed in accordance with ASTM F1980. These studies demonstrated that sterility, package integrity, and product functionality could be maintained as follows:

STEEN Solution™: 2 years
XVIVO Disposable Lung Circuit™: 2 years
XVIVO Organ Chamber™: 4 years
XVIVO Disposable Lung Cannula Set™: 4 years

# 6. Electrical Safety Testing

An independent laboratory has evaluated the electrical safety of the XPS™ device. The test results demonstrate that the XPS™ System meets the applicable requirements of IEC 60601-1, the European standard for general safety requirements for medical electrical equipment, as summarized in Table 4.

Table 4 - Electrical Safety Testing Summary

| Standards and Approvals | |
| --- | --- |
| IEC 60601-1-1 | Medical electrical equipment – Part 1: General requirements for basic safety and essential performance Collateral Standard: Safety requirements for medical electrical systems |
| IEC 60601-1-2 | Medical electrical equipment – Part 1-2: Collateral Standard: Electromagnetic compatibility-Requirements and tests |
| IEC 60601-1-4 | Medical electrical equipment – Part 1-4: Collateral standard: Programmable electrical medical systems – Evidence checklist |
| IEC 60601-1-8 | Medical electrical equipment – Part 1-8: Collateral standard: General requirements, tests and guidance for alarm systems in medical electrical equipment and medical electrical systems |
| EN 1041 | Information supplied by the manufacturer with medical devices |
| EN 980 | Graphical symbols for use in the labeling of medical devices |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{11}

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 12

| Standards and Approvals | |
| --- | --- |
| ISO 15223 | Medical devices symbols to be used with medical device labels, labeling and information to be supplied |

7. Electromagnetic Compatibility (EMC) Testing

The XPS™ System was tested by an independent laboratory to demonstrate that it meets the requirements for conducted and radiated emissions, electrostatic discharge immunity, radiated electromagnetic immunity, electrical fast transient/burst immunity, and conducted disturbance induced by radio frequency fields. The test results demonstrated that the XPS™ System meets the applicable requirements of the 2001 version of IEC 60601-1-1-2, the standard for electromagnetic capability (EMC) for medical electrical equipment.

8. System Reliability

The reliability of the main components of the XPS™ System has been shown via the individual component manufacturers' data from the specific products, which are 510(k) cleared and CE marked. The main components are the following:

- CardioHelp pump (K102726, CE Marked)
- C2 Ventilator (K092148, CE Marked)
- Variotherm Heater/Cooler (CE Marked)

B. Animal Studies

1. Animal and Rejected Human Lung Testing

Three (3) porcine lungs and one (1) rejected human lung were perfused using the XPS™ System with STEEN Solution™ Perfusate on distinct dates under the direction of different transplant surgeons. Each perfusion was done under controlled conditions, using the procedure outlined in the Vitrolife/XVIVO “NOVEL LUNG TRIAL- Normothermic Ex Vivo Lung Perfusion (EVLP) as an Assessment of Extended/Marginal Donor Lungs (Protocol number VSS-NA-001).” The lungs were removed from the donor following standard lung recovery procedures (e.g., hypothermic flush with Perfadex® and placed in cold (ice) storage during transportation). Upon arrival at the test site, the lungs were removed from the hypothermic container and placed in a sterile basin for temporary storage. The straight pulmonary artery (PA) cannula from the XVIVO Lung Cannula Pack™ was selected and attached to the PA using umbilical tape to secure the cannula in place. The cone-shaped left atrial (LA) cannula was selected to connect the LA using a 4.0 running monofilament suture to provide effective connection for reliable outflow drainage. The perfusion tubing from the MAQUET disposable lung circuit was connected to the lungs using straight 3/8” hose connectors. The shape and size of the cannulas were designed to safely hold open the pulmonary artery and left atrial cuff to allow the fluid to move smoothly through the lung and to monitor the

{12}

pressures in the lung, while visualizing the flow of the solution. Extracorporeal circulation (i.e., flow) was provided by the $\mathrm{XPS^{TM}}$ System. Table 5 provides data obtained from the $\mathrm{XPS^{TM}}$ software indicating that the various components of the $\mathrm{XPS^{TM}}$ system were functioning properly.

Table 5 - Pre-Clinical Results

| DATE | Test Subject | CardioHelp Pump | C2 Ventilator | Variotherm HC U | PGM Sensors |
| --- | --- | --- | --- | --- | --- |
| 9/3/10 | Porcine | Pass | Pass | Pass | Pass |
| 9/4/10 | Porcine | Pass | Pass | Pass | Pass |
| 10/26/10 | Porcine | Pass | Pass | Pass | Pass |
| 12/6/10 | Human | Pass | Pass | Pass | Pass |

The results of the four (4) laboratory tests (three (3) porcine and one (1) human lung) show that the $\mathrm{XPS^{TM}}$ ventilator, pump (and associated disposables including deoxygenator membrane), and PGM worked safely and efficiently together during both animal (pig) and human lung perfusion tests and similarly to what was expected based on the published (and unpublished) data from the University Health Network, Toronto group during its human ex vivo lung perfusion ("HELP") clinical trial. The XVIVO Disposable Lung Circuit™ built by MAQUET Cardiopulmonary AG perfused both pig and human donor lungs in the $\mathrm{XPS^{TM}}$ System using the STEEN Solution™ Perfusate efficiently and safely with results within normal expected ranges.

# X. SUMMARY OF PRIMARY CLINICAL STUDIES

Data from two (2) clinical trials, the HELP Trial and the NOVEL Trial, were considered to support H120003 for the safety and probable benefit of EVLP when used to reassess initially unacceptable donor lungs perfused at near normal body temperature (normothermia) in an ex vivo setting. Both trials were sponsored by Vitrolife, Inc., which became XVIVO Perfusion, Inc., in late 2012.

Table 6 - Supporting Clinical Studies

| | EVLP-Transplanted | Cold Storage(Control) |
| --- | --- | --- |
| HELP Trial (Canadian Trial)*: Normothermic EVLP for an Improved Assessment of Donor Lungs for Transplantation | n= 50 | n= 253 |
| NOVEL and NOVEL Extension Trials (U.S. Trials): Normothermic EVLP as an Assessment of Extended/Marginal Donor Lungs | n= 110 | n= 116 |

* Cypel M., et al., 2012

PMA P180014: FDA Summary of Safety and Effectiveness Data

{13}

In the earlier Canadian Trial (HELP Study, 2008-2010, Toronto), STEEN Solution™ was perfused with available off-the-shelf equipment. This hardware and single-use disposable equipment set was functionally equivalent to the subsequent components of the XPS™ System and, in fact, provided a basis for the development of the XPS™ System. This study was not powered to show statistically significant differences in the predefined endpoints. A brief summary of the HELP Trial appears below.

Data from the U.S. clinical trial (NOVEL Trial, 2011-2013) were considered as the pivotal data to support the safety and probable benefit of EVLP using the XPS™ System with STEEN Solution™ Perfusate for H120003. The NOVEL study was expanded after the approval of the HDE application and the full cohort, comprised of the NOVEL (2011-2014) and the NOVEL Extension (2014-2017) studies, was the basis for the PMA approval decision. A summary of the NOVEL and NOVEL Extension is presented in subsection 2.A, below.

1. Canadian HELP Study (N=22, plus an additional compassionate use extension of 39, for a total N=61)

HELP Trial Study Design

The HELP study was a prospective, non-randomized, single-center study that reviewed clinical outcomes between initially rejected donor lungs treated with four (4) hours of EVLP using STEEN Solution™ (study group) and all other lung transplants performed during the same study period and preserved using standard static cold storage (CS) methods with Perfadex™ Preservation Solution (Control Group).

