A multicentre randomised controlled trial to compare the efficacy of ex- vivo normothermic machine perfusion with static cold storage in human liver transplantation

Investigators: Prof Constantin Coussios (Oxford, UK) Mr Simon Knight (Oxford, UK) Mr David Nasralla (Oxford, UK) Professor Andreas Paul (Essen, Germany) Professor Jaques Pirenne (Leuven, Belgium) Professor Juan Carlos Garcia-Valdecasas (Barcelona, Spain) Professor Mr Wayel Jassem (King’s College, London, UK) Mr Hynek Mergental (Birmingham, UK) Mr Andrew Butler (Cambridge, UK) Mr Charles Imber (Royal Free, London, UK) Mr Zeeshan Akhtar (Oxford, UK) Sponsor: University of Oxford


Secondary
To compare graft and patient survival between NMP and SCS livers.
To compare evidence of post-reperfusion syndrome between NMP and SCS livers on implantation.
Assess mean arterial pressure (MAP) preand post-reperfusion in the context of vasopressor use.
To compare biochemical liver function between NMP and SCS livers.
Bilirubin, GGT, AST and INR at days 1-7, day 30 and months 6, 12 and 24 posttransplant. Lactate at days 1-7 whilst the patient is still on ITU/HDU To compare evidence of ischaemic reperfusion injury between NMP and SCS livers.
Post-reperfusion biopsies will be compared to baseline pre-reperfusion biopsies and graded according to standard histological criteria.
To compare evidence of ischaemic cholangiopathy between NMP and SCS livers.
Evidence of biliary stricturing on magnetic resonance cholangiography (MRCP) at 6 months post-transplant.
To assess the ability of perfusion parameters and biomarkers in perfusion fluids to predict clinical outcomes following transplantation.
Perfusion flows and pressures, perfusate blood gas parameters, bile production, perfusate biochemical markers and proteomics.
To assess the feasibility and safety of NMP as a method of organ storage and transportation.
Adverse events, transplantation and organ discard rates.
To assess the health economic implications of normothermic liver perfusion.
Quality of life measures (EQ-5D-5L) at 30 days and 6 months. Logistical and healthcare costs, and resource use,

INVESTIGATOR SIGNATURE PAGE
I agree to conduct this clinical study in accordance with the design and specific provisions of this protocol and will only make changes in the protocol after notifying the sponsor.
I understand that I may terminate or suspend enrolment of the study at any time if it becomes necessary to protect the best interests of the study subjects as advised by the DMC. This study may be terminated by the University of Oxford, with or without cause.
I agree to personally conduct or supervise this investigation and to ensure that all associates, colleagues, and employees assisting in the conduct of this study are informed about their obligations in meeting these commitments.
I will conduct the study in accordance with Good Clinical Practice, the Declaration of Helsinki, and the moral, ethical and scientific principles that justify medical research. The study will be conducted in accordance with all relevant laws and regulations relating to clinical studies and the protection of patients.
I will ensure that the requirements relating to Research Ethics Committee (REC) and regulatory authority review and approval are met. I will provide the University of Oxford with any material that is provided to the REC or regulatory authority for ethical approval.
I agree to maintain adequate and accurate records and to make those records available for audit and inspection in accordance with relevant regulatory requirements.
I agree to promptly report to the DMC, REC, regulatory authorities and sponsor any changes in the research activity and all unanticipated problems involving risks to human subjects or others. Additionally, I will not make any changes in the research without REC and regulatory approval, except where necessary to ensure the safety of study participants.

TITLE
"A multicentre randomised controlled trial to compare the efficacy of ex-vivo normothermic machine perfusion with static cold storage in human liver transplantation."

TRIAL REGISTRATION
This trial protocol will be registered with the ISRCTN prior to study commencement.

PREVIOUS VERSIONS
Details of previous versions and amendments to this protocol are detailed in appendix A6.

STUDY FUNDING
European Union Seventh Framework Programme (FP7): the COPE study (Work Package WP2).
The funding agency will not take part in nor has the ultimate authority over the study design; collection, management, analysis or interpretation of data; writing of the report; and the decision to submit the report for publication.

SPONSOR
The trial is sponsored by Oxford University.

Chief and central investigators
The chief and central investigators will be responsible for:  Design and conduct of the study  Preparation of protocol and revisions  Preparation of trial-specific standard operating procedures (SOP)  Advising the REC, CA and DMC chairman of serious adverse events (SAE)  Organisation and management of device training  Organisation and management of logistics  Publication of study reports The chief investigator has overall responsibility for these activities.

Local lead investigators
One investigator per contributing centre will be nominated as a local co-ordinator. Responsibilities will include:  Obtaining local research ethics committee and research governance approval (aided by the central investigators)  Identification and recruitment of patients to the study  Conducting clinical procedures in accordance with the protocol and standard operating procedures  Data collection and completion of electronic CRFs  Follow-up of study participants Local investigators will also be members of the trial management committee.

Clinical Trials Unit
The Surgical Intervention Trials Unit (SITU) is the clinical trials unit for this trial and will be responsible for:  Participant randomisation  Database design  Management of data collection  Statistical analysis of trial data  Providing data for regular DMC meetings  Monitoring the trial  Site initiation and close-out

Database design
The SITU will develop and maintain the trial database and electronic clinical reporting forms. The database will be designed to provide the IT system for the randomisation procedure, and will have reporting functionality in order to capture data entry and verification, which the trial data manager will utilise.

Trial Co-ordinator
The trial co-ordinator will be based in SITU and ensure that regulatory standards are maintained and the trial is conducted according to the principles of GCP. The Trial Co-ordinator will also be responsible for:  Organisation of trial management committee meetings  Ensure local investigator's compliance with the trial protocol  Clinical monitoring  Regular reporting to the sponsor on the progress of the clinical investigation.

Trial statistician
The trial statistician will be responsible for the development of the statistical analysis plan and the subsequent analysis of trial data.

TRIAL MANAGEMENT COMMITTEE
The trial management committee (TMC) will be responsible for:  Agreement of the final protocol  Reviewing progress of the study and, if necessary, agreeing to changes to the study protocol and/or standard operating procedures to facilitate the success of the study  Reviewing new studies that may be of relevance to the current protocol  Receiving and reviewing details of any concerns raised by the DMC  Reporting periodically to the COPE Management Board on the progress of the trial All lead investigators will be trial management committee members, with one representative from each of the participating centres. These individuals are:

DATA MONITORING COMMITTEE
The data monitoring committee (DMC) is responsible for:  Agreeing a charter for the conduct of the DMC  Reviewing data from the study according to the schedule set out in the protocol  Reviewing serious adverse events (device related or not) and any device deficiencies As a result of the reviews the DMC may make recommendations to the TMC, including premature termination of the trial, should they feel it is indicated.
Members of the data monitoring committee are:

CONSORTIUM FOR ORGAN PRESERVATION IN EUROPE (COPE)
The Consortium for Organ Preservation in Europe (COPE) is the official organ task force of the European Society for Organ Transplantation (ESOT) and is funded by a European Commission FP7 Award. The Consortium brings together academic institutions, clinical and scientific experts and SMEs from across Europe to work together on advancing organ preservation techniques.
There are currently a range of clinical and translational studies, known as Work Packages (WP), being funded by COPE with the aim of advancing and developing organ preservation technology.

COPE MANAGEMENT BOARD
The COPE Management Board will consist of the COPE co-ordinator, administrators, Work Package leaders and representatives from the SMEs involved in the various COPE trials, including OrganOx. The TMC will report to the Management Board periodically with updates on trial progress. If the COPE Management Board have concerns about the progress of the trial, the TMC will be informed of this with information about specific issues that need to be addressed. If required, the COPE Management Board can receive advice from the External Advisory Board, as described below.
The COPE co-ordinator and administrators are also required to periodically provide updates to the EU Scientific Officer on the progress of all Work Packages within the COPE consortium.

EXTERNAL ADVISORY BOARD
The external advisory board (EAB) will act as a source of independent, impartial advice to the COPE Management Board should they request it. This may be in the context of concerns or conflicts arising from recommendations by the DMC, or regarding unforeseen circumstances that arise during the course of the trial.
The EAB will consist of independent clinicians with expertise in transplantation and an ethicist. These members are still to be confirmed.
An overview of the governance framework for COPE, as described above, is illustrated in Appendix 5.

LIVER TRANSPLANTATION -A SUCCESSFUL THERAPY
Liver transplantation is the only effective treatment for many patients with liver disease. For patients with liver failure, techniques for supporting liver function provide only limited and temporary benefit as a bridge to transplantation or liver regeneration in the case of acute liver failure. In contrast to kidney dialysis support in renal failure, there is no artificial means to support a patient with liver failure for an extended period. Since the first clinical case in 1963, transplantation of the liver has developed to an established and standardised procedure as cure for acute and chronic liver disease. The results of liver transplantation have greatly improved and patient survival rates of over 90% at one year and 70% at five years are routinely achieved for elective liver transplantation. As a result, liver transplantation has become the mainstay of treatment for an increasing spectrum of patients with chronic liver disease, metabolic liver disease, acute liver failure and some liver cancers.

LIVER FAILURE EPIDEMIOLOGY
In European countries cirrhosis is the most common indication for liver transplantation (68%), other indications include cancer (14%), and acute hepatic failure (8%) [www.eltr.org]. The main causes for cirrhosis in Europe are virus related cirrhosis and alcoholic liver disease [www.eltr.org]. In the UK it is predicted that HCV-related cirrhosis and deaths from hepatocellular carcinoma (HCC) will increase substantially during the next decade. Similarly the incidence of liver disease related to both obesity and alcohol consumption continues to increase despite health campaigns. It is very likely that the demand for donor livers suitable for transplantation will continue to rise exacerbating the existing organ shortage.

THE DONOR SHORTAGE
Over the last two decades, liver transplantation has become a victim of its own success: many more patients are referred for transplantation, but the number of suitable grafts from deceased organ donors has increased more slowly. The donor organ shortage constitutes a serious risk for patients with liver failure. It is the principal cause of increasing waiting lists and the death of patients on the waiting list worldwide. 739 liver transplants were carried out from deceased donors in the UK during 2011-2012. However, 1105 patients were registered for liver transplantation in the same period. This shortfall is typical of liver transplantation services around the world. In many countries, a patient is now more likely to die on the waiting list for a transplant than in the 12 months after the operation.
Great efforts have been made in recent years to increase the referral of organ donors: this has resulted in a 50% increase in the number of donors referred in the UK, but this is largely due to an increase in the number of donors declared dead by cardiovascular criteria ('donation after circulatory death', DCD) and other 'extended criteria' organ donors (older age, steatosis etc.). There has been a much smaller increase in the number of standard ('ideal') organ donors, declared dead by neurological criteria ('donation after brain death', DBD). 132 patients underwent liver transplantation with DCD donor organs in 2012-13 (from a total of 507 DCD donors).

APPROACHES TO THE DONOR ORGAN SHORTAGE
Organ donor rates in different countries range widely: using 2011 data, the highest rate occurred in Spain with 35.3 donors per million population per year (DPMP), compared to the UK (17 DPMP, Germany -14.7 DPMP, Portugal -28.1 DPMP, France -25 DPMP, Belgium 30.1 DPMP) (Council of Europe data). It is generally accepted that these discrepancies are due not only to cultural distinctions but also due to political and legal differences. There is debate as to whether the situation is simpler in those countries that practice "presumed consent" (opting out) [1]. Most countries, however, practice a system of "opting in," whereby consent must be sought from the family of the donor.
Due to the critical shortage of donor organs, clinicians are continually searching for ways to overcome the discrepancy between demand and availability of donor livers for transplantation. Additional sources of organs include those from living donors, organs from DCD (donation after circulatory death) donors (previously known as 'non-heart-beating donors') and other less-than-optimal ('higher risk') deceased donors.

