Accepted Manuscript. Evolving Trends in Machine Perfusion for Liver Transplantation

Transcription

1 Accepted Manuscript Evolving Trends in Machine Perfusion for Liver Transplantation P. Dutkowski, J.V. Guarrera, J. de Jonge, P.N. Martins, R.J. Porte, P.A. Clavien PII: S (19) DOI: Reference: YGAST To appear in: Gastroenterology Accepted Date: 20 December 2018 Please cite this article as: Dutkowski P, Guarrera J, de Jonge J, Martins P, Porte R, Clavien P, Evolving Trends in Machine Perfusion for Liver Transplantation, Gastroenterology (2019), doi: doi.org/ /j.gastro This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

2 Evolving Trends in Machine Perfusion for Liver Transplantation P Dutkowski 1, JV Guarrera 2, J de Jonge 3, PN Martins 4, RJ Porte 5, PA Clavien 1 1 : Department of Surgery & Transplantation, University Hospital Zurich, Switzerland 2 : Division of Liver Transplant and HPB Surgery, Rutgers New Jersey Medical School, University Hospital, Newark, NJ, USA 3 : Department of Surgery Erasmus University Medical Center Rotterdam, The Netherlands 4 : Department of Surgery, Transplant Division, University of Massachusetts, UMass Memorial Medical Center, Worcester, USA 5 : HPB and Liver Transplant Surgery, University of Groningen, University Medical Center Groningen, The Netherlands Word count: 2137 (max 2400) Tables: 1 Figures: 1 References: 51 The authors declare the following conflicts of interest: PD, JdJ, RP, and PAC are involved in a European RCT of Hypothermic Oxygenated PErfusion (HOPE) in DBD liver transplant recipients JdJ and RP are involved in a European RCT of dual Hypothermic Oxygenated PErfusion (D- HOPE) in DCD liver transplant recipients JVG is involved in Research Support and Consulting, Lifeline Scientific-Organ Recovery Systems PAC and PD have filed patents on long term ex situ perfusion technology along with ETH (Eidgenössische Technische Hochschule) Corresponding author: P.-A. Clavien Department of Surgery & Transplantation, University Hospital Zurich Raemistrasse 100, CH-8091 Zurich Phone: Fax: clavien@access.uzh.ch 1

3 List of Abbreviations NRP: normothermic regional perfusion NMP: normothermic machine perfusion HOPE: hypothermic oxygenated perfusion D-HOPE: dual hypothermic oxygenated perfusion DBD: donation after brain death DCD: donation after circulatory death LT: liver transplantation AST: aspartate aminotransferase ALT: alanine aminotransferase ADP: adenosine diphosphate ATP: adenosine triphosphate RET: reverse electron transfer RCT: randomized clinical trial INR: international normalized ratio CCI: Comprehensive complication index IFOT: ischemia free organ transplantation ROS: reactive oxygen species NADH: nicotine adenine dehydrogenase DAMPs: Danger associated molecular patterns 2

4 The imbalance between available grafts for transplantation and demands has moved the focus of many investigators on the search for novel strategies to rescue organs, previously considered to be unsuitable for transplantation. In this setting, machine perfusion is recognized as one of the most significant improvements in the field of transplantation over the past 20 years (1). Besides potentially improving organ shortage by repairing putative irreversible injuries, dynamic preservation strategies may offer an opportunity to test organ quality before implantation or to manipulate some functions; for example, by mitigating the immune response (2). For livers, two main perfusion approaches are currently debated in the clinic, a) perfusion with blood or alternative oxygen carriers at physiologic normothermic or subnormothermic conditions, or b) perfusion with cooled oxygenated artificial fluids. The aim and mechanisms of both approaches differ greatly. Normothermic machine perfusion (NMP) is used to minimize the duration of cold storage, and is applied either in situ, e.g. in donors before procurement, known as normothermic regional perfusion (NRP)(3), or ex situ during or after organ transport to the recipient center (4). The first randomized NMP trial on predominantly livers donated after brain death (DBD) showed excellent 1-year graft survival (95%) by continuous application of NMP during the entire preservation period with, however, similar survival outcome in the control static preservation group (5). Hypothermic machine perfusion(6)(7) and hypothermic oxygenated perfusion (HOPE and D- HOPE)(8)(9)(10) are currently performed after cold storage, i.e., just prior to implantation in a recipient. Recent observational studies reported a 5-year graft survival of 94 % in HOPE treated livers despite using higher risk livers with longer donor warm ischemia(11). Besides mortality, however, morbidity measured up to one year after transplantation appears equally important (12). For example, a particularly distressful type of preservation injury occurs on the bile ducts, often weeks after transplantation, and this risk is mainly observed in livers donated after circulatory death (DCD)(13). Of note, both, upfront NMP or NMP after cold storage, e.g. an end-ischemic normothermic perfusion, failed to protect significantly DCD liver grafts from such biliary injury (5)(14). In contrast, end-ischemic HOPE has been shown to decrease significantly histological signs of bile duct injury after reperfusion (10), and achieved a rate of less than 5 % non-anastomotic strictures in recent reports on DCD livers (15). These results, however, should be interpreted with caution given the observational study design. 3