Initially rejected lungs were defined as those not meeting the clinical donor lung criteria, based on the 2003 International Society of Heart and Lung Transplantation (ISHLT) consensus document on lung transplant acceptability criteria (Orens et al., 2003).

After four (4) hours of EVLP perfusion, the donor lung was evaluated for a delta PO₂ &gt; 350mmHg and stable pulmonary vascular resistance (PVR), peak airway pressure, and lung compliance (i.e., &lt; 15% deterioration). If meeting these transplantability criteria, the donor lungs were considered acceptable for transplantation.

During the initial phase, this study included three (3) standard criteria lung transplants in a safety pilot study. In addition, the study included 19 initially unacceptable lung donors for transplantation. A subsequent compassionate use extension arm was added to the study, increasing the sample size with 39 additional patients for a total of 61. Data from the study were reported to FDA at various stages of the HELP study, as they became available during the review of H120003, and were also published by different authors at different times, thus the sample sizes in the various analyses are not consistent, as seen in the tables below. All the included donor lungs were transplanted after EVLP normothermic preservation.

Donor/recipient selection was based on first available lungs that did not meet the

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 14

{14}

criteria for standard, "ideal" donor lung (Orens et al., 2003, if not 'standard,' proceed through EVLP), and recipient match.

A study design limitation, which resulted from ethical considerations, was the inability to randomize the initially rejected donor lungs to 'EVLP' or static cold storage.

The study's primary endpoint was the incidence of primary graft dysfunction (PGD) Grades 2 and 3 at 72 hours after transplantation. The study was not powered to demonstrate statistical differences across study groups for the endpoints.

# HELP Trial Results

# Primary Graft Dysfunction Grades

The primary endpoint in the study (e.g., incidence of PGD Grades 2 and 3 at 72 hours after transplantation) showed that the EVLP recipient group had no significant difference in comparison to those in the Control Group. PGD Grade 2 at 72 hours was $11\%$ and $23\%$ in the EVLP and Control Arms, respectively, while PGD Grade 3 at 72 hours was $3\%$ and $11\%$ in the EVLP and Control Arms, respectively (see Table 7). Similarly, Cypel et al., 2012, reported that PGD Grade 3 at 72 hours was $2\%$ and $8.5\%$ in the EVLP and control arms, respectively (see Table 8).

Table 7 - PGD Grades, HELP Trial

| Toronto General Hospital Patients - PGD | | | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| PGD Grade | Controls N=103 | | | EVLP N=35 | | |
| | T 0 hr | T 24 hrs | T 72hrs | T 0hr | T 24hrs | T 72hrs |
| 1 | 72 | 55 | 63 (61%) | 25 | 28 | 30 (86%) |
| 2 | 16 | 33 | 24 (23%) | 5 | 5 | 4 (11%) |
| 3 | 15 | 9 | 11 (11%) | 5 | 2 | 1 (3%) |
| No value obtained | 0 | 6 | 5 | 0 | 0 | 0 |

Note: Extubated patients were not given a PGD score

PMA P180014: FDA Summary of Safety and Effectiveness Data

{15}

Table 8 - Recipient Outcomes in ex vivo Lung Perfusion, HELP Trial

| Variable | EVLP (n=50) | Controls (n=253) | P value |
| --- | --- | --- | --- |
| PGD 3 at 72h (%) | 2 | 8.5 | 0.14 |
| ECLS (%) | 2 | 2.7 | 1.00 |
| Mechanical ventilation (d) | | | 0.30 |
| Median | 2 | 2.2 | |
| Range | 1-101 | 1-43 | |
| ICU stay (d) | | | 0.32 |
| Median | 4 | 4.5 | |
| Range | 1-100 | 1-257 | |
| Hospital stay (d) | | | 0.11 |
| Median | 20 | 23 | |
| Range | 7-156 | 1-299 | |
| 30-d mortality (%) | 4 | 3.5 | 1.00 |
| Anastomotic stricture requiring intervention (%) | 4 | 4 | 1.00 |

EVLP, Ex vivo lung perfusion; PGD, primary graft dysfunction; ECLS, extracorporeal life support; ICU, intensive care unit
Source: Table obtained from Cypel et. al., 2012, includes "compassionate extension."
Controls: Standard static cold storage.

## Survival Analyses

Thirty-day mortality was reported as 4% and 3.5% for the EVLP and control arms, respectively (p=1.0). Table 9 below presents survival data at 1, 2, and 3 years post-transplant.

Survival at 3 years was comparable across arms, 67% (n=28) versus 71.2% (n=163) for the EVLP and control arms, respectively. The early deaths in the study group were attributed to postoperative complications (i.e., retroperitoneal bleeding and sepsis). It was concluded that these complications "were not directly related to the allograft."

It should be noted that the total number of recipients in the HELP trial's EVLP arm was 61, but at 3 years, only 28 were included in the survival analysis. The status of the remaining 33 patients is unknown (see Table 9). Finally, since the data were obtained from the Applicant's HDE application (H120003), as well as from the cited publications, there are some discrepancies in the patient numbers, since different analyses included different sample sizes.

PMA P180014: FDA Summary of Safety and Effectiveness Data

{16}

Table 9 - HELP Study Survival Outcomes and Highest Predicted FEV1 Data

| | EVLP | N | Control | N | Significance |
| --- | --- | --- | --- | --- | --- |
| Survival 1 year | 83.7% | 49 | 85.1% | 262 | P=0.83 (F) |
| Survival 2 years | 75.0% | 44 | 78.4% | 236 | P=0.69 (F) |
| Survival 3 years | 67.9% | 28 | 71.2% | 163 | P=0.82 (F) |
| Number of acute rejections/year | 0.54±0.72 | 39 | 0.47±0.65 | 204 | P=0.54 (MW) |
| Highest Predicted FEV1 (only double lungs) | 73.5%±28% | 35 | 71.8%±25% | 220 | P=0.67 (ST) |

F=Fisher's exact test; MW=Mann-Whitney; ST=Student's T-test.

The updated survival data from Toronto General Hospital is listed in Table 10 below.

Table 10 - HELP Survival (last follow up - May 24, 2013*)

| Toronto General Hospital Patients-Survival | | |
| --- | --- | --- |
| | Control (N=397) | EVLP (N=74) |
| Alive | 309 (78%) | 57 (77%) |
| Expired | 88 (22%) | 17 (23%) |
| Survival Day (Range) | Mean 687 (1-1709) | Mean 629 (7-1702) |
| | Median 597 | Median 531 |

* The HELP trial was conducted at the Toronto General Hospital, University of Toronto, Canada, from September 25, 2008 to February 28, 2010, and the results from this study were published in Cypel et al., 2010. The HELP Study included 22 lung transplant recipients of EVLP lungs during its "initial phase," and subsequently added 39 more transplanted patients during the "compassionate extension phase" of the study. The update on this study was published by Cypel et.al., 2012.

## Allograft Function Analyses

Pulmonary function test (PFT) data over time were not available, limiting FDA's ability to draw valid conclusions. The prospective collection of PFT data was not part of the original HELP study protocol. The limited data available for H120003 included only the highest predicted FEV1 (\%) on double lung transplants (Table 9).