Living donation
Living donation is one potential means to increase the number of liver transplants, using surgical techniques developed for ´liver splitting´ (a technique for using a single liver for transplantation into two recipients). The major limitations of this technique include, first, the fact that most patients do not have a willing and suitable living donor, and, second, real concerns about the risks to the healthy donor. The reported risk of donor death is estimated at 0.2% but the risk of serious complications is much higher [2]. For these and other reasons living donor transplantation has had limited impact on the shortage of donor livers for transplantation in most countries.

Higher risk donor organs
Much emphasis is now placed on optimising the condition of those organs that are available, to enable more higher risk organs to be transplanted safely. The use of a higher risk organ does constitute a greater risk to the recipient, with a higher probability that the organ will never function and require immediate replacement (primary non-function, PNF), that it will function poorly and place the patient at risk of other complications (early allograft dysfunction; EAD) or that it will lead to later complications including multiple stricturing of the biliary tree (ischaemic cholangiopathy; IC).
The serious effects of the organ shortage, with many patients dying on the waiting list, has led to increased interest in using donor livers which were formerly thought unsuitable for transplantation. The use of these ´extended criteria donors´ (ECD; also called 'marginal' or 'high risk') donor livers for liver transplantation is now seen as essential if liver transplant units are to address the demand. Several donor parameters have been identified as relative risk factors for poor outcome, including age; steatosis; DCD donation; split livers; prolonged cold ischaemia time (>12 hours). These were all developed using North American data and formulated into an algorithm known as the 'Donor Risk Index' (DRI) [3], and later validated using European data [4].

Donation after circulatory death (DCD)
For many years (following the establishment of brain death criteria) in most countries, deceased donor organs were almost exclusively sourced from donors declared dead by neurological criteria (donation after brain death, DBD). In such cases the donor remains on ventilatory support and cardiopulmonary function is maintained until the donor has been transferred to the operating room, preliminary dissection has been performed, cannulae placed and the organs are ready for cooling. This enables organs to be removed with minimal interruption to oxygenation before cooling and preservation. In contrast, in the case of DCD donors, death is certified after cardiac arrest using cardiovascular parameters; this causes an inevitable period of oxygen deprivation between cardiac arrest and cold preservation of the organs. DCD donors have been classified as 'uncontrolled', in which death is not predicted (typically after failed resuscitation in an emergency room) and 'controlled' in which death is anticipated (typically under circumstances where life-support is withdrawn from a patient in whom continued treatment has been deemed to be futile).
In the latter situation, because death is anticipated and the transplant team can be mobilised in advance, the period of 'warm ischaemia' sustained by the donor organs is usually much shorter; livers from this source have been transplanted in substantial numbers in recent years, although very selectively, but with higher rates of primary non-function and other post-operative complications than DBD livers, with worse long-term graft survival [5][6][7]. This increase in complications results in an increased cost incurred with DCD transplantation [8].
In contrast, livers from uncontrolled DCD are not generally used for transplantation because of a high rate of primary non-function in the experience of those centres that have attempted this [9]. It has been estimated, however, that effective use of these potential donors would add greatly to the number of organ donors [10]. This contrasts with the situation in renal transplantation in which organs from uncontrolled DCD have been used by some centres with outcomes that compare well with those from controlled DCD donors [11]. Following renal transplantation, a period of 'delayed graft function' is acceptable because it is possible to support the patient with dialysis whilst the transplanted organ recovers from the ischaemic injury; an equivalent delay in initial function of a transplanted liver is fatal without urgent re-transplantation.

Viability assessment
The safety and utility of DCD liver transplantation would be greatly improved by a reliable test to quantify the ischaemic/preservation injury and assess viability in order to predict the outcome after transplantation. Even after careful pre-retrieval donor selection, up to 45% of retrieved livers from DCD donors are discarded due to doubts about viability [12]. In the UK 2011/12 only 689 of 782 (88%) of all deceased donor livers retrieved were actually transplanted, falling to 132 of 178 (74%) for livers from DCD donors (NHSBT data). An effective means of pre-transplant viability assessment would not only allow greater use of higher risk donors but also minimise the risk of primary non-function by identifying and excluding non-viable organs before subjecting a patient to the risk of surgery.

Ex vivo reconditioning
A further strategy in the quest to use higher risk donor organs successfully is that of 'reconditioning' after retrieval -using techniques to reverse the injury sustained by the organ before and during the process of retrieval and treating the organ in such a way as to minimise the immediate damage that occurs after transplantation (ischaemia-reperfusion injury). Treatment of the organ during preservation has major logistic and ethical advantages over any attempt to achieve the same effects by treating the donor (therapeutic interventions before declaration of death that are not of potential benefit to the donor are not currently permitted). Many cytoprotective strategies have been tested in experimental models of transplantation and several have been shown to have therapeutic potential, including various antioxidants, inflammation inhibitors, vasodilating agents, inhibitors of chemotaxis or neutrophil infiltration.
At the moment the flow of oxygenated blood ceases, the supply of oxygen, cofactors, and nutrients stops along with the means of disposal of metabolic waste products. Anaerobic metabolism continues (at a temperature-dependent rate), leading to depletion of energy stores, mainly adenosine tri-phosphate (ATP), with a concomitant build-up of an acidotic milieu. ATP is required for energy-dependent cellular functions, including the integrity of sodium/potassium pumps that maintain electrolyte balance across cell membranes and ATP depletion leads to loss of trans-cellular electrolyte gradients, cell swelling, influx of free calcium, and the subsequent activation of phospholipases. The breakdown of ATP during ischaemia also generates substrates for the production of reactive oxygen intermediates on reperfusion and initiates the cascade of ischaemic injury. Prevention of ATP depletion is therefore an important target of innovative preservation methods. It has been shown that providing an oxygen supply to the organ can prevent ATP depletion and preserve viability following cardiac arrest in a porcine liver transplant model [13].

COLD STORAGE
Organs retrieved for transplantation undergo injury at several consecutive stages: 1) warm ischaemia prior to preservation, 2) cold preservation injury, 3) ischaemic rewarming during surgical implantation and 4) reperfusion injury. These consecutive events lead to a cumulative cellular injury that may not be compatible with recovery after transplantation.
Standard clinical practice involves flushing and cooling the liver in situ with preservation solution; University of Wisconsin [UW] solution is used most commonly although Histidine-Tryptophane-Ketoglutarate [HTK] solution is also widely used. Typically several litres of cold preservation fluid are used both in situ and after removing the organ from the donor and before packing for transport and storage. Additional cooling may be provided by topical frozen saline slush both in situ and ex situ. After retrieval, the organ is placed in sterile plastic bags for transportation and stored in an ice-box in preservation solution until transplantation. Although the available preservation solutions differ in chemical composition, the function is essentially the same: to prevent cellular swelling and death caused by fluid shifts as the membrane ion-exchange pumps cease operating in the cold environment. Although cold preservation slows metabolism by 1.5-to 2-fold for every 10°C drop in temperature, considerable metabolic activity still occurs at 1°C. This leads to accumulation of metabolic products which act as substrates for metabolism that takes place when the organ is re-perfused with oxygenated blood -the basis of the ischaemia-reperfusion phenomenon [14] In organs retrieved from DCD donors the deleterious effects of cold ischaemia are superimposed on the injury sustained during warm ischaemia, which causes rapid depletion of ATP. There are some differences in the pattern of injury sustained during warm and cold ischaemia; the latter causes initial injury to sinusoidal endothelial cells whereas warm ischaemia is more damaging to hepatocytes and cholangiocytes.
Cold storage causes injury to the graft regardless of other factors. It has been known for many years that extended preservation time with cold preservation solution has a deleterious effect on organ viability [15,16] with a clear correlation between cold-ischaemia time and post-operative primary graft function. The rate of primary non-function was reported in a large registry analysis as 5.8% [17], although individual centre reports show a range of incidence. Although a strong correlation has been shown between preservation injury and subsequent acute rejection [18], this has not been confirmed in other reports [19]. Preservation injury is a more critical issue in higher risk donor organs -whereas good quality livers can tolerate preservation periods even up to 18 hours, higher risk grafts must be implanted much more quickly in order to reduce the risk of potentially fatal graft dysfunction after transplantation.

MACHINE PERFUSION
Early in the history of organ preservation and transplantation, pioneers in the field investigated machine perfusion. In the first half of the twentieth century Carrel, for example, perfused organs with normothermic, oxygenated serum and demonstrated gross viability for several days [20]. A number of early successful clinical liver transplants carried out by Dr Thomas Starzl, used machine perfusion with diluted blood under cold hyperbaric conditions [21]. This technique never gained popularity, partly due to its complexity and logistic challenges but also the subsequent introduction of effective cold flush preservation solutions. Subsequently research into extracorporeal machine perfusion largely centred on liver support [22].

HYPOTHERMIC MACHINE PERFUSION (HMP)
It has been demonstrated experimentally by several groups that hypothermic machine perfusion of the kidney significantly improves the preservation compared with static cold storage, in terms of immediate graft function and medium-term outcome of deceased kidneys and DCD organs specifically [23,24]. The mechanism of benefit is not fully understood, but probably relates both to the removal of metabolic products as well as the delivery of oxygen (although in the absence of formal oxygenation or specialist oxygen carrier molecules, oxygen delivery is likely to be limited).
HMP offers the additional benefit of some (limited) assessment of pre-transplant organ viability by measurement of perfusion pressures and also perfusate αGST levels and other biomarkers as a marker of cellular injury. It also provides the potential for local therapeutic intervention. Several groups have claimed advantages with hypothermic machine perfusions in liver transplantation and the first human clinical study of ex-vivo liver HMP is currently underway [25,26].

NORMOTHERMIC MACHINE PERFUSION (NMP)
There is accumulating evidence of the superiority of a more physiological approach using oxygenated blood at normal body temperature. Several studies demonstrate that the quality of preservation can be improved substantially by warm perfusion, by combining the avoidance of cooling with the maintenance of a supply of oxygen and nutrition [27,28]. These data are mainly based on organ retrieval, preservation, ex-situ-evaluation and transplantation in large animal models (pigs). In the experimental setting normothermic perfusion has been shown to resuscitate porcine livers subjected to 60 to 90 minutes of warm ischaemia [14,29]. Warm perfusion has the added advantage of allowing more effective viability assessment of the organs while on the circuit (using multiple perfusion dynamic and biochemical parameters), contrasting with the more limited parameters available on the hypothermic circuit [27]. Preclinical liver transplant experiments in the pig model from a number of centres [27,28,30] have shown that the normothermically preserved liver can be transplanted reliably and successfully after warm ischaemic injuries that do not allow survival using cold preservation. If these results were translated into clinical practice, this would generate a large new source of donor organs.
Normothermic perfusion has been shown as feasible in the setting of human renal transplantation [31], with a potential benefit of a reduction in the rate of delayed graft function in organs retrieved from extended criteria donors [32].