5 An important point, still poorly investigated, is whether machine perfusion techniques have any benefits in low-risk liver grafts, which are associated with excellent short and long-term outcome using simple static preservation approaches. Such grafts include DBD livers without relevant steatosis or DCD livers with short warm and cold ischemia times (16)(17)(13). For example, transplantation with cold stored livers not associated with steatosis, fibrosis, or prolonged warm ischemic time achieved excellent 92 % one-year graft survival (12). Correspondingly, the control group in the recent NMP trial showed a 96 % one-year graft survival by simple cold storage (5) indicating that the trial was performed using mostly high quality grafts and low risk recipients. This issue might be also important in the ongoing HOPE trial, where hypothermic machine perfusion targets exclusively DBD livers with no specific restriction on high-risk grafts (18). Therefore, a true benefit of machine perfusion in this trial would be linked to the inclusion of a relative high percentage of injured livers. Despite the overwhelming enthusiasm for machine perfusion technologies (19)(20)(21), the routine replacement of cold storage by dynamic perfusion approaches is difficult to justify and certainly not cost efficient. A relevant benefit is to be expected only in conditions with increased risk of graft loss, such as advanced donor age, macrovesicular steatosis, and longer donor warm or cold ischemia. The current unsolved issue is at what threshold of risks we should use those novel strategies. There is a clear lack of data and consensus in this area, and this should be a high priority for added knowledge. Current randomized machine perfusion trials Do they help? No doubt that RCTs are needed to define the role of machine perfusion in liver 4 transplantation. Such trials must secure not only safety and benefits, but also costeffectiveness, as some perfusion strategies are associated with exorbitant costs. Therefore, investigators must put particular attention in the study design to provide convincing information through sufficient sample sizes and clinically relevant endpoints. Optimally, to justify a change in organ preservation policies, novel approaches should demonstrate significant improvement in allograft function, liver utilization rates, outcome of recipients including ICU and hospital stays, graft and patient survivals, and at least one-year posttransplant morbidity (12). In contrast, the use of surrogate markers of liver injury, such as

6 the release of transaminases (AST/ALT) must be strictly avoided as a primary endpoint, since correlation between AST or ALT release and outcome of a graft after transplantation is inconsistent, especially in DCD liver grafts. Serum transaminase levels after transplantation may confer some predictive value regarding graft survival only when markedly elevated, e.g. > 5,000 U/L) (22). A recent large trial on NMP, unfortunately, chose peak serum AST levels as the primary endpoint (5), which not only poorly reflect whole graft injury, but these markers are washed out together with other cytoplasmic enzymes already during ex situ machine liver perfusion prior to implantation of the graft in the recipient. This is an obvious bias, as wash out of any liver enzymes is not part of cold stored grafts, e.g. in the control group. Surprisingly, the authors failed to report AST concentrations in the perfusate (5). Second, availability of more sensitive markers of allograft function would be essential to demonstrate protective effects. Unfortunately, no such markers exist, and the most accurate clinical prediction for liver graft dysfunction following transplantation relies on lactate and bilirubin clearance together with the synthesis of coagulation factors. Importantly, dynamic changes of these laboratory parameters (e.g. the slope of serum INR or bilirubin) are superior to any single peak values (23). Accordingly, a definition of liver graft dysfunction based on peak transaminases in combination with later (e.g., 7 day) single levels of INR and bilirubin at arbitrary cut-off points (24) must be avoided in perfusion trials. The only completed RCT (5), regrettably, focussed on such definition, and thereby failed to offer any valuable information about the role of NMP on liver graft function. Third, rescuing damaged organs, otherwise discarded, is a major target of new technologies in this area. The policy to discard a graft for transplantation is, however, far from being standardized, greatly differs among centres and individual surgeons, and is prone to major selection biases in clinical trials (25). For example, in the recent perfusion RCT (5), a quarter of accepted livers were discarded in the cold storage group by the respective recipient centres, although initially randomized because deemed transplantable by the donor surgeon. This figure contrasts with only 12% in the perfusion group (5). Such discrepancy between accepting grafts for the trial, but not for implantation suggests an additional flaw in the study design. Future trials require a strict commitment in the decision to use or not a liver graft before randomization, regardless if perfusion or cold storage is 5