PMA P180014: FDA Summary of Safety and Effectiveness Data

{17}

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 18

# 2. NOVEL and NOVEL Extension Study

## A. Study Design

The main clinical study used in support this PMA application was the NOVEL study, including its NOVEL Extension portion. The NOVEL study is the Applicant's original clinical study for the XPS™ System, dating back to the approval of the original IDE (G100104) in March of 2011. A cohort from that study was used in the approval of the Applicant's HDE H120003. Upon the approval of the HDE in August of 2014, FDA and the Applicant worked to modify and expand the NOVEL clinical protocol to collect data for this PMA, as well as to satisfy the post-approval study (PAS) requirements of the HDE approval. As such, the study was expanded to encompass 220 subjects (110 in each arm), which would be the cohort used for the PMA application, and 20 investigational sites. This PMA cohort would include the 62 subjects (31 in each arm) that had already been submitted in H120003. In this PMA application, the Applicant has designated the study prior to these modifications and expansions the "NOVEL study," while the portion after the modifications were implemented has been designated the "NOVEL Extension study."

The study remained as a controlled, non-randomized study with the Control Arm being standard-criteria lungs stored and transported with standard-of-care (SOC), cold, static, preservation solution. The Control lungs were contemporaneous to the EVLP lungs and were transplanted at the same centers as those from the EVLP arm. The timeframe of the study was through the first year after transplantation, although the HDE's PAS, which was built into the same protocol as the NOVEL study, followed patients to 3 years post-transplantation.

Other modifications included in the NOVEL study protocol at the time of H120003 approval included changes in the primary endpoints, changes in the definitions and categorization of adverse events, changes in the follow-up period (follow-up was increased to three years post-transplantation), some changes in the study entry criteria, and some changes in the acceptability criteria for EVLP, as well as the transplantability criteria post-EVLP. These protocol changes were largely informed by the HDE application and discussions held at the GU Advisory Panel meeting for the HDE.

Additionally, there were 16 more subjects per arm that were enrolled for the purposes of the HDE's PAS. In total, the PAS would encompass 252 subjects, 126 in each arm.

Patients in the NOVEL and NOVEL Extension were treated between 2011 and 2017. The database for this PMA reflected data collected through July 2018 and included 226 patients, in addition to the 16 HDE PAS subjects. There were 17 investigational sites that participated in the study.

The study evaluated the EVLP treatment with the XPS™ system of double and single lungs. In some cases, double lungs were instrumented in the system and

{18}

treated, but were then split and transplanted in different recipients. As such, a total of 332 donor grafts, single or double, were enrolled into the study, resulting in 110 transplanted recipients, 106 EVLP-treated donor lungs that were deemed not transplantable after EVLP, and 116 transplanted control recipients. Eleven (11) donor double lungs were split resulting in 22 single-lung recipients.

Also, it is important to reiterate that the study focused on the treatment of previously unsuitable or rejected donor lungs, considered to be non-ideal, marginal or extended-criteria. Control lungs were standard-criteria donor lungs that were transported and preserved with cold, static storage.

Once accepted for EVLP, the donor lungs would be flushed with cold preservation solution (Perfadex), packaged according to industry standards, and transported on ice to the recipient/transplant center. The XPS™ System resides at the transplant site and does not travel with the donor lungs; therefore, the donor lungs must first travel on ice and then undergo EVLP once they reach the transplant site.

Once received by the transplant site, the lungs would be unpacked, the EVLP cannula would be sutured to the left atrial cuff and pulmonary artery, and the lungs would then be placed on the XPS™ system to begin the EVLP treatment. The EVLP treatment includes antegrade perfusion to remove any clots that might have formed in the pulmonary artery during transport. Graft preparation time is measured from the time of unpacking to the start of antegrade perfusion. The lungs would be warmed and perfused on the EVLP circuit for a minimum of 3 hours and a maximum of 6 hours. Physiological parameters would be collected every hour and x-rays would be taken at 1 hour and possibly later in the EVLP period, if requested by the transplant team. The x-ray provides secondary confirmation of improvement if the reason for initial rejection was pulmonary edema. Edema could also be evaluated by the surgeon (e.g., by lifting the lung to determine whether it has become less boggy and heavy, or by visually inspecting the frothing coming from the lung).

After EVLP, the lungs would be assessed for transplantability and deemed acceptable for transplantation or not. If accepted for transplantation, the lungs would be taken off of the XPS™ System, cooled and placed back on ice in the standard, sterile method used for organ storage. According to the Applicant, this second cooling period is necessary to decrease the chance of lung degradation during the implantation procedure.

1. Clinical Inclusion and Exclusion Criteria

The following pre-EVLP eligibility criteria were to be followed for donor lungs:

The donor lung must meet the following inclusion criteria to proceed with EVLP:

- PaO2/FiO2 ≤ 300 mmHg at the time of clinical evaluation, OR

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 19

{19}

- If PaO2/FiO2 &gt; 300 mmHg, the donor must have one or more of the following risk factors:
- Multiple blood transfusions (&gt; 10)
- Pulmonary edema detected via chest x-ray, bronchoscopy or palpation of lungs.
- Donation after circulatory death (DCD).
- Investigator evaluation of the donor lung as “unsuitable” for transplant.

Donor lungs were not permitted to enroll in the study if they met any of the following exclusion criteria:

- Significant active pneumonia and/or persistent purulent secretions on bronchoscopy or as determined by investigator.
- Known significant aspiration of gastric contents within the lung.
- Significant mechanical lung injury or trauma determined by chest x-ray, bronchoscopy, CT scan or visual inspection.
- Active infectious disease such as HIV, hepatitis B or C, or syphilis. (If infectious disease information is not available at the start of EVLP, this criterion can be assessed during or after EVLP but prior to transplant.)

The following post-EVLP transplantability criteria were used in order to proceed to transplant:

- Surgeon must be clinically satisfied with the lung evaluation.
- Stability or improvement in all lung function parameters (PVR, compliance, airway pressure) during perfusion.
- ΔPO2 ≥ 350 mmHg at two time points during EVLP.

If two (2) ΔPO2 ≥ 350 mmHg could not be obtained, adaptive eligibility criteria could be used. At least three (3) of the four (4) following criteria would need to be met:

- One ΔPO2 ≥ 350 mmHg or absolute PO2 ≥ 400 mmHg.
- Chest x-ray findings with absence or improvement of pulmonary edema/infiltrates.
- Static compliance &gt; 35 for a single lung or &gt; 60 for double lungs.
- Absence of consolidation by palpation.

The adaptive eligibility criteria were introduced in the NOVEL Extension portion of the study in 2014 after the HDE’s approval.

The donor lung would be excluded from transplant if any of the following criteria were met:

- All ΔPO2s &lt; 350 mmHg (measured with FiO2 set at 1.0) or all absolute PO2s are &lt; 400 mmHg.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 20

{20}

- Greater than 10-15% overall deterioration of lung function across all parameters (PVR, compliance, airway pressure) with chest x-ray findings showing deterioration.
- Donor lung is positive for infectious diseases such as HIV, hepatitis B or C, or syphilis.

Regarding the recipients, the protocol included the following recipient inclusion and exclusion criteria:

A recipient may enroll in the study if they meet the following inclusion criteria:

- Requires single or bilateral lung transplant.
- Male or female, 18 years of age or older.
- Subject or subject’s representative provides a legally effective informed consent.

A recipient may not enroll in the study if they meet any of the following criteria:

- Recipient is HIV positive.
- Recipient has active Hepatitis.
- Investigator believes that the recipient has another infection that excludes them from transplant in the study.
- Recipient is to receive a multi-organ transplant.
- Recipient is on hemodialysis or has chronic severe renal dysfunction (severe renal dysfunction is defined as a glomerular filtration rate of 29 mL/min/1.73m² or less).
- Recipient is to have planned concurrent cardiac procedures.
- Recipient is a re-transplant (re-transplant is defined as a recipient having the removal and transplant of a previously transplanted lung. A recipient with a previously single lung transplant is eligible to enroll in the trial if it is for the other lung and within 6 months of previous transplant).
- Recipient is on Nova Lung, ECMO, or other invasive mechanical ventilation at time of transplant (Continuous Positive Airway Pressure (CPAP) and Bilevel Positive Airway Pressure (BIPAP) are not exclusionary).