PRECLINICAL INVESTIGATIONS
The experimental precursor to the proposed normothermic perfusion system was developed 15 years ago and the method of perfusion of the isolated pig liver with autologous blood has since been extensively tested and refined, although the overall design of the perfusion circuit remains unchanged [33]. The circuit incorporates a centrifugal pump, membrane oxygenator, and heat exchanger. Arterial perfusion is directly pumped and the portal vein is perfused via a soft-shell reservoir using gravitational force. The addition of various substrates to the perfusion solution enables maintenance of metabolic function [34].
The initial preservation experiments were carried out to compare preservation by warm perfusion with conventional cold preservation [29]. Porcine livers were retrieved and stored for a period of 24 hours, either flushed with UW solution and placed in an icebox or attached immediately to the preservation circuit. Both groups of livers were then reperfused on the circuit for 24 hours (as a surrogate for transplantation) and markers of cellular injury and of synthetic and metabolic liver function were measured. These experiments demonstrated significant superiority of normothermic machine perfusion in terms of haemodynamic, biochemical and histological parameters Subsequent experiments investigated the use of oxygenated, normothermic perfusion in an experimental setting that reflected the clinical situation of DCD donor organ retrieval [14]. Perfusion with normothermic blood was again compared with static cold storage after 60 min of warm ischaemia. Normothermic perfused livers demonstrated recovery of function by synthetic function, substrate utilisation and perfusion haemodynamics. Furthermore these livers displayed less cellular injury as shown by hepatocellular enzymes. In contrast, cold stored livers showed no evidence of viability during reperfusion and massive necrosis on histological examination.
It is recognised that the combination of warm ischaemia and conventional cold preservation leads to a poor outcome in DCD liver transplantation [7]. In the experimental setting, it is possible to institute warm perfusion with minimal exposure of the organ to cooling. However, in contrast, the logistics of clinical multi-organ retrieval in a distant donor hospital are complex and would be simplified by a period of cold preservation prior to normothermic preservation. This would enable the liver to be retrieved in the normal way, transported in an ice box and then attached to the perfusion machine once back at the base hospital. This scenario was simulated in the same experimental model by inserting a period of cold preservation prior to normothermic preservation [35]. Porcine livers were subjected to 60 minutes of warm ischaemia and then assigned to either normothermic preservation for 24 hours or cold preservation in University of Wisconsin solution for 4 hours followed by 20 hours normothermic preservation to achieve a total preservation time of 24 hours [35]. Livers that underwent normothermic preservation throughout had superior bile production, metabolic activity (base deficit and greater glucose use), and less hepatocellular damage (transaminase levels), and sinusoidal endothelial cell dysfunction (hyaluronic acid). The histology of livers that had been exposed to 4 hours of cold preservation before normothermia showed more necrosis and destruction of architecture. A similar study investigated 60 minutes of warm ischemia followed by 1 hour of cold preservation before 23 hours of normothermic perfusion [36]. This also showed evidence of increased hepatocellular injury, sinusoidal cell injury, but no detriment in terms of protein synthesis (factor V), bile production or histological features. These studies, therefore, demonstrated the need for the warm preservation device to be transportable so that normothermic preservation can be instituted with a minimal period of cooling at the time of organ retrieval.
In order to confirm these results in a preclinical model of organ transplantation, a series of liver transplants in a pig model was performed [27]. In these experiments pig livers were cold-preserved or warm-preserved (using the same machine perfusion methodology as before) for either 5 hours or 20 hours, followed by liver transplantation. As a model of DBD and DCD clinical scenarios, organs were cold-perfused in situ either at the time of cessation of circulation (as in a DBD organ donation) or after 40 and 60 minutes of warm ischaemia (simulating DCD organ donation). The two preservation times were selected because 5 hours is comfortably within, and 20 hours substantially beyond, the limit of the conventional cold preservation technology in pigs (in which a generally accepted limit for survival is 12 hours). Similarly the 40 and 60 minute periods of warm ischaemia are considerably longer than would be acceptable in current clinical practice where warm ischaemia rarely exceeds 30 minutes. Indeed, success at 40 minutes would raise the realistic prospect of transplantation of donor livers from uncontrolled DCD donors.
There was no difference in outcome between the two groups at 5 hours of preservation. After 20 hours of preservation, there were significant advantages consistently in warm compared to cold preservation of both DBD and DCD organs. These advantages applied to postoperative enzyme release and animal survival. Notably, in the 20 hour warm-preserved groups, there was no difference in survival or postoperative transaminase levels in recipients of DBD compared to DCD (40 minute warm ischaemia) donor organs (86% versus 83%). At 60 minutes of warm ischaemia and 20 hours normothermic preservation, however, there were no survivors.
Analysis of haemodynamic, synthetic and metabolic parameters showed that those groups of livers that subsequently went on to successful transplant were predictable before transplantation on the basis of portal flow/pressure, acid-base homeostasis and several other biochemical parameters [27]. It may be concluded that normothermic perfusion, in this context, is not only a more effective means of organ preservation than conventional cold storage, but also that this method can be configured to provide an effective means of viability assessment [37].
The prototype version of the automated clinical investigation device has been tested and demonstrated to be effective during pre-clinical studies in which human livers, discarded as unsuitable for transplantation, were perfused for 24 hours. 13 such livers were perfused with human blood and the perfusion characteristics and control algorithms have been shown to be equally applicable to human as to pig livers (manuscript in preparation). More recently, the clinical trials device has been tested, using livers declined for clinical transplantation, and all key functional aspects of the device shown to be operational, including particularly transport to the donor hospital, automation and 24 hour perfusion.

PHASE 1 CLINICAL TRIAL DATA
A phase 1 clinical trial was opened at King's College Hospital in 2012 and extended to the Queen Elizabeth Hospital, Birmingham in 2013. The first patient was transplanted with a normothermically-perfused liver in February 2013. As of December 27 th 2013, the trial completed recruitment and transplanted the twentieth recipient with a liver preserved using the OrganOx metra device (in the configuration intended for the COPE study). In all these cases, perfusion parameters were stable with good acid-base maintenance (indicators of good outcome). Postoperatively, all patients have made good recoveries.

HYPOTHESIS
Normothermic machine perfusion (NMP) is superior to static cold storage (SCS) of human liver allografts for reduction of preservation injury.

Primary objective
To compare the effect of NMP to SCS in the prevention of preservation injury and graft dysfunction, as measured by peak transaminase levels in the first week following transplantation.

Primary outcome measure/endpoint
The primary endpoint is defined as the difference in peak serum aspartate transaminase level (AST) within 7 days post-transplant between the two treatment arms. Serum AST will be measured daily during the first post-transplant week, and the peak level will be defined as the highest of these values (in IU/L). In order to ensure consistency, the first post-transplant measurement should be taken at 12 to 24 hours post-reperfusion.
A number of studies have demonstrated a relationship between peak AST in the early post-transplant period and patient survival, graft survival, early graft dysfunction and primary non-function following liver transplantation [19,38]. Peak AST is also significantly elevated in liver allografts with histological evidence of moderate to severe perfusion injury [39,40].

Objective Outcome Measures
To compare graft and patient survival between NMP and SCS livers.
1. Primary non-function: irreversible graft dysfunction requiring emergency liver replacement during the first 10 days after liver transplantation, in the absence of technical or immunological causes. 2. Graft survival at 30 days and 6, 12 and 24 months following transplantation. 3. Patient survival at 30 days and 6, 12 and 24 months following transplantation.
To compare biochemical liver function between NMP and SCS livers. To compare evidence of reperfusion injury between NMP and SCS livers.
Histological evidence of reperfusion injury in post-reperfusion biopsies (taken immediately prior to abdominal closure). These will be compared to baseline pre-reperfusion biopsies (on removal of the liver from SCS/NMP) and graded using standard histological criteria [39,45] To compare evidence of ischaemic cholangiopathy between NMP and SCS livers.
Evidence of biliary stricturing on magnetic resonance cholangiography (MRCP) at 6 months post-transplant.
To assess the ability of perfusion parameters and biomarkers in perfusion fluids to predict clinical outcomes following transplantation.
1. Perfusion parameters (logged automatically by the device): a. Arterial and caval pressures (in mmHg) b. Arterial, portal and caval flow rates (in mmHg) c. pO2, pCO2 and pH d. Blood temperature ( o C), Glucose (mmol/L) and bile production (ml/h) 2. Perfusate ALT and AST at 15 minutes, 1 hour and the end of NMP 3. Perfusate IL6, TNF, vWF at 15 minutes, 1 hour and the end of NMP 4. In addition to these pre-specified outcomes, additional biological samples will be taken for the COPE WP7 bioresource at specified timepoints as detailed in appendix A1.
To assess the feasibility and safety of NMP as a method of organ storage and transportation.

THE COPE BIORESOURCE (WP7)
Work Package 7 (WP7) of the COPE project aims to establish a bio-repository using biological specimens from clinical trials to support research work relevant to improving the outcomes of kidney and liver transplantation. The additional samples collected for the bioresource will enable supplementary, separate evaluation of the different liver preservation modalities by:  assessing (immuno) histopathological criteria of donor livers in both arms (NMP and SCS)  examining the relationship between histological markers and donor demographics, donor management, preservation method and outcomes after transplantation in both arms Serum, perfusate, urine, bile and tissue samples will be used for research into molecular mechanisms of injury and repair during organ preservation. Assumed 'gold standard' markers will be validated and compared to novel molecular signatures and degradation products. These have been developed from identified pathophysiological pathways that might have potential to predict function and outcomes after transplantation.
It is recognised that high risk donor livers are increasingly being utilized to address the organ deficit. However, no accepted, universal, risk stratification tool exists to identify the quality of these organs and predict the outcomes of transplantation. Histological criteria have been suggested to evaluate extended criteria donor organs on pre-implantation biopsies. However, these have not been validated.
Whilst the implications of warm ischaemia on cellular injury are well recognised, our understanding of the molecular mechanisms that lead to liver dysfunction and early graft lost are not fully developed. End-organ injury as a result of cerebral damage and warm ischaemia leads to donor livers being further susceptible to preservation damage and ischaemia reperfusion injury. The ability of normothermic perfusion to condition donor livers will be assessed and compared to controls. Perfusion parameters for standardised liver assessment have also not been universally or widely accepted.
Thus, the aims of sample collection for WP7 are:  Relate pathology to demographics, function and outcomes, and evaluate current scoring systems  Use of next generation mass-spectrometry to analyse relevant segments of the portal triad and of bile with genomics, proteomics and metabolomics techniques to identify candidate biomarkers and validate both new and existing markers predicting outcome of transplantation.  To establish a simple assay for measurement in routine practice.  To perform multivariate analyses on perfusion parameters, combined with histological and molecular markers to develop a composite liver grading scoring system.  To identify novel pathways of injury and repair in donor livers.
WP7 will support further sub-studies and promote collaborations, through establishing an access policy for samples to be used in research projects outside of the funding scope of the FP7 grant.
The nature and timing of samples to be collected for the bioresource as part of the present study are detailed in appendix A1.

TRIAL DESIGN
This is a multicentre randomised controlled, non-blinded, clinical trial comparing static cold storage (SCS) versus normothermic machine perfusion (NMP) for organ preservation prior to liver transplantation. Following assessment of donor and recipient eligibility and confirmation of consent, the liver will be randomised to either NMP or SCS. At the end of preservation, the liver will be transplanted and the patient managed according to standard local practice and protocols.
Enrolled patients will participate in the study for 6 months, with outcomes assessed during the initial inpatient stay and at study visits at day 30 post-transplant and at month 6 post-transplant. Additional biochemical and survival data will be collected from routine clinical measurements taken in participating centres at 12 months and 24 months post-transplant.
Data will be collected into a secure central online electronic database using electronic case report forms.
Primary outcomes will be analysed and reported 30 days following enrolment of the last patient to the study. The study will close after the final patient has completed 24 months follow-up.
Anticipated flow of patients through the trial is depicted in figure 1.