7 used, to minimize differences in drop outs. In addition, the quality of all grafts need to be documented in terms of donor age, steatosis, cold and warm ischemia times to increase comparability of discard reasons among centers (15). Fourth, hospital stay after liver transplantation varies greatly among centres despite adjusting risk, as illustrated in a recent benchmark analysis in large volume centers recording a range of hospitalization between 6 and 24 days in low risk transplants (12). The reason behind such considerable differences in hospital stay is mostly not treatment based, but reflects discharge policies of transplant centers, e.g. in-hospital waiting for rehabilitation places vs. unrestricted transfer without rehabilitation. This indicates major limitations in interpreting such parameter in studies. Fifth, graft and patient survivals are logically convincing parameters, but there is a need for adequate follow-up to assess safety or superiority in machine perfusion trials. A minimum observation of 1-year appears advisable, based on the fact that most graft losses related to preservation injury develop all along the first year after LT (1)(12). Finally, measuring morbidity after transplantation is crucial, as it affects quality of life of patients and influences overall cost. (26). Capturing complications after surgery, however, is challenging since multiple definitions of postoperative complications exist. The Clavien- Dindo classification, which relies on the need for treatment to correct a negative event, or its surrogate metric, the comprehensive complication index (CCI )(27), are currently widely used in surgery and transplantation (28) with the availability of reference values provided in a recent multicenter benchmark study covering one year after transplantation (12). In summary, most of the available endpoints in LT are weak and should be carefully selected in RCT s. Based on the fact that complications after liver transplant are frequent and have unequivocally the highest clinical relevance in terms of cost and quality of life, we suggest a cumulative assessment of post-transplant complications within one year, besides recipient and graft survival, as preferred primary endpoints for future machine perfusion studies. This is the case for example in the two ongoing multicentric randomized machine perfusion trials on hypothermic liver perfusion (HOPE, D-HOPE)(18). For convincing effectiveness against cold static preservation, any novel perfusion technology should prove superiority in relevant endpoints before claiming its general use (Table 1). 6

8 7 Future randomized trials Future machine perfusion trials should compare the different competing techniques using clinically relevant endpoints. To properly conduct such studies, an objective and adequate knowledge of the various perfusion technologies is mandatory, and thus should optimally be restricted to few experienced centres. We would urge the transplant community including the industry to rather focus on supporting current research in highly engaged perfusion teams, instead of suggesting widespread use of one specific perfusion technology. Accessible registries to allow research in long-term outcome should also be implemented(29). Mechanism of different machine perfusion techniques what is currently known? The precise underlying mechanism for protection of liver grafts by machine perfusion remains under debate. For NMP, the current explanation is a reduction of cold ischemia by minimizing the anaerobic metabolism with less accumulation of citric acid cycle products and improved ATP restoration (30). The extreme of such strategy is an ischemia free organ transplantation (IFOT), where donor livers are put on a normothermic circuit even before procurement, similar as in NRP, but, however, with continuing nomothermic perfusion until implantation (31). It might be more efficient to develop repair strategies of pre-transplant or procurement injuries, which are common in the clinic (e.g. fatty liver, prolonged cold or warm ischemia). Endischemic NMP in discarded human livers with longer periods of cold ischemia failed to protect against biliary injury; the most prevalent cause of graft loss after DCD liver transplantation (14). Continuous or endischemic NMP after ischemia triggers an inflammatory cascade, mostly related to mitochondrial derived reactive oxygen species (ROS) (32)(33) (Figure 1), which resemble in vivo conditions (34). Evidence for an activation of danger signals by normothermic perfusion techniques is also well described in normothermic perfusion of other organs including heart- (35), kidney- (36), and lung (37). The target of all ex situ normothermic interventions is, therefore, to capture ROS and danger molecules (DAMPs and pro-inflammatory cytokines), for example by perfusate scavengers (33) or by implementation of cytokine filters into the circuit. In contrast, hypothermic perfusion techniques mitigate oxidative stress despite high availability of oxygen in the cold (38), probably due to changes in mitochondrial electron transfer, e.g.