2. Follow-up Schedule

All patients were followed through the pre-operative and peri-operative periods, as well as post-transplantation. Post-transplantation, patients were evaluated on Day 7, Day 30, at hospital discharge, and at 12 months post-transplantation. Additionally, through the H120003 PAS, they were evaluated for survival, lung function, adverse events and lung-related complications, such as Bronchiolitis Obleterans Syndrome (BOS), at 2 and 3 years post-transplantation.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 21

{21}

Preoperatively, the donor lungs were screened for their EVLP assessment. The following data were collected:

- Gender;
- Age;
- Donor type;
- Cause of death;
- PaO2/FiO2;
- Cytomegalovirus;
- Smoking history;
- Analysis of x-ray;
- Secretions, if any;
- Endobronchial assessment;
- Results of sputum gram stain, if available;
- Confirmation of lung eligibility, per the approved protocol.

During EVLP, the following parameters were collected:

- Preservation information (e.g., start and stop times of EVLP procedure, start and stop times of cold preservation, volume of preservation solution used, type of preservation used, macroscopic lung evaluation);
- Ex vivo data and x-rays (i.e., x-rays at 1 hour and 3 hours, if necessary, during EVLP);
- Post-EVLP assessment (according to pre-specified criteria);
- Confirmation of lung eligibility post-EVLP (according to pre-specified criteria).

Post-operatively, for the time up to 72 hours post-transplantation, the following data were collected:

- Post-transplant lung status evaluations at 0, 24, and 72 hours;
- PGD scoring;
- Adverse events.

At Day 7, the following data were collected:

- Adverse events;
- Intubation status (including re-intubation and length of time of intubation);
- ICU status (including re-admissions);
- Hospitalization status (including re-admissions);
- Ventilator and ECMO status.

At Day 30, the following data were collected (could be collected in patient visit or via phone):

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 22

{22}

- Survival status (primary cause of death, if applicable);
- Ventilator and ECMO status;
- Adverse events;
- Assess re-admission status.

At hospital discharge, the following data were collected:

- Date of discharge;
- Amount of oxygen the patient was on at discharge;
- Length of stay in hospital (i.e., from date of transplant to date of discharge).

At the 12-month post-transplantation point, the following data were collected:

- Survival status (primary cause of death, if applicable);
- Safety assessments (e.g., graft function);
- Adverse events.

For the H120003 PAS, the following data were collected at 2 and 3 years post-transplantation:

- Survival status (primary cause of death, if applicable);
- Re-transplantations;
- Hospitalizations (including for reasons of infections);
- Evidence of noncompliance;
- Graft status (i.e., functioning or non-functioning, and if failed, reason(s) for failure or rejection);
- Graft function;
- Lung function (FEV1, oxygen requirement at rest, BOS evaluation and evaluation for bronchial strictures);
- Adverse events.

Adverse events and complications were recorded at all visits.

The key timepoints are shown below in the tables summarizing safety and effectiveness.

3. Clinical Endpoints

Regarding safety and effectiveness, the co-primary endpoints were:

- 1-year Survival;
- Rate of Grade 3 primary graft dysfunction (PGD) at 72 hours post-transplant.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 23

{23}

With regard to success/failure criteria, the clinical protocol specified that success would only be met if both the comparisons between survival rates and rates of PGD Grade 3 between the Treatment and Control Arms were successful. The null hypothesis was that the Treatment Arm was non-inferior to the Control Arm for each of the co-primary endpoints. A non-inferiority margin of 12% was specified. The primary endpoints would be evaluated using a 2-sided 95% (adjusted Wald) confidence interval. If the upper confidence limits for each of the difference in rates is no more than 0.12, then the EVLP Group would be considered non-inferior to the Control Group. The protocol specified that "accidental deaths (e.g., automobile accident)" would be excluded from the survival analysis.

The secondary endpoints were:

- Pulmonary function tests (FEV1) at 3, 6, 9, and 12 months;
- PGD scores at 24 and 48 hours post-transplantation;
- Intensive care unit (ICU) length of stay;
- Hospital length of stay;
- Post-transplantation use of extracorporeal membrane oxygenation (ECMO) due to lung function issues;
- Duration of post-transplantation mechanical ventilation;
- Quality of life and functional status at 1 year post-transplantation.

## B. Accountability of PMA Cohort

At the time of database lock, of 226 patients (recipients) enrolled in the PMA study, 226 patients (100%) were available for analysis at the completion of the study (the 12-month post-transplantation visit), and could be determined to be dead or alive at that time point. There were no patients lost to follow-up in this period.

Regarding donor lungs, a total of 332 unique donor lungs were enrolled in the NOVEL/NOVEL Extension study, including 116 standard criteria, control donor lungs. Of the 216 donor lungs entered into the EVLP Group, there were 177 double lungs and 39 single lungs. Of the 177 double lungs, 89 were not transplanted after EVLP and 88 were accepted for transplantation. Of these 88 double lungs accepted for transplantation, 63 were transplanted as double lungs into 63 recipients, 11 were split and one of the split lungs was transplanted as a single lung into 11 recipients (the other lung was deemed unsuitable for transplantation after EVLP), and 7 were transplanted as single lungs into 14 recipients (split lungs where both sides were used). Of the 39 single lungs treated with EVLP, 17 were not transplanted and 22 were transplanted into 22 recipients. In all, a total of 110 recipients were transplanted with EVLP-treated lungs and 106 donor lungs were discarded after EVLP.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 24

{24}

# C. Study Population Demographics and Baseline Parameters

The demographics of the study population are typical for a transplant study performed in the U.S. The donor and recipient characteristics for the subjects included in the study are summarized in Table 11, below.

The UNOS control was used for post-hoc comparisons to the EVLP Arm. This group was comprised of transplant recipients from the same centers as those involved in the NOVEL/NOVEL Extension study, but excluding the following subjects:

EVLP subjects
- Pediatric subjects (recipients $&lt; {18}$ )
Ventilator use at time of transplant
ECMO at time of transplant
History of HIV
- Multi-organ transplant
Re-transplant

Table 11 - Donor Demographics

| Donors | NOVEL EVLP Not Transplanted N=106 | | NOVEL EVLP Transplanted N=110 | | NOVEL Control Transplanted N=116 | | UNOS Control Transplanted N=4898 | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Donor Lung Type | | | | | | | | |
| Bilateral Lungs | 89 | 84.0% | 88 | 80.0% | 85 | 73.3% | Data not available from UNOS | |
| Single Lung | 17 | 16.0% | 22 | 20.0% | 31 | 26.7% | | |
| Donor Gender | | | | | | | | |
| Female | 34 | 32.1% | 30 | 27.3% | 45 | 38.8% | 1917 | 39.1% |
| Male | 72 | 67.9% | 80 | 72.7% | 71 | 61.2% | 2981 | 60.9% |
| Donor Type | | | | | | | | |
| Brain Dead | 66 | 62.3% | 82 | 74.5% | 115 | 99.1% | 4790 | 97.8% |
| Donation After Circ Death | 40 | 37.7% | 28 | 25.5% | 1 | 0.9% | 108 | 2.2% |
| Donor CMV | | | | | | | | |
| Negative | 40 | 37.7% | 54 | 49.1% | 50 | 43.1% | 1899 | 38.8% |
| Positive | 64 | 60.4% | 56 | 50.9% | 66 | 56.9% | 2991 | 61.1% |
| Unknown | 2 | 1.9% | 0 | 0.0% | 0 | 0% | 8 | 0.2% |
| Cause of Death | | | | | | | | |
| Trauma | 43 | 40.6% | 42 | 38.2% | 45 | 38.8% | 1114 | 22.7% |
| CVA | 28 | 26.4% | 25 | 22.7% | 27 | 23.3% | 2097 | 42.8% |
| Hypoxia | 30 | 28.3% | 36 | 32.7% | 37 | 31.9% | 1553 | 31.7% |
| Other | 5 | 4.7% | 7 | 6.4% | 7 | 6.0% | 134 | 2.7% |
| Smoking Status | | | | | | | | |
| Never | 45 | 42.5% | 49 | 44.5% | 59 | 50.9% | Data not available from UNOS | |
| Current | 42 | 39.6% | 43 | 39.1% | 36 | 31/0% | | |
| Former | 10 | 9.4% | 14 | 12.7% | 11 | 9.5% | | |
| Unknown | 9 | 8.5% | 4 | 3.6% | 10 | 8.6% | | |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{25}