TRIAL PARTICIPANTS
Participants will be adult patients active on the waiting list for liver transplantation at any of the participating transplant centres (other than those with acute/fulminant liver failure).

ELIGIBILITY CRITERIA
All eligibility criteria must be met at the time of randomisation.

DONOR CRITERIA
Inclusion: Donors over the age of 16 years. Liver allografts from donation after brain death (DBD), standard and extended criteria donors (SCD, ECD) and donation after circulatory death (DCD) donors.
Exclusion: Living donors; liver intended for split transplant; donor age <16 years; liver in which investigator is unwilling to randomise to either arm.

RECIPIENT CRITERIA
Inclusion: Adult patients (18 years or more), active on the waiting list for liver transplantation; able to give informed consent.
Exclusion: Age less than 18 years; acute/fulminant liver failure; transplantation of more than one organ (e.g. liver and kidney); refusal of informed consent; unable to give informed consent.

TRIAL PROCEDURES
The participant timeline is illustrated in appendix A2. The following section provides details of this timeline and all study procedures.

RECRUITMENT
The emergency nature of liver transplantation means that once a potential recruit is called in for a transplant there will only be a 3-4 hour window for the consent and screening process to occur. This does not allow sufficient time for the potential participant to consider the implications of participating in the study. For this reason, all patients who fulfil the entry criteria and who are on the waiting list for liver transplantation at the participating centres will be approached in advance of the study either during a routine clinic appointment, inpatient admission or in advance of discussion by letter. If the patient expresses interest in the study, a faceto-face meeting will be arranged during a routine admission or outpatient appointment. Detailed information will be given both verbally and in the form of a patient information sheet. The study coordinator and/or a medically qualified researcher will give information.

RECIPIENT CONSENT (STUDY VISIT 1)
Patients will be given a minimum of 24 hours (if required) to consider the information about the trial before deciding whether to give written consent to participation. The study coordinator and the liver transplant coordinators at each centre will maintain a list of consenting patients.
When a suitable donor organ becomes available and this is allocated to the recipient patient who has provided consent, then the recipient patient will be asked to affirm the consent that has already been provided by signing and dating the informed consent form for a second time; this process may also be done with a documented phone call. If the patient is not willing to proceed in the study at this stage, then the organ retrieval and preservation will be carried out using conventional procedures. The recipient will also be required to complete a surgical consent form for the transplant procedure as per standard local policy.
The participant must personally sign and date the latest approved version of the informed consent form before any study specific procedures are performed. Written and verbal versions of the participant information sheet and informed consent form will be presented to the participants detailing the exact nature of the study, the implications and constraints of the protocol, the known side effects and any risks involved in taking part. It will be clearly stated that the participant is free to withdraw from the study at any time for any reason without prejudice to future care, and with no obligation to give the reason for withdrawal. The participant will be allowed as much time as possible to consider the information, and the opportunity to question the investigator, or other members of the clinical or research team, to decide whether to participate in the study.
Written informed consent will be documented by means of a dated signature from the participant and dated signature from the person who presented and obtained the informed consent. The person who obtains consent must be: 1. suitably qualified and capable of providing information about the study; 2. capable of answering questions about the study or ensuring that such questions are answered by a suitably qualified individual; 3. authorised to do so by the Chief Investigator. A copy of the signed Patient Information Leaflet and the signed and dated consent form will be given to the participant. The original signed form will be retained at the study site and a copy will be placed in the medical notes.
Subjects are free to withdraw consent at any time, irrespective of their initial consent. Subjects who withdraw consent before 6 months post-operatively will be replaced.
Each subject must also give permission for the sponsor's representatives to review their hospital records for the purpose of source document verification.
The subject's general practitioner/family doctor will be informed of their participation in the study. A letter to the participants GP will be produced.

DONOR CONSENT
Explicit consent from the donor family is not required for participation in the study as the donation process is unaltered by participation in the trial (no intervention occurs prior to donation), and the intervention does not affect the transplantability of the donor organ.
During the course of the study, donor details will be kept anonymous (specific study identification codes will be used for each study donor). Donor data will only be made available to authorised staff of the study sponsor, its authorised representatives and regulatory authorities

SAMPLES FOR BIORESOURCE
Additional biological samples will be obtained to be stored centrally for use in future studies of the pathobiology of liver transplantation (COPE WP7, as described in section 3.3). A materials consent from the recipient will be obtained to specifically address the collection of these histological specimens and plasma, serum and urine samples. Refusal to consent for the storage of these samples will not preclude inclusion in the trial.
All samples will be collected in accordance with national regulations and requirements including standard operating procedures for logistics and infrastructure. In the UK, samples will be taken in appropriately licensed premises, stored and transported in accordance with the HTA guidelines and local trust policies. Samples for long-term storage will be kept in the Oxford Radcliffe bioresource. The stored tissues will be held under an extension of the University of Oxford's HTA license (12217).

RECIPIENT ASSESSMENT
All patients on the transplant waiting list in participating centres will have been screened for suitability for transplantation; further screening assessment is not required as part of the present trial. On offer of a suitable donor organ consent will be confirmed as described in section 6.2.1. The online randomisation tool will require confirmation that the recipient meets the inclusion criteria for the trial and require the entry of baseline demographic information prior to randomisation and release of the randomisation code.
On admission to hospital, the recipient will be assessed for fitness to proceed to transplant according to local procedures. If a recipient is deemed unfit for transplant at the time of admission, they will no longer be active on the transplant waiting list and as such will be excluded from the trial. Any data collected from these individuals will not be included in the trial and will be deleted.

DONOR ASSESSMENT
The transplant recipient co-ordinator is usually an advanced practice nurse, one of whose roles is to facilitate calls for organ offers and co-ordinate all aspects of the transplant process. On receiving an organ offer, the local recipient co-ordinator/surgeon will ascertain baseline demographic information from the offering organisation to assess eligibility of the liver for inclusion in the trial.

SEQUENCE GENERATION
Participants will be randomly assigned to NMP or SCS with 1:1 allocation as per a computer generated randomisation schedule using variable block randomisation using the following stratification factors:participating (recipient) centre and by donor type (DBD or DCD).

ALLOCATION CONCEALMENT MECHANISM
Allocation concealment will be ensured by use of central computerised randomisation (with telephone backup). Allocation will not be revealed until the patient has been recruited to the trial and donor and recipient baseline characteristics have been recorded. Random permuted block length will be used; block sizes will not be disclosed.

IMPLEMENTATION
Prior to study enrolment, the local investigator will confirm the availability of the NMP device. Once informed consent has been obtained from the potential recipient of an organ offer, the local investigators will login to an online data collection and randomisation tool. This will require confirmation of consent, as well as compliance with inclusion and exclusion criteria. Baseline recipient and donor characteristics will be recorded by the recipient co-ordinator and are required prior to release of the randomisation group. This will ensure that the patient is enrolled to the trial prior to allocation.

BLINDING/MASKING
Whilst it is not possible to blind the local investigators to the method of organ preservation, outcome assessors will be blinded where possible. This includes the histopathologist interpreting the biopsy specimens as well as the radiologist interpreting the 6-month MRCP images.

DONOR DEMOGRAPHICS
Donor demographics to be recorded will include the following:  Age  Sex  Race (White, African-American, other)  Cause of death (CVA, hypoxia, trauma, other)  Type of donor (DBD, DCD)  Donor height  Donor risk index (DRI) [3]  Peak serum AST  Peak serum ALT  Peak serum sodium  Peak GGT  Length of ITU stay  BMI

RECIPIENT DEMOGRAPHICS
Recipient demographics to be recorded will include the following:  Age  Sex  Aetiology of liver disease  Indication for transplant  MELD score (based on INR, creatinine, bilirubin)  BMI  Creatinine clearance A baseline assessment of quality of life using the EQ-5D-5L will be performed at the time of initial consent to the study, which may be when the patient was first put on the liver transplant waiting list.

NMP GROUP
If the liver is randomised to the NMP group, arrangements will be made to transport the device to the donor hospital (see section 7.6). The recipient co-ordinator will also request that the donor co-ordinator arrange for 3 units of donor-type red blood cells to be cross-matched at the donor centre for use in the OrganOx device. Following the routine retrieval procedure at the donor hospital the liver will be placed in ice-cold perfusion solution (according to local protocol) on the back-table, and prepared for cannulation. The procedure for preparing the device for use and placing the organ on the device is described in detail in the instructions for use (IFU) document (version 3.0, 25/05/2012). The device is then transported to the recipient transplant centre. The procedure for removing the liver from the device is also described in the IFU. Implantation and reperfusion of the liver proceed as per the usual practice of the implanting centre. The duration of machine perfusion will be dictated by recipient theatre logistics and local policy, but should not be less than 4 hours or more than 24 hours.
If cannulation proves impossible, the liver will be transported using standard static cold storage as described below. Results will be analysed in the randomised group (intention-to-treat).

SCS GROUP
Following the routine retrieval procedure, the liver will be placed in ice-cold perfusion solution (according to local protocols) on the back-table, followed by storage in cold perfusion solution within an icebox. The organ will be transported to the recipient centre, and removed from storage prior to implantation for standard backtable preparation. The duration of cold storage will be dictated by logistics and local policy.

RECORDING OF OPERATIVE AND PERFUSION PARAMETERS
The retrieving and implanting research teams will record the following data:

Donor timings
The times to be recorded for DBD donors (SCS group) are as follows:  Cessation of donor circulation (cross clamp)  Start of cold perfusion (should be the same unless technical problem)  Liver removal and placement on ice  Removal from ice (recipient operation)  Portal reperfusion  Arterial reperfusion The times to be recorded for DCD donors (SCS group) are as follows:  Withdrawal of support  Onset of functional warm ischaemia (SBP < 50 mmHg)  Cessation of donor circulation  Start of cold perfusion  Liver removal and placement on ice  Removal from ice (recipient operation)  Portal reperfusion  Arterial reperfusion In addition, the following times will be recorded for all (DBD and DCD) machine perfused livers:  Initiation of normothermic machine preservation  Cessation of normothermic machine preservation (cold flush)

Preservation parameters
In addition to timings, a number of other preservation parameters will be recorded. These will include: In addition to these pre-specified outcomes, additional biological samples will be taken for the COPE WP7 bioresource at prespecified timepoints as detailed in appendix A1.
At the end of preservation a sample of perfusate/storage solution will be taken for microbiological culture (SCS and NMP groups).

Operative parameters
These will include:  Total operative time: defined as time from knife-to-skin to skin closure.  Anastomotic time (secondary warm ischaemia): defined as time between removal of organ from ice (SCS) or perfusion device (NMP) to organ reperfusion (portal or arterial)  Presence of post-reperfusion syndrome (defined as a decrease in mean arterial pressure (MAP) of more than 30% from the pre-implantation baseline value for more than one minute during the first five minutes after reperfusion [43,44])  Use of vasopressors prior to and after reperfusion  Intraoperative transfusion of blood products measured in units.  The use of veno-venous bypass or porto-caval shunts  Type of caval anastomosis (standard end-end, piggyback (end-side or side-side)

Histological evidence of ischaemia-reperfusion injury
Graft biopsies will be taken immediately prior to abdominal closure and examined for evidence of reperfusion injury. These biopsies will be compared to the baseline biopsies prior to organ reperfusion and graded according to standard histological criteria [39,45]. The trial histopathologist, who will be blinded to the method of perfusion, will assess all biopsies.