9 down-regulated electron transfer at temperatures below the Arrhenius break point temperature of 15 C. At the same time, metabolism of accumulated electron donors, such as NADH and succinate, leads to ATP re-synthesis (39)(40) with surprisingly high efficiency, compared to normothermic conditions, possibly due to less electron and proton leakage in the cold (38). End-ischemic cold oxygenated perfusion therefore allows a higher upload of cellular energetic levels in liver grafts procured after circulatory death (DCD), compared to normothermic perfusion(41) (Figure 1). These observations support possible complementary effects of, for example, combining first hypothermic and afterwards normothermic machine perfusion, when applied after a period of cold ischemic storage(42)(43). Future outlook Future investigations on perfusion should include the search of protective mechanisms, and evaluate effective repair of injured organs. Thereby, perfusion technologies convey a great potential to influence cellular metabolism before implantation, modulating inflammation, immune response and perhaps triggering anticancer pathways. It is central that expensive and cumbersome RCTs in the field of organ perfusion select exclusively clinically relevant endpoints. Additionally, prediction of organ function during machine perfusion will likewise be possible by detailed perfusate analysis including metabolomics and proteomics, leading to safer utilization of marginal grafts. We anticipate that within the next few years a far better understanding will be available about which perfusion strategies should be applied best and for which liver graft. 8

10 References 1. Dutkowski P, Linecker M, Deoliveira ML, Müllhaupt B, Clavien PA. Challenges to liver transplantation and strategies to improve outcomes. Gastroenterology. 2015; 2. Clavien PA, Muller X, de Oliveira ML, Dutkowski P, Sanchez-Fueyo A. Can immunosuppression be stopped after liver transplantation? The Lancet Gastroenterology and Hepatology Oniscu GC, Randle L V., Muiesan P, Butler AJ, Currie IS, Perera MTPR, et al. In situ normothermic regional perfusion for controlled donation after circulatory death - The United Kingdom experience. Am J Transplant. 2014; 4. Ravikumar R, Jassem W, Mergental H, Heaton N, Mirza D, Perera MTPR, et al. Liver Transplantation After Ex Vivo Normothermic Machine Preservation: A Phase 1 (Firstin-Man) Clinical Trial. Am J Transplant. 2016; 5. Nasralla D, Coussios CC, Mergental H, Akhtar MZ, Butler AJ, Ceresa CDL, et al. A randomized trial of normothermic preservation in liver transplantation. Nature. 2018; 6. Guarrera J V., Henry SD, Samstein B, Odeh-Ramadan R, Kinkhabwala M, Goldstein MJ, et al. Hypothermic machine preservation in human liver transplantation: The first clinical series. Am J Transplant. 2010; 7. Guarrera J V., Henry SD, Samstein B, Reznik E, Musat C, Lukose TI, et al. Hypothermic machine preservation facilitates successful transplantation of orphan extended criteria donor livers. Am J Transplant. 2015; 8. Dutkowski P, Schlegel A, De Oliveira M, Müllhaupt B, Neff F, Clavien PA. HOPE for human liver grafts obtained from donors after cardiac death. J Hepatol. 2014;60(4): van Rijn R, Karimian N, Matton APM, Burlage LC, Westerkamp AC, van den Berg AP, et al. Dual hypothermic oxygenated machine perfusion in liver transplants donated after circulatory death. Br J Surg. 2017; 10. van Rijn R, van Leeuwen O, Matton A, Burlage L, Wiersema-Buist J, van den Heuvel M, et al. Hypothermic oxygenated machine perfusion reduces bile duct reperfusion injury after transplantation of donation after circulatory death livers. Liver Transplant. 2018;(Jan 25.). 11. Schlegel AA, Muller X, Kalisvaart M, Muellhaupt B, Perera M, Isaac J, et al. Outcomes of liver transplantations from donation after circulatory death (DCD) treated by hypothermic oxygenated perfusion (HOPE) before implantation. J Hepatol. 2018; Muller X, Marcon F, Sapisochin G, Marquez M, Dondero F, Rayar M, et al. Defining Benchmarks in Liver Transplantation: A Multicenter Outcome Analysis Determining Best Achievable Results. Ann Surg. 2018; 13. Laing RW, Scalera I, Isaac J, Mergental H, Mirza DF, Hodson J, et al. Liver transplantation using grafts from donors after circulatory death: A propensitymatched study from a single centre. Am J Transplant. 2016;(5):n/a-n/a. 14. Watson CJE, Kosmoliaptsis V, Pley C, Randle L, Fear C, Crick K, et al. Observations on