Table 12 –NOVEL EVLP Donor Characteristics (before EVLP)

| Donor PaO2 | NOVEL EVLP | | | |
| --- | --- | --- | --- | --- |
| | Not Transplanted
N=106 | | Transplanted
N=110 | |
| PaO2 < 300 | 31 | 29.3% | 40 | 36.4% |
| PaO2 ≥ 300 | 75 | 70.7% | 70 | 63.6% |

Table 13 – Marginal Reasons (for all donor lungs with PaO2 ≥ 300)

| Marginal Reason | Count | Share of Responses | Rate per Case |
| --- | --- | --- | --- |
| > 10 Blood Tranfusions | 18 | 0.0726 | 0.124 |
| Pulmonary Edema | 82 | 0.3306 | 0.566 |
| DCD Donor | 54 | 0.2298 | 0.372 |
| Surgeon’s Opinion | 90 | 0.3669 | 0.621 |

The NOVEL and NOVEL Extension study enrolled previously unacceptable lungs that were offered multiple times to recipient sites and were not accepted, and were thus bound for disposal. For the donor lungs accepted for EVLP, the Organ Placement Organizations (OPOs) made an average of 23 placement attempts (median 9, range 1-199) before a study site accepted the organ for EVLP. This number of placement attempts is higher than the typical number for standard criteria lungs per Harhay et al., 2019, who claims that 70% of organs are placed by match sequence number 10. For most lungs, the OPO continued to attempt placement even after study enrollment due to the fact that the EVLP sites had asked for pump waivers, meaning that if the lung was evaluated not to be transplantable after EVLP, then they would not have to pay for the lung. In those cases where pump waivers were requested, the mean number of additional offers made was 30 (median 23, range 0-317), for a total of 53 refusals (median 32, range 1-383). Without the option of EVLP, the OPO would have stopped lung placement efforts at the last Match Attempt listed and these lungs would not have been used for transplant.

After EVLP, the following donor lung characteristics were noted.

PMA P180014: FDA Summary of Safety and Effectiveness Data

{26}

Table 14 – Post-EVLP Donor Lung Evaluation

| EVLP Run Data | NOVEL EVLP | |
| --- | --- | --- |
| | Not Transplanted
N = 106 | Transplanted
N = 110 |
| Best PO2 | | |
| Best PaO2 (Mean) | 98.4 | 102.0 |
| Best PaO2 (Median) | 92.5 | 91.0 |
| Best Δ PO2 (Mean) | 327.9 | 418.5 |
| Best Δ PO2 (Median) | 340.0 | 417.5 |
| Key Parameters Influencing Decision to Transplant | | |
| Median PAP (mmHg) | 7.5 | 7.8 |
| Median LAP (mmHg) | 4.0 | 4.0 |
| Median PVR (mmHg) | 219.5 | 185.5 |
| Median CStat (dynes) | 85.0 | 103.0 |
| Median PaO2 (Median) | 84.0 | 82.0 |

The major reasons for not transplanting lungs after EVLP are summarized in Table 15.

Table 15 – Post-EVLP Donor Lung Evaluation

| | EVLP Not Transplanted
N=106* | |
| --- | --- | --- |
| Reasons why EVLP-treated lungs were considered unacceptable for transplant | | |
| Edema | 83 | (78%) |
| PaO2 ≤ 350 mmHg | 77 | (73%) |
| Surgeon’s Clinical Decision | 74 | (70%) |
| Compliance | 51 | (48%) |
| Bronchial Finding | 46 | (43%) |
| PVR | 38 | (36%) |
| Airway Pressure | 31 | (29%) |
| Fluid Level in Pump | 26 | (25%) |
| Logistics | 5 | (5%) |

*Only unique lungs are counted for this analysis (i.e., a bilateral set of lungs that was split between two (2) recipients is only counted once). One hundred eight (108) EVLP non-transplants were performed.

The cold ischemia time and total out of body time for the study are summarized in Tables 16 and 17, below. As expected, the cold ischemia time for EVLP donor lungs was higher than that observed in the control arm, and similarly, the total out-of-body time was higher in the EVLP arm, compared to the control.

PMA P180014: FDA Summary of Safety and Effectiveness Data

{27}

Table 16 – Cold Ischemia Time

| | | EVLP
N=110 | Control
N=111* |
| --- | --- | --- | --- |
| Quantiles | | | |
| 100.0% | Max | 905 | 675 |
| 99.5% | | 905 | 675 |
| 97.5% | | 773 | 601 |
| 90.0% | | 688 | 459 |
| 75.0% | Q3 | 595 | 390 |
| 50.0% | Median | 494 | 317 |
| 25.0% | Q1 | 381 | 247 |
| 10.0% | | 287 | 179 |
| 2.5% | | 236 | 116 |
| 0.5% | | 195 | 111 |
| 0.0% | Min | 195 | 111 |
| Summary Statistics | | | |
| Calculable | | 110 | 111 |
| Not Calculable | | 0 | 5 |
| Mean | | 494.4 | 320.4 |
| Standard Deviation | | 146.0 | 108.8 |
| Standard Error (Mean) | | 13.9 | 10.3 |
| Upper 95% Mean | | 522.0 | 342.8 |
| Lower 95% Mean | | 466.9 | 301.9 |
| Interquartile range | | 214 | 143 |

*5 control subjects had incalculable total ischemic time.

Table 17 – Total Out of Body Time

| | | EVLP
N=110 | Control
N=111* |
| --- | --- | --- | --- |
| Quantiles | | | |
| 100.0% | Max | 1125 | 675 |
| 99.5% | | 1125 | 675 |
| 97.5% | | 1087 | 601 |
| 90.0% | | 928 | 459 |
| 75.0% | Q3 | 830 | 390 |
| 50.0% | Median | 732 | 317 |
| 25.0% | Q1 | 627 | 247 |
| 10.0% | | 519 | 179 |
| 2.5% | | 450 | 116 |
| 0.5% | | 375 | 111 |
| 0.0% | Min | 375 | 111 |
| Summary Statistics | | | |
| Calculable | | 110 | 111 |
| Not Calculable | | 0 | 5 |
| Mean | | 735.4 | 322.4 |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{28}

| | EVLP N=110 | Control N=111* |
| --- | --- | --- |
| Standard Deviation | 157.0 | 108.8 |
| Standard Error (Mean) | 15.0 | 10.3 |
| Upper 95% Mean | 765.1 | 342.8 |
| Lower 95% Mean | 705.7 | 301.9 |
| Interquartile range | 203 | 143 |

*5 control subjects had incalculable total ischemic time.

The recipient characteristics for the study are summarized in Table 18.