CONCOMITANT CARE
All other aspects of the retrieval procedure will be carried out according to local policies and national guidelines.
Recipient management including the implantation procedure, postoperative care, immunosuppression and other medications, and post-transplant monitoring will follow local protocols.

INPATIENT STAY
Patients will be assessed daily by their clinical team and managed according to normal local protocols.

Outcome assessment
The following biochemical outcomes will be recorded:  Reoperation rate  Any other adverse event Severity will be graded according to the Clavien-Dindo classification [45] as described in Appendix A4.

Immunosuppression
Details of induction immunosuppression and maintenance immunosuppression (including doses) at day 7 posttransplant will be recorded.

STUDY VISIT 2 -DAY 30
This visit will, where possible, coincide with a routine outpatient appointment. If the recipient is an inpatient, assessment will be made in hospital where appropriate.

Outcome assessment
The following biochemical outcomes will be recorded at day 30 post-transplant:  Serum bilirubin (measured in μmol/l)  Serum gamma-glutamyl transferase (GGT; measured in IU/L)  Serum aspartate transaminase (AST; measured in IU/L)  International normalised ratio (INR) Other outcomes to be recorded include:  Graft and patient survival at day 30 post-transplant  Requirement for renal replacement therapy (HD, HF, HDF) at any time  Estimated Glomerular Filtration Rate (eGFR)  Healthcare resource use (by means of a patient-completed log, hospital admissions data and medical records)  Quality of life (measured by means of the EQ-5D-5L questionnaire).

Safety outcomes
 Recipient infection (defined as a positive microbiological culture result)  Biopsy-proven acute rejection episodes  Biliary complications (biliary strictures -anastomotic and non-anastomotic, bile duct leaks)  Vascular complications (bleeding, hepatic artery stenosis, hepatic artery thrombosis, portal vein thrombosis, portal vein stenosis)  Reoperation rate  Any other adverse event Severity will be graded according to the Clavien-Dindo classification [45].

Immunosuppression
Details of maintenance immunosuppression (including doses) at day 7 and day 30 post-transplant will be recorded.

STUDY VISIT 3 -MONTH 6
This visit will, where possible, coincide with a routine outpatient appointment. If the recipient is an inpatient, assessment will be made in hospital where appropriate.

Outcome assessment
The following biochemical outcomes will be recorded at month 6 post-transplant:  Serum bilirubin (measured in μmol/l)  Serum gamma-glutamyl transferase (GGT; measured in IU/L)  Serum aspartate transaminase (AST; measured in IU/L)  International normalised ratio (INR) Other outcomes to be recorded include:  Graft and patient survival at month 6 post-transplant  Requirement for renal replacement therapy (HD, HF, HDF) at any time  Estimated Glomerular Filtration Rate (eGFR)  Healthcare resource use (by means of a patient-completed log, hospital admissions data and medical records). The resource use log will be collected from the patient at this visit.
 Quality of life (measured by means of the EQ-5D-5L questionnaire).

MRCP evidence of ischaemic cholangiopathy
All study participants will undergo magnetic resonance cholangiopancreatography (MRCP) with T2-weighted turbo-spin echo sequences at 6 months post-transplant unless contraindicated. Contraindications to MRI include:  Implanted pacemaker, defibrillator or metal heart valve  Implanted pump device (such as an insulin or pain medication pump)  Inner ear implant  Aneurysm clip within the brain  Intrauterine device (IUD)  Metal in the eyes (at any time), or history of being a metal worker  Current pregnancy  Intractable claustrophobia The trial radiologist, who will be blinded to the preservation method, will assess all MRCPs. Evidence of ischaemic cholangiopathy will be taken as the presence of extra-anastomotic biliary structuring in the absence of hepatic artery thrombosis [47,48].

Safety outcomes
 Recipient infection (defined as a positive microbiological culture result)  Biopsy-proven acute rejection episodes  Biliary complications (biliary strictures -anastomotic and non-anastomotic, bile duct leaks)  Vascular complications (bleeding, hepatic artery stenosis, hepatic artery thrombosis, portal vein thrombosis, portal vein stenosis)  Reoperation rate  Any other adverse event Severity will be graded according to the Clavien-Dindo classification [45].

Immunosuppression
Details of maintenance immunosuppression (including doses) at 6 months post-transplant will be recorded.

LATER OUTCOMES
Whilst the end-point for trial participation will be 6 months, we also intend to collect graft and patient survival data, adverse event information and serum biochemistry values from consenting participants at 12 and 24 months post-transplant. These data will be collected from participating centers and will be measured as part of routine clinical care, not requiring additional visits or interventions on the part of the trial participant. The biochemical parameters to be recorded are:  Serum bilirubin (measured in μmol/l)  Serum gamma-glutamyl transferase (GGT; measured in IU/L)  Serum aspartate transaminase (AST; measured in IU/L)  International normalised ratio (INR)

HEALTH ECONOMIC ANALYSIS
An economic analysis will be performed by the trial Health Economist, with the objective of estimating average costs and effectiveness in each arm of the study. This will inform a cost-effectiveness analysis using a health service perspective and incremental cost effectiveness ratios (ICER's) will be reported. Quality adjusted survival will be obtained by administration of the EuroQol EQ-5D-5L questionnaire (Appendix A3). Quality of life data will be collected at baseline (pre-transplant, at time of consent) and at each study follow-up visit following liver transplantation (day 30 and month 6).
Costs will be estimated based upon measured resource use and national unit costs. Resources will include machine and disposables costs, immunosuppression and other drugs, inpatient hospital stays (including intensive care days), radiological investigations, biopsies and other procedures, outpatient visit and visits to the family doctor. Resource use will be identified from case report forms, hospital episode statistics/insurer claims and from patient self-reporting using a simple log/questionnaire (to assess out-of hospital resource use). These questionnaires will be kept by the patient during the study and collected at the final study visit. Resource use will be transferred to an eCRF, and the original document kept at the participating centre as source material.

PARTICIPANT RETENTION
All randomised patients completing the 6-month follow-up assessment will be regarded as having completed the primary study. All patients will be encouraged to complete study follow-up, and all reasonable efforts will be made to ensure completeness of follow-up. Measures include ensuring that assessments are made, where possible, at routine hospital visits rather than additional appointments, and that patients do not incur extra financial costs (e.g. travelling costs) as a result of study participation.
It is understood that study participants may withdraw consent for study participation at any time irrespective of their reasons. The investigators may also withdraw a recipient from the study in order to protect their safety and/or if they are unwilling or unable to comply with the required study procedures. We will keep all data accrued to the point of withdrawal unless the participant requests otherwise, as is stipulated in the trial consent form.
Possible reasons for investigator-led withdrawal of a participant from the trial include:  Ineligibility either arising during the trial or retrospectively having been overlooked at screening  Significant protocol deviation  Significant non-compliance with trial requirements  An adverse event which leads to inability to comply with trial procedures  Disease progression which results in an inability to comply with trial procedures  Withdrawal of consent  Loss to follow-up In the event a patient withdrawing from the trial, the reason for withdrawal must be documented on the eCRF. Such patients will be asked whether they consent to data accrued before the date of withdrawal being included in the trial analysis.

DEFINITION OF THE END OF THE TRIAL
Data will be collected from participants for 24 months post-transplant. Once all data from all samples has been processed, then the database will be locked and the trial will end.
The procedures for the early termination/suspension of the study at one or more clinical sites in light of safety or compliance concerns are detailed in section 8.3.

ORGANOX LIMITED
OrganOx Limited is a late-stage medical device development company that was founded in April 2008 as a spin-out from the University of Oxford.

THE ORGANOX METRA
The OrganOx metra is a normothermic preservation device for use in human liver transplantation. It perfuses the donor liver with blood, oxygen and nutrients, as well as a number of medications, at normal body temperature to mimic ideal physiological conditions and preserve the organ for up to 24 hours. The device provides information as to the haemodynamic, synthetic and metabolic function of the liver whilst being perfused which may assist the clinician in assessing the organ's suitability for transplantation.

THE ORGANOX METRA BASE UNIT
The OrganOx metra normothermic perfusion device incorporates a centrifugal pump, an oxygenator, oxygen concentrator, heat exchanger, reservoir, flow probes, pressure sensors, infusions and blood gas analyser together with tubing and connector components. The device is comprised of three main components:  a reusable base unit which contains software and hardware  a disposable plastic circuit  a set of perfusion solutions suitable for 24 hours perfusion

DISPOSABLE SET
The disposable set used with the core base unit of the OrganOx metra contains all the disposables used with each organ recovery on the metra and comprises: 1. A disposable tubing set, including a blood reservoir, perfusion lines, a blood oxygenator and centrifugal pump-head together with flow and pressure sensors. 2. An organ storage bowl which is pre-connected to the tubing set to contain the organ while on the device. 3. Cannulae for the coeliac artery, portal vein and inferior vena cava with easy connection attachment to the perfusion circuit. 4. A cannula and connection point for bile collection 5. Blood gas sensors for monitoring pO2, pCO2 and pH by means of on-line blood gas analysis.

PERFUSION SOLUTIONS
For the present study all the additives necessary to perfuse and maintain the organ during the storage process, with the exception of sodium taurocholate, are not included and will be sourced locally (OrganOx will provide a list of recommended suppliers in the Instructions for Use (IFU) document). These solutions include bolus injections (given at the start of perfusion) and the maintenance infusions (given throughout perfusion).
The primary perfusion fluid for the liver comprises packed red blood cells, supplemented by colloid solution to normalise the haematocrit and osmolarity-these two components are not included and will be sourced locally.
Before connection of the liver the blood-based perfusate is supplemented with:  Cefuroxime (antibiotic).  Heparin (anticoagulant) to prevent thrombosis in the circuit. In clinical use, a half-life of ~90 minutes is assumed; on this basis heparin is also given as a maintenance infusion.
 Sodium bicarbonate (buffer) for adjusting the pH of the perfusate before the liver is placed on the device.
 Calcium gluconate to correct the binding of citrate to calcium.
During the perfusion the following are infused at a constant rate:  Parenteral nutrition solution -a source of amino acids and glucose for liver maintenance.  Insulin to control the perfusate glucose level  Heparin to maintain anticoagulation.  A 2% solution of sodium taurocholate in isotonic saline to compensate for loss of bile salts.  Prostacyclin to optimise micro-perfusion.
The primary fluid for perfusing the organ is packed red cells supplied from blood transfusion centres and supplemented by a commercially-available colloid solution to normalise the haematocrit and osmolarity. Further additions are made to the perfusate to support the liver (in a similar mode to current preservation solutions such as Viaspan® or those used in machine preservation devices such as the Waters RM3® or the Organ Recovery Systems LifePort®). All solutions required will be attached to the circuit during set-up and before the liver is attached. The organ retrieval team will provide the solutions necessary for perfusion with the metra including the packed red blood cells. All solutions are prepared immediately before the organ is attached to the device and contain sufficient solution for 24 hours operation, the intended maximum perfusion time for a liver on the device.

DEVICE SAFETY
In designing the metra, OrganOx has made every attempt to maintain the current practices of organ retrieval and transplant teams, in order to minimise the risk of complications or errors that would prevent a successful retrieval. From a regulatory standpoint, it is important to note that the metra is an organ preservation system and its use does not involve direct connection to either the donor or recipient at any time.
The device has been designed according to ISO 13485, the standard that stipulates the requirements for a comprehensive management system for the design and manufacture of medical devices. In addition ISO 14971 specifies a process for a manufacturer to identify the hazards associated with medical devices to estimate and evaluate the associated risks, to control these risks, and to monitor the effectiveness of the controls. As part of the development of the device an extensive risk analysis has been undertaken and the risks identified and minimised in accordance with this standard. As a result, any remaining risk can only be investigated by a clinical transplant study. The OrganOx perfusion system is based on the principle that all the perfusion solutions, additives and packed red cells must be removed from the organ prior to transplant. Therefore following the completion of the perfusion, the perfusion solution is flushed out of the organ with HTK solution. OrganOx has deliberately designed the operation of the device such that it will require minimal changes to current transplant clinical practice.