11 10 the ex situ perfusion of livers for transplantation. Am J Transplant. 2018; 15. Muller X, Schlegel A, Würdinger M, Wendt M, Kron P, Eshmuminov D, et al. Hypothermic oxygenated perfusion (HOPE) for futile DCD liver grafts. HPB. 2018;in press. 16. DeOliveira ML, Jassem W, Valente R, Khorsandi SE, Santori G, Prachalias A, et al. Biliary complications after liver transplantation using grafts from donors after cardiac death: results from a matched control study in a single large volume center. Ann Surg. 2011;254(5):716-22; discussion Schlegel A, Kalisvaart M, Scalera I, Laing R, Mergental H, Mirza D, et al. The UK DCD Risk Score: A new proposal to define futility in donation-after-circulatory-death liver transplantation. J Hepatol. 2018;Mar(68(3)): Jochmans I, Akhtar MZ, Nasralla D, Kocabayoglu P, Boffa C, Kaisar M, et al. Past, Present, and Future of Dynamic Kidney and Liver Preservation and Resuscitation. Am J Transplant. 2016; 19. Ceresa CDL, Nasralla D, Jassem W. Normothermic Machine Preservation of the Liver: State of the Art. Curr Transplant Reports. 2018; 20. Schneeberger S. Life of a liver awaiting transplantation. Nature. 2018; 21. Laing RW, Mergental H, Mirza DF. Normothermic ex-situ liver preservation: The new gold standard. Current Opinion in Organ Transplantation Rosen HR, Martin P, Goss J, Donovan J, Melinek J, Rudich S, et al. Significance of early aminotransferase elevation after liver transplantation. Transplantation. 1998; 23. Agopian VG, Harlander-Locke MP, Markovic D, Dumronggittigule W, Xia V, Kaldas FM, et al. Evaluation of Early Allograft Function Using the Liver Graft Assessment Following Transplantation Risk Score Model. JAMA Surg. 2017; 24. Olthoff KM, Kulik L, Samstein B, Kaminski M, Abecassis M, Emond J, et al. Validation of a current definition of early allograft dysfunction in liver transplant recipients and analysis of risk factors. Liver Transplant. 2010; 25. Marcon F, Schlegel A, Bartlett DC, Kalisvaart M, Bishop D, Mergental H, et al. Utilization of Declined Liver Grafts Yields Comparable Transplant Outcomes and Previous Decline Should Not Be a Deterrent to Graft Use. Transplantation. 2018; 26. Vonlanthen R, Slankamenac K, Breitenstein S, Puhan MA, Muller MK, Hahnloser D, et al. The impact of complications on costs of major surgical procedures: A cost analysis of 1200 patients. Ann Surg. 2011; 27. Slankamenac K, Graf R, Barkun J, Puhan M a, Clavien P-A. The comprehensive complication index: a novel continuous scale to measure surgical morbidity. Ann Surg. 2013;258(1): Dindo D, Demartines N, Clavien P-A. Classification of Surgical Complications. Ann Surg. 2004;240(2): Quintini C, Martins PN, Shah S, Killackey M, Reed A, Guarrera J.V., et al. Implementing an innovated preservation technology: The American Society of Transplant Surgeons