Table 18 - Recipient Demographics

| Donors | NOVEL EVLP N=110 | | NOVEL Control N=116 | | UNOS Control N=4898 | |
| --- | --- | --- | --- | --- | --- | --- |
| Recipient Gender | | | | | | |
| Female | 41 | 37.3% | 53 | 45.7% | 1947 | 40.3% |
| Male | 69 | 62.7% | 63 | 54.3% | 2924 | 59.7% |
| Recipient CMV | | | | | | |
| Negative | 51 | 46.4% | 56 | 48.3% | 2266 | 46.3% |
| Positive | 59 | 53.6% | 58 | 50.0% | 2552 | 52.1% |
| Not Done | 0 | 0.0% | 2 | 1.7% | 79 | 1.6% |
| Primary Diagnosis | | | | | | |
| Emphysema/COPD/A1 | 48 | 43.6% | 43 | 37.1% | 1442 | 29.4% |
| Fibrosis | 47 | 42.7% | 42 | 36.2% | 2836 | 57.9% |
| Cystic Fibrosis | 7 | 6.4% | 13 | 11.2% | 505 | 10.3% |
| Primary Pulmonary HTN | 0 | 0.0% | 3 | 2.6% | 115 | 2.4% |
| Other | 8 | 7.3% | 15 | 12.9% | Not available | |
| Recipient Race | | | | | | |
| Amerind/Alaska Native | 0 | 0.0% | 1 | 0.9% | Data not available from UNOS | |
| Black/African American | 6 | 5.5% | 4 | 3.4% | | |
| White | 102 | 92.7% | 111 | 95.7% | | |
| Other | 1 | 0.9% | 0 | 0.0% | | |
| Unknown | 1 | 0.9% | 0 | 0.0% | | |
| Recipient Ethnicity | | | | | | |
| Hispanic/Latino | 4 | 3.6% | 3 | 2.6% | Data not available from UNOS | |
| Not Hispanic/Latino | 104 | 94.5% | 112 | 96.6% | | |
| Unknown | 2 | 1.8% | 1 | 0.9% | | |
| Transplanted Lung | | | | | | |
| Bilateral | 63 | 57.3% | 81 | 69.8% | 3491 | 71.3% |
| Single Left | 25 | 22.7% | 21 | 18.1% | 772 | 15.8% |
| Single Right | 22 | 20.0% | 14 | 12.1% | 635... | 13.0% |
| Single/Double | | | | | | |
| Double | 63 | 57.3% | 81 | 69.8% | 3491 | 71.3% |
| Single | 47 | 42.7% | 35 | 30.2% | 1407 | 28.7% |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{29}

D. Safety and Effectiveness Results

1. Safety and Effectiveness Results

Patient Survival

One of the components of the primary endpoint was one-year survival. The survival data (considering all cause mortality) are summarized in Table 19, below. As can be seen, survival for the EVLP Arm was worse than that observed in the Control Arm (86% vs. 94%, respectively), and the patient survival co-primary endpoint is not met according to the definitions pre-specified in the clinical protocol.

The Applicant categorized several deaths, seven (7) in the EVLP Arm and two (2) in the Control Arm, as “accidental” or adjudicated to have “Confounding Risk Factors,” meaning that in the Applicant’s assessment, they were deemed not related to the EVLP treatment, nor to the transplantation procedure. There was a provision in the clinical protocol for adjudicating deaths as “accidental (e.g., automobile accident).” When those “accidental” deaths are excluded from the analysis, the Applicant is able to meet non-inferiority (by the pre-specified delta of 12%), with a one-year survival rate of 93% for the EVLP Arm, compared to 96% for the Control Arm. This analysis introduces uncertainty, since it is difficult to assess bias in the adjudication. There were no “accidental deaths” related to accidents, such as automobile accidents. Table 19 provides the all-cause survival data; therefore, it includes the “accidental deaths.” Similarly, Figure 3 provides the Kaplan-Meier curve for all-cause mortality. Table 20 lists the deaths occurring in the study, including the causes of death.

Table 19 – 1 Year Survival (All Cause Mortality) Rate Difference

| One Year Survival
(All Cause Mortality) | NOVEL EVLP | | NOVEL Control | | UNOS Control | |
| --- | --- | --- | --- | --- | --- | --- |
| | N = 110 | | N = 116 | p-value
(Fisher’s) | N = 4063* | |
| Survived to 1 Year | 95 | (86%) | 109 | (94%) | 0.0718 | 3556 |
| Expired Before 1 Year | 15 | (14%) | 7 | (6%) | | 507 |
| Case Not At 1 Year
(not included in survival %) | 0 of 110 cases | | 0 of 116 cases | | | 835 of 4898 cases |
| | 0 living, 0 expired | | 0 living, 0 expired | | | 793 living, 42 expired |
| * Subjects transplanted less than 1 year from the cutoff date are not included in the analysis. | | | | | | |

An independent three (3) member safety committee (comprised of two (2) lung transplant surgeons and one (1) lung transplant pulmonologist) performed a quarterly review of a listing of safety data for the EVLP and Control Arms to assess if the events that occurred outside of the expected events in this population. This included quarterly review and adjudication of all Major Lung Events (MLEs) and Deaths as per the study protocol safety charter. The committee reviewed causality, cause of death, MLE type, and

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 30

{30}

provided clinical justification for the deaths removed from the specific cause survival analysis. If an event was considered an Unanticipated Adverse Device Effect (UADE), the safety committee and the Independent Safety Monitor (ISM) adjudicated and assessed unreasonable risk.

In addition to the uncertainly over the "accidental" deaths, the Applicant has also expressed concerns regarding the NOVEL and NOVEL Extension's Control Group. They state that the Control Group was subject to selection bias, since some investigators in the study failed to enroll Control patients concurrently with their EVLP patients, as evidenced in the monitoring visits performed by the Applicant to their investigational sites. The Applicant states that the Control Group performance (e.g., 94% and 96% 12-month all cause mortality and specific mortality survival, respectively) is excessively high, and thus, the study's Control Group is not a fair comparator for the EVLP data.

To bring some perspective to the EVLP data, the Applicant has provided a comparison between the EVLP results and data from the United Network Organ Sharing (UNOS) Scientific Registry of Transplant Recipients (SRTR). The UNOS data provided are for transplants that took place during the time of the NOVEL and NOVEL Extension studies and were collected at the same investigational sites. They exclude the NOVEL patients (EVLP and Control), but are inclusive of all other transplants performed that fit the NOVEL study entry criteria. As can be seen, the patient survival data are comparable between the EVLP Arm and UNOS data (all-cause mortality survival rates of 86% for the EVLP Group and 88% for the UNOS Control Group). The Applicant states the comparison between the EVLP Group and the UNOS data is more indicative of real-world evidence.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 31

{31}

Figure 3 - Kaplan-Meir Curve - NOVEL EVLP, NOVEL Control and UNOS - All Cause Mortality

| Group | Number failed | Number Censored | p-value ChiSq |
| --- | --- | --- | --- |
| EVLP | 15 | 95 | Log-Rank: <0.001 Wilcoxon: <0.001 UNOS data is lower |
| NOVEL Control | 7 | 109 | |
| UNOS | 1341 | 3557 | |
| Combined | 1363 | 3761 | |