DEVICE LABELLING
All components of the OrganOx metra system (reusable base unit and disposable set) will be labelled by OrganOx as "Exclusively for Clinical Investigation". Labelling will also include the Sponsor name, contact details and a unique trial identifier.

DEVICE ACCOUNTABILITY
Device accountability will be undertaken at the each local site throughout the study for the reusable unit(s) and disposable sets (sterilisation/assembly batch number and disposable set number). The manufacturer and lot number for each perfusion solution will also be recorded on the case report forms (CRFs). The site will maintain a log of usage of both the retained unit, disposable set and perfusion solutions used throughout the study recording the lot number used against each subject (on the CRF).
At the end of each procedure the OrganOx metra, and any unused disposable and perfusion solutions will be removed from the donor hospital and returned to the investigator centre. Details of total numbers of disposable sets taken on-site and off-site will also be recorded.

DEVICE MAINTENANCE
Device cleaning and routine maintenance will be the responsibility of the local investigator storing the device. Full details for cleaning and routine maintenance required will be provided in the instructions for use (IFU), and appropriate training will be provided as part of the device training described in section 11.3.

LOGISTICAL CONSIDERATIONS
The OrganOx metra device has a number of constraints that create specific logistical considerations:  The device is large, and therefore may not fit in normal retrieval transportation depending on local procedures.
 The device has a battery life of 2.5 hours, and so transportation with a means of providing AC power is required.
 The device requires the manual assessment and input of perfusate glucose every 4 hours during perfusion.

LOGISTICAL ARRANGEMENTS FOR UK SITES
The OrganOx metra device will be stored and maintained at a central location in Oxford. When an eligible donor organ is offered to a participating recipient centre, the local recipient co-ordinator/surgeon will check to see if the named recipient has consented to take part in the study. If this is the case, then the coordinator will check whether the OrganOx metra device and support team are available (it will not be possible to support multiple retrievals simultaneously). If the device and support team are available, then the named recipient will be contacted, informed consent confirmed verbally over the telephone, and checks made that the recipient still fulfils all inclusion/exclusion criteria. The recipient co-ordinator/surgeon will then use the online randomisation tool to enter baseline information and randomise the recipient to treatment or control arms.
The recipient co-ordinator/surgeon will then contact the lead retrieval surgeon to inform him/her of the allocation group. If the liver has been randomised to machine perfusion, the recipient co-ordinator/surgeon will also contact the central co-ordinating team to arrange dispatch of the device to the donor hospital in a specially-adapted vehicle with an AC power invertor. The recipient co-ordinator will also request that the donor co-ordinator arrange for 3 units of donor-type red blood cells to be cross-matched at the donor centre for use in the OrganOx metra device. Following retrieval, the same vehicle will then transport the device with the liver to the recipient centre. One or more members of the research team from Oxford will accompany the device, to provide support and training in the use of the device and assist in the collection of samples.
Once the liver is removed from the device at the recipient centre, the vehicle will transport the device and researchers back to Oxford for cleaning and retrieval of perfusion logs.

LOGISTICAL ARRANGEMENTS FOR EUROPEAN SITES
An OrganOx metra device will be stored and maintained at a central location in each of the participating centres. When an eligible donor organ is offered to the recipient centre, the local recipient coordinator/surgeon will check to see if the named recipient has consented to take part in the study and confirm availability of the OrganOx metra device. If the device is available, the named recipient will be contacted, informed consent confirmed verbally over the telephone, and checks made that the recipient still fulfils all inclusion/exclusion criteria. The recipient co-ordinator/surgeon/research team member will then use the online randomisation tool to enter baseline information and randomise the recipient to treatment or control arms.
If the liver has been randomised to machine perfusion, the recipient co-ordinator will request that the donor co-ordinator arrange for 3 units of donor-type red blood cells to be cross-matched at the donor centre for use in the OrganOx metra device. The device will be transported to the donor hospital in a vehicle equipped with a power invertor, and from the donor hospital back to the participating recipient centre once retrieval is complete. One or more members of the research team from the participating centre will accompany the device, to place the liver onto the device and collect the necessary study samples.

DATA MONITORING COMMITTEE
The trial has a data monitoring committee (DMC) which consists of at least four independent members, including clinicians with relevant expertise and a statistical expert, independent from the Investigators and the funding source. The DMC will periodically review accruing data to safeguard the interests of the trial participants, potential participants and future patients and assess the safety of the interventions. The DMC will advise the Trial Management Committee if, in its view, the study should be terminated due to major clinical disadvantages in one of the study arms.
A separate DMC charter will contain full details of the committee and its roles and reporting structure.

INTERIM ANALYSES
Interim analyses of primary and secondary outcomes are not planned. They will only be performed if requested by the DMC on the grounds of participant safety.

DEFINITIONS Adverse Event (AE)
Any untoward medical occurrence, unintended disease or injury, or untoward clinical signs (including abnormal laboratory findings) whether or not related to the study intervention.

Serious Adverse Event (SAE)
An adverse event that:  Led to death  Resulted in serious deterioration in the health of the subject that: o resulted in a life-threatening illness or injury o resulted in a permanent impairment of a body structure or a body function o required in-patient care or prolongation of hospitalisation o resulted in medical or surgical intervention to prevent life-threatening illness or injury or permanent impairment to a body structure or a body function.
This includes device deficiencies that might have led to a serious adverse event if: a) suitable action had not been taken or b) intervention had not been made or c) circumstances had been less fortunate.
These are handled under the SAE reporting system.
Planned hospitalisation for a pre-existing condition, or a procedure required by the trial protocol, without serious deterioration in health, is not considered a serious adverse event.

Adverse Device Effect (ADE)
An adverse event related to the use of an investigational medical device. This definition includes any events resulting from insufficient or inadequate instructions for use, deployment, implantation, installation, or operation, or any malfunction of the investigational device. This definition also includes any event resulting from user error or form intentional misuse of the investigational device.

Serious Adverse Device Effect (SADE)
Any untoward medical occurrence that can be attributed wholly or partly to the device, which resulted in any of the characteristics of a serious adverse event as described above.

Unanticipated Serious Adverse Device Effects (USADE)
Any serious adverse device effect which, by its nature, incidence, severity or outcome, has not been identified in section 8.2.2.

Device Deficiency
Inadequacy of a medical device with respect to its identity, quality, durability, reliability, safety or performance. Device deficiencies include malfunctions, use errors and inadequate labelling. Device deficiencies resulting in SADEs will be managed as detailed in section 8.2.3.
Device deficiencies that did not lead to an adverse event, but could have led to a medical occurrence if suitable action had not been taken, or intervention had not been made or if circumstances had been less fortunate will also be managed as detailed in section 8.2.3.

Use error
Act or omission of an act that results in a different medical device response than intended by the manufacturer or expected by the user. Use error includes slips, lapses and mistakes. An unexpected physiological response of the subject does not itself constitute a use error.

Severity definitions
The following definitions will be used to determine the severity rating for all adverse events: Mild: awareness of signs or symptoms, that does not interfere with the subject's usual activity or is transient that resolved without treatment and with no sequelae.
Moderate: a sign or symptom, which interferes with the subject's usual activity.
Severe: incapacity with inability to do work or perform usual activities.  Clinically significant abnormal laboratory finding or other abnormal assessments that is associated with the condition being studied (unless judged by the investigator as more severe than expected for the patient's condition).

ANTICIPATED ADVERSE EVENTS
The investigator will exercise his/her medical judgment in deciding whether an abnormal laboratory finding or other abnormal assessment is clinically significant. However, if in the opinion of the investigator, the frequency or severity of the event is greater than would be expected then it must be reported.

PROCEDURES FOR RECORDING ADVERSE EVENTS AND DEVICE DEFICIENCIES
It is the responsibility of the local investigator to ensure that all adverse events (including ADEs) and device deficiencies occurring during the course of the study are recorded. This will include but not be limited to:  A description of the event  The dates of the onset and resolution  Action taken  Outcome  Assessment of relatedness to the device  Whether the AE is serious or not  Whether the AE arises from device deficiency  Whether the AE arises from user error Adverse events that occur during the course of the study should be treated by established standards of care that will protect the life and health of the study subjects It is the responsibility of the local investigator to collect all directly observed adverse events and all adverse events spontaneously reported by the subject. In addition each subject should be questioned about adverse events at each visit. Adverse events should be recorded on provided serious adverse event data collection forms.

REPORTING PROCEDURES FOR ALL SERIOUS ADVERSE EVENTS
Reporting of all Serious Adverse Events will be done in accordance with the European Commission Guidelines on Medical Devices Serious Adverse Event Reporting (MEDDEV 2.7/3; December 2010).
It is the responsibility of the local investigator to ensure that all adverse events which fall in to the category of Serious Adverse Events (SAEs) and any device deficiencies (including Serious Adverse Device Effects (SADEs)) are reported to the co-ordinating centre, chief investigator, central investigators and, if required, to their local R&D department as soon as possible after becoming aware of the event but no later than 24 hours. Details to be included in the report are as section 8.2.3.
Adverse event and serious adverse event reporting will be via the electronic data collection tool using the COPE SAE form, with SAEs being automatically forwarded to the Trials Co-ordinator and clinical reviewers by the reporting tool. The clinical reviewers are the Chief Investigator and Central Investigators, who are independent of any of the recruiting clinical sites. Reporting by Fax will provide a backup system (+44 (0) 1865 572398) in the event that the online data collection tool is unavailable. The Fax machine is located in the central coordinating centre based in SITU.
Within the following 5 working days, the local investigator may be required to provide additional information on the SAE or device deficiency in the form of a written narrative. This should include a copy of the completed SAE form, and any other diagnostic or relevant information that will assist the understanding of the event.
Significant new information on ongoing serious adverse events should be provided promptly to the coordinating centre and clinical reviewers using the same electronic COPE SAE form.
On submission of an electronic SAE form, the co-ordinating centre and all of the clinical reviewers will be immediately notified by email. They will review SAEs and, if they feel they pose an immediate risk to patient health or safety, then they will report them to the DMC immediately and to the device manufacturer, competent authority and the REC within 2 calendar days of the Chief Investigator becoming aware of the event.
All other reported SAEs will be reported to the DMC and competent authority within 7 calendar days of notification, if appropriate. This will not include SAEs that may be expected as part of the risks of liver transplant surgery. Adverse device events (SADEs, USADEs) and device deficiencies will also be reported to the device manufacturer. All SAEs will be followed up to resolution. The DMC will review the accumulating data at regular intervals.
In the other European centres SAE reporting will be done in accordance with their local regulatory requirements. The Chief Investigator will also inform all investigators concerned and the device manufacturer of relevant information about USADEs that could adversely affect the safety of participants.