Table 20 - Patient Deaths - First Twelve Months Post-Transplantation

| Site | Subject | Study Arm | Diagnosis Requiring Tansplant | Date of Transplan | Date of Death | Primary Cause of Death |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | 123 | Control | Fibrosis | 4/3/14 | 6/4/14 | Antibody mediated rejection |
| 3 | 301 | EVLP | Fibrosis | 10/5/11 | 10/15/11 | Reperfusion injury due to cytokines |
| 3 | 302 | Control | Fibrosis associated with short telomere | 1/20/12 | 9/26/12 | Renal failure |
| 4 | 409 | EVLP | Fibrosis | 4/22/12 | 11/10/12 | Respiratory failure |
| 4 | 412 | EVLP | Fibrosis | 10/4/12 | 2/22/13 | Acute rejection |
| 4 | 427 | Control | Emphysema, COPD, Alpha 1 Antitrypsin Deficiency | 7/15/17 | 12/29/17 | Acute on chronic hypercarbic respiratory failure |
| 5 | 504 | Control | Fibrosis | 3/3/12 | 8/9/12 | Airway stenosis and respiratory failure |
| 5 | 513 | EVLP | Fibrosis | 7/18/12 | 2/1/13 | Complications from aortic injury |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{32}

| Site | Subject | Study Arm | Diagnosis Requiring Transplant | Date of Transplant | Date of Death | Primary Cause of Death |
| --- | --- | --- | --- | --- | --- | --- |
| 5 | 522 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 9/12/12 | 6/11/13 | BOS, B-cell lymphoma/leukemia which caused a discontinuation of immunosuppression |
| 5 | 572 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 5/21/17 | 12/22/17 | Bacterial septicemia |
| 6 | 609 | EVLP | Fibrosis | 3/2/13 | 10/17/13 | Massive hemoptysis secondary to a bronchovascular fistula that occurred following stent placement due to bronchial stenosis |
| 6 | 620 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 6/17/13 | 12/21/13 | Renal failure |
| 6 | 625 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 11/3/13 | 8/5/14 | Lung cancer (RLL squamous cell carcinoma) |
| 6 | 630 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 6/16/15 | 7/5/15 | Liver failure |
| 6 | 640 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 12/20/16 | 9/8/17 | Lymphoma |
| 7 | 703 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 7/16/13 | 2/22/14 | Graft vs. Host Disease |
| 8 | 809 | Control | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 3/6/15 | 12/25/15 | Septic shock caused by aspiration pneumonia |
| 9 | 905 | EVLP | Fibrosis | 10/12/16 | 11/28/16 | Sepsis due to colon perforation with diverticulitis |
| 11 | 1105 | Control | Scleroderma | 5/13/15 | 4/26/16 | Intracranial hemorrhage |
| 11 | 1111 | EVLP | Emphysema/COPD/ Alpha 1 Antitrypsin Deficiency | 3/9/17 | 7/17/17 | Septic shock |
| 16 | 1603 | EVLP | Fibrosis | 3/22/17 | 4/30/17 | Cardiopulmonary arrest |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{33}

| Site | Subject | Study Arm | Diagnosis Requiring Transplant | Date of Transplant | Date of Death | Primary Cause of Death |
| --- | --- | --- | --- | --- | --- | --- |
| 17 | 1705 | Control | Fibrosis | 9/25/16 | 8/15/17 | Respiratory failure, sepsis |

The deaths adjudicated by the Applicant and their independent Safety Committee as being “accidental” or unrelated to the device are:

- Unrelated control deaths: 302, 1105
- Unrelated EVLP deaths: 513, 620, 625, 640, 703, 905, 1111

As discussed earlier, when looking at all-cause mortality, the NOVEL study does not meet its pre-specified primary endpoint, which includes 12-month survival. Instead of falling within the pre-specified 12% non-inferiority margin, the observed value for survival for the EVLP Arm is 15% different (worse) than the Control Group (as indicated by the upper bound of the two-sided 95% confidence interval). FDA concludes that the all-cause mortality data should be used when describing the survival co-primary endpoint, since that was the analysis pre-specified in the clinical protocol. As such, EVLP subjects fared worse than the Control Group, although the concerns raised by the Applicant about the Control Group should be considered. The FDA also believes the comparison of the NOVEL EVLP data to data collected from UNOS is valuable and can provide an indication of where EVLP-treated lungs might fit in the lung transplant field.

## PGD Grade

The second component of the primary endpoint was the incidence of PGD Grade 3 at 72 hours post-transplantation. Table 21 shows the PGD data for 24, 48 and 72 hours post-transplantation before adjudication. Considering these unadjudicated data, the PGD Grade 3 at 72 hours post-transplantation were 14% for the EVLP Group vs. 7% for the Control Group. This comparison does not meet the success criteria pre-specified in the clinical protocol.

The PGD data were adjudicated by the Applicant’s Clinical Events Committee. During adjudication, there were nine (9) patients (all in the EVLP arm) who were deemed to have received prophylactic Extracorporeal Membrane Oxygenation (ECMO). These were patients who were placed on ECMO prior to the transplantation surgery and were then kept on ECMO post-transplantation until they recovered sufficient function. According to the protocol, any patient on ECMO when assessed for PGD would be automatically graded Grade 3, per the International Society for Heart and Lung Transplantation (ISHLT) guidelines; however, according to the study’s Statistical Analysis Plan, these “prophylactic” ECMO patients would be excluded from the primary analyses. As shown below, the incidence of PGD Grade 3 at 72 hours post-transplantation was higher in the EVLP Arm, compared to Control, and when all ECMO patients are included in the

PMA P180014: FDA Summary of Safety and Effectiveness Data

{34}

analyses, pre-specified 12% non-inferiority is not met. When the prophylactic ECMO patients are excluded, however, the PGD Grade 3 rates are more comparable. Retrospectively, it is difficult to determine whether patients on ECMO therapy are being truly treated prophylactically, or if they are being given ECMO therapeutically. As such, the Summary of Safety and Effectiveness Data (SSED) and labeling include all Grade 3 subjects, even if Grade 3 is due to ECMO. After independent adjudication, but still including all ECMO subjects, the PGD Grade 3 rates were 16% and 9% in the control and EVLP Groups, respectively.

The PGD Grade 3 data at 72-hours post-transplantation does have a favorable comparison to the UNOS control data provided by the sponsor (31.3% Grade 3 at 72 hours post-transplantation, N=1200), although it should be noted that PGD grade data are not always available in the SRTR database, and the SRTR registry is not a reliable comparator for PGD. Looking at other comparators, the Applicant has cited the Lung Transplant Outcomes Group (LTOG), which was a U.S. National Institutes of Health (NIH) sponsored, multicenter, prospective cohort study designed to evaluate risk factors for, and rates of, PGD. This study included 10 centers and its primary outcome was PGD at 48 or 72 hours post-transplantation. According to the LTOG data, the rate of PGD Grade 3 at 72 hours post-transplantation was 16.8% (Diamond et al., 2013), which compares favorably with the NOVEL and NOVEL Extension EVLP 72-hour PGD data, which was 16%. Additionally, per Christie et al., 2005, the national incidence of 72-hour Grade 3 PGD is between 10% and 30%.

In order to monitor safety in real-time and continually assess the safety of the device, the Safety Committee was un-blinded to Treatment Arms and could not be used to adjudicate PGD at 72 hours, as this could bias the adjudication. Accordingly, all of the 72-hour PGD scores were adjudicated by 2 blinded independent transplant pulmonologists per the study protocol. The adjudicators' responsibility was to perform PGD adjudication to determine PGD score based on the 72 hour raw, blinded, de-identified chest x-ray images and a clinical database extract of Arterial Blood Gases (ABGs) using the ISHLT Determination. If there was non-consensus between the Investigator and Primary Adjudicator, all reports and images were reviewed and assessed by a secondary adjudicator. The majority PGD score determined the final score. When one adjudicator provided a score of 3, which was not in consensus with the other adjudicator, a second adjudicator review took place. If there was a non-consensus decision between the investigator, primary adjudicator, and secondary adjudicator, a second adjudicator review took place and a consensus was made between the Primary and Secondary adjudicator.