STUDY SUSPENSION OR EARLY TERMINATION
The DMC or sponsor may recommend suspension or termination of the study either at an individual investigation site or the entire study for significant and documented reasons. An investigator, ethics committee or regulatory authority may suspend or prematurely terminate participation in the study at the investigation sites for which they are responsible. If suspicion of an unacceptable risk to subjects arises during the study, or when so instructed by the ethics committee or regulatory authorities, the sponsor shall suspend the study while the risk is assessed. The sponsor shall terminate the study if an unacceptable risk is confirmed.
The sponsor shall consider terminating or suspending the participation of a particular study site or investigator in the study if monitoring or auditing identifies serious or repeated deviations on the part of an investigator.
If suspension or premature termination occurs, the terminating party shall justify its decision in writing and promptly inform the other parties with whom they are in direct communication. The chief investigator and sponsor shall keep each other informed of any communication received from either the ethics committee or the regulatory authority.
If, for any reason, the sponsor suspends or prematurely terminates the study at an individual investigation site, the sponsor shall inform the responsible regulatory authority as appropriate and ensure that the Ethics Committee is notified, either by the chief investigator or by the sponsor. If the suspension or premature termination was in the interest of safety, the sponsor shall inform all other investigators.
If suspension or premature termination occurs, a) the sponsor shall remain responsible for providing resources to fulfil the obligations from the study protocol and existing agreements for following up the subjects enrolled in the study, and b) the chief investigator or authorized designee shall promptly inform the enrolled subjects at his/her study site, if appropriate.
Device deficiencies and use errors not falling into the categories of ADEs or SADEs should be reported via the online data collection tool and will be collected by the study investigators for investigation by the manufacturer.
SAE reporting will continue until the last patient recruited has completed 12 months of follow-up. Patients transferred back for on-going care to referring centres will have their data including AEs related to the outcome measures collected by data collection forms sent to the patient's specialist. Patient cards will be provided to all participants of the study, with a contact telephone number (research nurse / researcher) to inform regarding the occurrence of SAEs.

OUTCOMES
Primary outcome reporting will occur after collection and analysis of data at 30 days following the enrolment of the last patient to the trial.
The intervention (NMP) will be compared against control (SCS) for all primary and secondary outcomes. Primary outcome, difference in peak serum (AST) level, will be analysed using ANOVA with adjustment for stratification factors. Binary outcomes will be assessed using chi-squared test and logistic regression to adjust for potential confounders. Continuous outcomes will be compared using the T-test if normally distributed, or by the Mann-Whitney U test. Time-to-event outcomes will be analysed using survival analysis methods, including Kaplan-Meier and Cox proportional hazards regression model with calculation of hazard ratios.
Outcomes will be reported with 95% confidence intervals and p-values to 3 decimal places. A p-value of less than 0.05 will be regarded as statistically significant.
A modified intention-to-treat analysis will be performed for all outcomes. Patients will be analysed in the groups to which they are randomly assigned, irrespective of whether the assigned method of preservation is actually used. Practically, this means that if a liver intended by randomisation for machine perfusion for any reason undergoes cold storage, it will be analysed in the assigned machine perfusion group. The analysis will exclude recipients for whom randomisation occurred, but were not transplanted for any reason. In the case of DCD livers in which the retrieval did not proceed, the reason will be documented along with the time waited before stand-down. In all other cases the reason for the liver not proceeding to transplantation will be documented and a narrative analysis of this data performed.
Full details of the proposed statistical analysis will be outlined in a separate document (the Statistical Analysis Plan; SAP).

ADDITIONAL ANALYSES
Analysis will be stratified by recipient centre and donor type (DBD or DCD). Subgroup analyses will be performed for donor type (DCD vs. DBD), by donor risk index (DRI) and by duration of machine preservation in the perfusion arm of the trial. We anticipate that any effect of machine perfusion may be greater in higherrisk donor livers (DCD and higher DRI).
Secondary analyses will also investigate the ability of perfusion parameters and biomarkers measured whilst the liver is perfused on the device to predict clinical outcome (i.e. viability assessment).

MISSING DATA
Withdrawals from the trial after implantation will be documented as per section 6.10, and a narrative analysis of withdrawals will be performed. Recipients withdrawing from the trial after implantation will be included in analysis using imputation methods for missing data, as described in the SAP.
It is anticipated very few patients will be lost to follow-up, but it is likely that not all measurements at all time points will be recorded for every recipient. Assuming that data are missing at random, with no effect of group allocation on the probability of missing data, multiple imputation methods will be used to estimate missing variables. Sensitivity analyses will ensure that these assumptions are correct.

SAMPLE SIZE
Data from 416 liver transplant recipients from University Hospital Essen demonstrate the geometric mean of peak AST to be 608.59 IU/L (the geometric mean is used as peak AST is non-normally distributed). 220 transplants (110 per arm) would have 90% power at 5% significance level to detect a 33% reduction (to 401.67 IU/L) in the geometric mean of peak AST.
2011/12 NHSBT data suggest that 12% of livers retrieved are not transplanted. Assuming losses of 15%, randomisation of 260 livers into the trial will be required to achieve adequate power.
Data from a study of hypothermic machine perfusion of human livers demonstrate an approximate 35% reduction in the peak AST with machine preservation when compared to historical controls undergoing static cold storage [25]. It is expected that normothermic machine perfusion will be at least as effective as hypothermic machine preservation in preventing reperfusion injury. In studies of a porcine model of DBD liver transplantation, normothermic machine perfusion led to a 52% reduction in peak post-transplant AST [27]. In a DCD model, the reduction was 73%.
Given the relationship between peak AST and primary non-function, graft and patient survival described above [19,38], a 33% reduction in peak AST is likely to represent a clinically significant difference in outcome between the study arms.
It is recognized that a proportion of DCD donor livers randomized in the study will not proceed to donation due to the donor not arresting within the time defined by local protocols. These livers will be replaced in the study so that the numbers above reflect the number of livers actually retrieved.

DEVIATION FROM THE STATISTICAL ANALYSIS PLAN
Any deviation from the original statistical analysis plan will require justification in the final study report.

SOURCE DATA
Source documents are where data are first recorded, and from which participants' eCRF data are obtained. These include, but are not limited to, hospital records (from which medical history and previous and concurrent medication may be summarised into the eCRF), clinical and office charts, laboratory reports, pharmacy records, subject diaries or logs, microfiches, radiographs, correspondence, device accountability records, recorded data from automated instruments, copies or transcriptions certified after verification as being accurate and complete, and at medico-technical departments involved in the clinical investigation. eCRF entries will be considered source data if the eCRF is the site of original recording (e.g. there is no other written or electronic record of the data). All documents will be stored safely in confidential conditions. On all trial-specific documents, other than the signed consent, the participant will be referred to by the trial participant number/code, not by name.

DATA RECORDING
Data collection will be achieved using secure internet-based forms. Data will be input by local study investigators/co-ordinators trained in the use of the system prior to receiving log-in details. Data will be uploaded to a central database maintained by the database manager at the Surgical Intervention Trials Unit (SITU).
Demographic data, laboratory results and survival data will be recorded locally on the data collection system. 6-month quality of life data from the EQ-5D-5L and details of resource utilisation will also be recorded.
All blood samples will be analysed in local laboratories and results recorded in common units.
Biopsies will be analysed centrally as part of the COPE bioresource (WP7) and the biopsy scores recorded in the main database.
MRCP scans will be analysed centrally following transfer of digital images to the trial radiologist. Assessment of ischaemic cholangiopathy will be documented via online case report forms.

DATA FORMS AND DATA ENTRY
As described in section 10.1.2, data will be entered onto online forms, which will be transmitted and stored in a database maintained on a central server at the University of Oxford. Validation rules prior to submission will ensure that data are entered in the correct format, within valid ranges and minimise the chance of missing data. Data already entered will be retrievable for viewing through the data entry system. The extent of an individual user's activity in the data entry system will be limited by privileges associated with his/her login and password.
All electronic data will be stored in a secure fashion on password protected central servers, with data identified only by the unique participant study ID. A separate database will be maintained to allow identification of study participants, if required for safety reasons, from their unique ID -this will be kept on a different server to the main database.

DISCREPANCIES AND MISSING DATA
The central database will be monitored for discrepancies and missing data. The surgical intervention trials unit (SITU) will be responsible for managing the database, and if such discrepancies are identified the trial manager will be responsible for identifying the problem and contacting the local centre to ensure resolution. The trial manger will be responsible for the production of weekly reports to each participating centre containing information and details of missing data or missed visits requiring completion.

SECURITY AND BACKUP OF DATA
The database will be accessible only over https so that the connection between the server and client is encrypted. No access will be allowed from any computer on any network (including from within the department) to the database server. The port used by MySQL server will be blocked. The access to databases will be controlled by username and password. When the user logs on to the first time, he/she will have to change the password. The password allowed has to adhere to the following rules - Passwords contain at least eight characters.  Passwords contain both alphabets and numbers.  Passwords contain a mixture of both upper case and lower case characters.  Passwords contain at least one numeric character.  Passwords contain a minimum of five different characters.  The maximum sequential repetition of a character allowed is two.  Passwords are changed at least once a year.
Each centre involved in the trial will have one Super User who will be responsible for assigning user name and passwords to individuals requiring access to the database. It will be responsibility of the Super User to revoke access for the users who are no longer required to use the database. The Database Manager will assign the username and password for the Super User. The Data Manager and the Super Users will regularly monitor the list of users granted access.
Each participating centre will have its own website and access to a website will only be allowed through a fixed set of IP addresses belonging to the institution. This will ensure that only a fixed set of users belonging to a specific institution can access the database.
The study will have a single database, which means that all the data for the study will go into a single database although it can be entered from different centres. This will ensure that the data collected across different centres is consistent and in the same format.
The database will only be accessible through a fixed list of websites. Where an individual requires access to the complete database, it will be possible to provide such individuals links to data dumps within the website.
The source data on the server resides in the database server MySQL which is installed on the drive where the operating system is installed. The data in the database for the trial will be backed up on a separate drive on the server. This will be done automatically once every day. The backed up data will also be encrypted and copied to another drive, the complete contents of which will be backed up to the University managed central back up service.
The backed up data will also be copied to an encrypted external hard drive at regular intervals at least once every month. The external drive will be stored at a different site within the University of Oxford and will always be locked.
Every time changes are made to a web application, the web applications will be backed up. The backed up web applications along with a copy of the active application will also be copied to an encrypted external hard drive at regular intervals along with the databases once every month.
At any time complete backups of the data are available at least in three different places -MySQL server where the source data is stored, data backup folders which are specific to a trial (on a separate drive to where the DBMS runs) and the encrypted external hard drive.

DATA ACCESS
The chief investigator and trial management committee members will have access to the full data sets. Local investigators not on the trial management committee will have access to datasets for the local site only.
Direct access will be granted to authorised representatives from the Sponsor, host institutions and the regulatory authorities to permit trial-related monitoring, audits and inspections.

DATA RETENTION
The trial coordinating centre (SITU) will archive all of the trial master file data in the central database and other relevant records and documents for up to 15 years after completion or discontinuation of the study or for the length of time required by relevant national or local health authorities.

DESCRIPTION OF HARDWARE AND SOFTWARE
The database will be developed using ASP.NET (.NET Framework 4.0/4.5) and the database server used will be MySQL (5.6.12). The database/web application will be hosted on an IIS (8.0) web server. The operating system on the server where the applications installed is Windows Server 2012 and it uses a RAID 6 configuration for replicating data across hard drives. The Data Manager will manage the server and the only other person who has access to the server is the departmental (Nuffield Department of Surgical Sciences) IT officer.

MONITORING
A Trials Co-ordinator, based at the clinical trials unit (SITU), will be responsible for each Investigator's compliance with the trial protocol and will perform source document verification during site visits..
A risk assessment will be carried out and the Monitoring Plan written as determined by the risk assessment.
The investigator and study personnel must set aside a reasonable amount of his / her time for these visits and the time of the relevant site personnel.

QUALITY ASSURANCE AUDITING AND INSPECTION
During the course of the study, the sponsor will appoint quality assurance personnel to provide audit of the administration and conduct of the study. The relevant competent authority could potentially conduct audits / inspections.
The Investigator and the relevant site personnel must set aside a reasonable amount of his / her time for study related monitors, audits and inspection by the authorised representatives of the sponsor, REC, government regulatory authorities, and institution compliance and quality assurance groups, and provide adequate access to all study related documents (e.g. source documents, regulatory documents, data collection instruments, study data etc.).

LOCAL INVESTIGATOR AND SITE PERSONNEL TRAINING
All key site personnel must undergo relevant training in advance of the site initiation in accordance with Good Clinical Practice (GCP) guidelines. Such training will be documented.
In addition, training on the investigational device will be provided in advance of recruitment of the first patient by the central co-ordinating centre. A record of all device training will be maintained. All personnel involved in randomisation and data entry will also be trained in the use of the online randomisation and data collection tool by members of the clinical trials unit, and records of such training will be maintained.

STUDY DOCUMENTATION
It is the responsibility of the local investigator to maintain complete, accurate and current study records. Each investigator will be provided with an investigator site file, online access to the case reporting system and other associated study specific documentation by the co-ordinating centre. Such records will be maintained during the course of the study and for up to 15 years following the date on which the study is terminated or completed, in accordance with local regulatory requirements.

DEFINITIONS
The investigators shall conduct this study in accordance with this protocol and any conditions of approval/notification imposed by the Research Ethics Committee and Competent Authority. Failure to comply with and/or inability to meet these regulations may jeopardize further participation of the investigator or investigative site in this and future clinical studies.
A "protocol deviation" is a failure to adhere to the requirements specified in this study protocol without adequate justification. Examples may include the enrolment of a study patient who does not meet all of the inclusion/exclusion criteria specified in section 5.3, or missed study procedures without documentation.
A "serious breach" is defined as a breach of GCP or the trial protocol which is likely to affect to a significant degree: a) The safety or physical or mental integrity of the subjects of the trial; or b) The scientific value of the trial

REPORTING OF PROTOCOL DEVIATIONS
All protocol deviations must be recorded and reported to the data monitoring committee. The DMC will review all deviations and assess their impact on patient safety. Serious breaches must be reported as per section 12.3.

REPORTING OF SERIOUS BREACHES
In the event that a serious breach is suspected the sponsor must be contacted within 1 working day. In collaboration with the chief investigator and the DMC, the serious breach will be reviewed by the sponsor and, if appropriate, the sponsor will report it to the REC committee, regulatory authority and the host institution within seven calendar days.

DECLARATION OF HELSINKI
The Investigator will ensure that this trial is conducted in accordance with the principles of the current revision of the Declaration of Helsinki on Biomedical Research involving Human Subjects.

ICH GUIDELINES FOR GOOD MEDICAL PRACTICE
The Investigator will ensure that this trial is conducted in full conformity with the relevant regulations and with the ICH Guidelines for Good Clinical Practice (CPMP/ICH/135/95; July 1996)

MEDICAL DEVICE REGULATIONS
The Investigator will ensure that this trial is conducted in full conformity with:

OTHER ETHICAL CONSIDERATIONS
It is possible that, through participation in this trial, incidental findings may be made that are unrelated to a participant's liver disease, transplantation or involvement in the trial, but are of relevance to the their health or wellbeing. In particular, such information may materialise in the findings on the MRCP scan performed at 6 months post-transplant as part of the trial. If this does happen, the patient will be informed of the findings and, with their consent, so too will their GP and other relevant members of their local health team.
Participation in this trial will not affect a patient's position on the liver transplant waiting list or their likelihood or receiving a liver transplant. Similarly, withdrawal of a participant from the trial at any point and for any reason will not affect their position on the liver transplant waiting list or their likelihood or receiving a liver transplant.

LOCAL REGULATORY APPROVAL
This protocol, site-specific informed consent forms and participant education and recruitment materials will be submitted for approval by the appropriate research ethics committee (REC), host institutions and regulatory authorities in each participating region. Before the study can begin, each investigator must have written evidence of favourable ethical opinion from an appropriate REC and approval from an appropriate regulatory authority.
Once approval has been granted, the Investigator is responsible for ensuring that he/she complies with the terms of the approval, namely with adverse event reporting, notification of amendments, interim, annual and final reports on the progress of the study.

PROTOCOL AMENDMENTS
Any change or addition to this study protocol which may impact on the conduct of the study, potential benefit to the patient or may affect patient safety, including changes of study objectives, study design, patient population, sample sizes, study procedures or significant administrative aspects will require a formal written amendment to the study protocol.
Administrative changes to the protocol are minor corrections and/or clarifications that have no effect on the way the study is to be conducted. These administrative changes will be agreed upon by the trial management committee and notified to the relevant RECs and regulatory authorities at the discretion of the trial management committee.
All other amendments to the protocol will be notified to the local regulatory authorities and research ethics committees for approval. Approved amendments will be circulated promptly to all investigators by the coordinating centre. Amendments will be tracked by version number and date in appendix A4 of this document.

REPORTING
The chief investigator shall submit once a year throughout the clinical trial, or on request, an annual progress report to the REC, host organisation and sponsor. In addition, the end of trial notification and final report will be submitted to the MHRA, the REC, host organisation and sponsor.

DONOR AND RECIPIENT CONFIDENTIALITY
All study-related information will be stored securely both at the study sites and the co-ordinating centre. Written information will be stored in locked filing cabinets in areas with limited access. All documentation and specimens will be identified by a unique study ID number to maintain participant confidentiality. Where this is not possible (e.g. informed consent forms), these will be stored separately from any study records identified by the unique study ID.
Participant's information will not be released outside of the study without the written consent of the participant, except as necessary by regulatory authorities or to their local hospital care team.

DECLARATION OF CONFLICTING INTERESTS
Professor Friend is a co-founder, Medical Director and Consultant to OrganOx Limited and also holds shares in the company. However, Professor Friend will not be involved in the selection, recruitment or transplanting of patients in this study.
Professor Coussios is a co-founder, Technical Director and Consultant to OrganOx Limited and also holds shares in the company. However, Professor Coussios will not be involved in the selection, recruitment or transplanting of patients in this study.

EXPENSES AND BENEFITS
Where possible, study visits and investigations will be conducted during routine hospital attendances. Reasonable travel expenses for any visits additional to normal care will be reimbursed on production of receipts, in accordance with the requirements of the Declaration of Helsinki 2008.

INDEMNITY INSURANCE
The University of Oxford has a specialist insurance policy in place, which would operate in the event of any participant suffering harm as a result of their involvement in the research (Newline Underwriting Management Ltd., at Lloyd's of London).
Host institution indemnity operates in respect of the clinical treatment that is provided. In the UK centres, this indemnity is provided by the NHS.

DATA ANALYSIS AND RELEASE OF RESULTS
By conducting the study, the local investigators agree that all information provided by the sponsor and coordinating centre will be maintained by the local investigators and the site personnel in strict confidence. It is understood that the confidential information provided to local investigators will not be disclosed to others without authorization from the sponsor and/or co-ordinating centre.
The scientific integrity of the study requires that all data must be analysed study-wide and reported as such.

PRIMARY OUTCOME PUBLICATIONS
Any publication arising from data collected as part of this study will be subjected to the agreed publication policies of the COPE trial management group. Publications will reflect the input of every centre. Reports relating to primary outcomes will be published in peer-reviewed journals of appropriate relevance. Individual centres will undertake not to report any trial data independently. A final report on the primary outcomes of the study will be compiled by the chief investigator and co-ordinating centre, and approved and signed off by each local investigator.

OTHER STUDY PAPERS, ABSTRACTS AND PRESENTATIONS
Study investigators wishing to publish secondary data analyses are encouraged to submit a proposal to the COPE publication committee for approval. If the publication committee accepts the proposal, then the author of the proposal may decide on the lead in each publication resulting from such a proposal.

A1.1 LOCALLY ANALYSED SPECIMENS
All specimens collected as part of the recipient's routine clinical care, such as postoperative blood samples for routine laboratory analysis, will be analysed and stored locally as per normal local procedure. Results required by the trial protocol will be uploaded to the online data collection system.

A1.2 CENTRALLY ANALYSED SPECIMENS
Study specific samples will be transported to the Oxford Radcliffe bioresource (as part of COPE WP7) for analysis. These samples include:  Perfusate samples from the perfusion device  Plasma, serum and urine samples as specified, other than those required for routine clinical care  Pre-and post-perfusion liver biopsies Provision will be made on the study consent form for storage and later use of these samples for ancillary studies. Shipping of all samples will be tracked with shipping and receipt logs maintained in each site investigators file.

Regulatory aspects
All samples will be collected in accordance with national regulations and requirements including standard operating procedures for logistics and infrastructure. In the UK, samples will be taken in appropriately licensed premises, stored and transported in accordance with the HTA guidelines and local trust policies. Samples for long-term storage will be kept in the Oxford Radcliffe bioresource. The stored tissues will be held under an extension of the University of Oxford's HTA license (12217).

Samples from the donor
In Belgium, Germany and Spain these samples do not require additional donor consent and will be taken during the retrieval process as part of this study for storage in the Oxford Radcliffe bioresource.
In England these samples will only be taken if the donor has been separately consented for participation in the NHSBT QUOD study (see A1.3 below). If consent for participation in the QUOD study has not been provided then these samples will not be taken. This will not preclude the donor liver from being used in this trial.

Blood samples
At each time-point where blood is collected  1x EDTA 6 ml separator tube will be obtained  1x Serum 6ml separator tube will be obtained To ensure minimal sample degradation and pre analytical variability, whole blood should be kept at room temperature prior to separation of plasma from cellular parts. Separation of cells from plasma and serum should be achieved by centrifugation at 1500g for 10 min at room temperature as close as possible to blood collection. Centrifugation will not be achievable in some scenarios e.g. during transportation etc. Standard Ordered Procedures will reflect practical time points for the handling and processing of samples. After centrifugation plasma and serum samples should kept at 4 o C.

Perfusate samples
Perfusate samples will be collected during normothermic machine perfusion. Two 6ml samples (EDTA and Serum) will be taken at all time-points and managed as per the blood samples above.

Biopsies
Biopsy segments will be divided into two and stored in pre-prepared formalin or RNAlater tubes. Samples stored in RNAlater will be frozen in liquid nitrogen and stored at -80 o C for long-term storage.

Consumables, tracking and tracing and logistics
Local sites will be provided with all of the required blood tubes, the required Standard Operating Procedures and the tracking and tracing software/hardware and programmes.

A1.3 THE QUOD STUDY
The Quality in Organ Donation (QUOD) study is an initiative set up by the University of Oxford in collaboration with NHSBT and National Organ Retrieval Service teams with the aim of identifying pathways of injury and repair to donor organs. In appropriately consented donors, blood, urine and tissue samples will be collected during the retrieval process. Samples will only be procured in hospitals with the appropriate HTA licence in place. Infrastructure to transport these samples back to the Oxford Radcliffe bioresource where they are stored for use in current and future research projects is in place. The aim of this bioresource is to support research into increasing our understanding of injury and repair mechanisms to donor organs and improving the quality of donor organs. The project started in Scotland in April 2013 and will commence in England in February 2014 (REC approval number: 13/NW/0017).