PMA P180014: FDA Summary of Safety and Effectiveness Data
Page 35

{35}

Table 21 – Primary Graft Dysfunction – Unadjudicated and Including Prophylactic ECMO – 24, 48 and 72 hours Post-Transplantation

| @ 24 Hours | EVLP
N = 110 | | Control
N = 116 | | p-value
(χ²) |
| --- | --- | --- | --- | --- | --- |
| Grade 0 | 24 | (22%) | 29 | (25%) | 0.0294 |
| Grade 1 | 34 | (31%) | 46 | (40%) | |
| Grade 2 | 24 | (22%) | 29 | (25%) | |
| Grade 3 | 28 | (25%) | 12 | (10%) | |
| @ 48 Hours | EVLP
N = 110 | | Control
N = 116 | | p-value
(χ²) |
| Grade 0 | 30 | (27%) | 29 | (25%) | 0.4847 |
| Grade 1 | 47 | (43%) | 55 | (47%) | |
| Grade 2 | 16 | (15%) | 21 | (18%) | |
| Grade 3 | 17 | (15%) | 11 | (9%) | |

@ 72 Hours

| All Subjects | EVLP
N = 110 | | Control
N = 116 | | p-value
(χ²) |
| --- | --- | --- | --- | --- | --- |
| Grade 0 | 37 | (34%) | 37 | (32%) | 0.3387 |
| Grade 1 | 42 | (38%) | 53 | (46%) | |
| Grade 2 | 16 | (15%) | 18 | (16%) | |
| Grade 3 | 15 | (14%) | 8 | (7%) | |

Other Safety and Effectiveness Data

Additional safety data are summarized in the tables below, including Intensive Care Unit length of stay, hospital length of stay, duration of mechanical ventilation, pulmonary function data, delayed extubations, re-intubations and tracheostomies, and ECMO use. These parameters show comparable performance between the EVLP and Control Groups.

Table 22 – Intensive Care Unit (ICU) Length of Stay (LOS, days)

| ICU Length of Stay | NOVEL EVLP
N = 110 | NOVEL Control
N = 116 |
| --- | --- | --- |
| | | |
| Mean LOS | 9.9 | 9.8 |
| Standard Deviation | 14.4 | 18.7 |
| 25th Percentile | 3 | 2 |
| Median LOS | 5 | 4.5 |
| 75th Percentile | 9 | 9 |
| Interquartile Range | 6 | 7 |
| Expired Prior to ICU Discharge | 4 | 0 |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{36}

Table 23 – Hospital Length of Stay (LOS, days)

| Hospital Length of Stay | NOVEL EVLP
N = 110 | NOVEL Control
N = 116 |
| --- | --- | --- |
| | | |
| Mean LOS | 23.9 | 28.5 |
| Standard Deviation | 24.4 | 41.7 |
| 25th Percentile | 11 | 10.25 |
| Median LOS | 16 | 14.5 |
| 75th Percentile | 24.75 | 24.25 |
| Interquartile Range | 13.75 | 14 |
| Expired Prior to Hospital Discharge | 5 | 0 |

Table 24 – Duration of Mechanical Ventilation (days)

| Intubation | NOVEL EVLP
N = 110 | NOVEL Control
N = 116 |
| --- | --- | --- |
| | | |
| Mean LOS | 7.0 | 5.7 |
| Standard Deviation | 24.7 | 21.8 |
| 25th Percentile | 1 | 1 |
| Median LOS | 1 | 1 |
| 75th Percentile | 3.75 | 2 |
| Interquartile Range | 2.75 | 1 |
| | | |
| Expired Prior to Extubation | 2 | 0 |

Table 25 – Pulmonary Function Test – FEV1% (Predicted) – at 3, 6, 9, 12 Months

| | 3 Months | | 6 Months | | 9 Months | | 12 Months | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| FEV1 | EVLP
N = 110 | Control
N = 116 | EVLP
N = 110 | Control
N = 116 | EVLP
N = 110 | Control
N = 116 | EVLP
N = 110 | Control
N = 116 |
| Mean | 69 | 73 | 71 | 74 | 72 | 74 | 72 | 76 |
| Median | 69 | 72 | 70 | 71 | 72 | 72 | 72 | 75 |
| Range | (19-111) | (22-125) | (22-123.8) | (26-136) | (28-120) | (26-150) | (23-115) | (21-144) |
| Not Done* | 11 | 9 | 14 | 9 | 21 | 9 | 20 | 11 |
| * A +/- 30 day window was allowed on all PFT evaluations. A PFT evaluation may not have been performed if the subject expired close to the 1/2/3 year timepoint, or if the subject was trached or hospitalized at the scheduled time of evaluation. | | | | | | | | |

PMA P180014: FDA Summary of Safety and Effectiveness Data

{37}

Table 26 – Delayed Extubations, Re-Intubations and Tracheostomies

| | EVLP
N = 110 | | Control
N = 116 | | p-value
Fisher's |
| --- | --- | --- | --- | --- | --- |
| Delayed Extubation | | | | | |
| Extubated Within 96 Hours of TX | 91 | (83%) | 101 | (87%) | 0.4570 |
| Not Extubated Within 96 Hours of TX | 19 | (17%) | 15 | (13%) | |
| Re-Intubations | | | | | |
| Not Re-Intubated After Initial Extubation | 88 | (80%) | 91 | (78%) | 0.8700 |
| Re-Intubated Within 1 Year of TX | 22 | (20%) | 25 | (22%) | |
| Tracheostomies (Due to Respiratory Failure) | | | | | |
| Not Trached Within 1 Year of TX | 94 | (85%) | 99 | (85%) | 1.0000 |
| Trached Within 1 Year of TX | 16 | (15%) | 17 | (15%) | |

Table 27 – ECMO Use and PFT Data

| | Count | | Average ECMO Days | FEV1 @3M | FEV1 @6M | FEV1 @9M | FEV1 @12M |
| --- | --- | --- | --- | --- | --- | --- | --- |
| EVLP | | | | | | | |
| All Subjects | 110 | (NA) | NA | 69.0 | 70.6 | 72.2 | 72.0 |
| No ECMO At Any Point | 87 | (79%) | NA | 69.0 | 69.9 | 70.9 | 70.7 |
| Intraoperative ECMO | 23 | (21%) | 8.5 | 69.2 | 73.5 | 77.8 | 77.8 |
| Prophylactic Use (% of intraop) | 7 | (30%) | 2.0 | 71.0 | 69.9 | 75.9 | 75.2 |
| Non-Prophylactic (% of intraop) | 16 | (70%) | 11.4 | 68.3 | 76.0 | 79.1 | 79.4 |
| Postoperative ECMO | 1 | (1%) | 4.0 | 65.0 | 72.0 | 72.0 | 72.0 |
| Control | | | | | | | |
| All Subjects | 116 | (NA) | NA | 72.9 | 74.2 | 74.0 | 75.7 |
| No ECMO At Any Point | 101 | (87%) | NA | 73.6 | 75.8 | 76.0 | 78.0 |
| Intraoperative ECMO | 15 | (13%) | 0.3 | 67.3 | 63.9 | 61.8 | 61.9 |
| Prophylactic Use (% of intraop) | 5 | (33%) | 0.4 | 69.3 | 69.2 | 68.8 | 68.2 |
| Non-Prophylactic (% of intraop) | 10 | (67%) | 0.2 | 66.3 | 61.0 | 58.3 | 58.8 |
| Postoperative ECMO | 7 | (6%) | 14.3 | 41.0 | 44.4 | 43.0 | 39.7…