12 (ASTS) Standards Committee White Paper on Ex Situ Liver Machine Perfusion. Am J Transplant. 2018;18: Ceresa CDL, Nasralla D, Coussios CC, Friend PJ. The case for normothermic machine perfusion in liver transplantation. Liver Transplant. 2018; 31. He X, Guo Z, Zhao Q, Ju W, Wang D, Wu L, et al. The first case of ischemia-free organ transplantation in humans: A proof of concept. Am J Transplant. 2018; 32. Chouchani ET, Pell VR, Gaude E, Aksentijević D, Sundier SY, Robb EL, et al. Ischaemic accumulation of succinate controls reperfusion injury through mitochondrial ROS. Nature. 2014; 33. Murphy MP. How mitochondria produce reactive oxygen species. Biochem J. 2009; 34. Van Golen RF, Van Gulik TM, Heger M. Mechanistic overview of reactive speciesinduced degradation of the endothelial glycocalyx during hepatic ischemia/reperfusion injury. Free Radical Biology and Medicine Kearns MJ, Miller SD, Cheung A, Bashir J, Wong S, Seidman MA, et al. A Rodent Model of Cardiac Donation after Circulatory Death and Novel Biomarkers of Cardiac Viability during Ex Vivo Heart Perfusion. Transplantation. 2017; 36. Hosgood SA, Moore T, Kleverlaan T, Adams T, Nicholson ML. Haemoadsorption reduces the inflammatory response and improves blood flow during ex vivo renal perfusion in an experimental model. J Transl Med. 2017; 37. Sadaria MR, Smith PD, Fullerton DA, Justison GA, Lee JH, Puskas F, et al. Cytokine Expression Profile in Human Lungs Undergoing Normothermic Ex-Vivo Lung Perfusion. Ann Thorac Surg. 2011; 38. Dufour S, Rousse N, Canioni P, Diolez P. Top-down control analysis of temperature effect on oxidative phosphorylation. Biochem J. 1996; 39. Westerkamp AC, Mahboub P, Meyer SL, Hottenrott M, Ottens PJ, Wiersema-Buist J, et al. End-ischemic machine perfusion reduces bile duct injury in donation after circulatory death rat donor livers independent of the machine perfusion temperature. Liver Transplant. 2015; 40. Op Den Dries S, Westerkamp AC, Karimian N, Gouw ASH, Bruinsma BG, Markmann JF, et al. Injury to peribiliary glands and vascular plexus before liver transplantation predicts formation of non-anastomotic biliary strictures. J Hepatol. 2014;60(6): Schlegel A, Kron P, Graf R, Dutkowski P, Clavien PA. Warm vs. cold perfusion techniques to rescue rodent liver grafts. J Hepatol. 2014; 42. Burlage LC, Karimian N, Westerkamp AC, Visser N, Matton APM, van Rijn R, et al. Oxygenated hypothermic machine perfusion after static cold storage improves endothelial function of extended criteria donor livers. HPB. 2017; 43. Boteon YL, Laing RW, Schlegel A, Wallace L, Smith A, Attard J, et al. Combined Hypothermic and Normothermic Machine Perfusion Improves Functional Recovery of Extended Criteria Donor Livers. Liver Transplant. 2018; 11

13 44. Bral M, Gala-Lopez B, Bigam D, Kneteman N, Malcolm A, Livingstone S, et al. Preliminary Single-Center Canadian Experience of Human Normothermic Ex Vivo Liver Perfusion: Results of a Clinical Trial. Am J Transplant. 2017; 45. De Carlis R, Di Sandro S, Lauterio A, Ferla F, Dell Acqua A, Zanierato M, et al. Successful donation after cardiac death liver transplants with prolonged warm ischemia time using normothermic regional perfusion. Liver Transplant. 2017; 46. Dutkowski P, Polak WG, Muiesan P, Schlegel A, Verhoeven CJ, Scalera I, et al. First Comparison of Hypothermic Oxygenated PErfusion Versus Static Cold Storage of Human Donation After Cardiac Death Liver Transplants: An International-matched Case Analysis. Ann Surg. 2015;262(5): Guarrera J V., Henry SD, Samstein B, Reznik E, Musat C, Lukose TI, et al. Hypothermic machine preservation facilitates successful transplantation of orphan extended criteria donor livers. Am J Transplant. 2015;15(1): Hessheimer A, Coll E, Valdivieso A, Gómez M, Santoyo J, Ramírez P, et al. Superior outcomes using normothermic regional perfusion in cdcd liver transplantation. ILTS Conf Lisbon Port. 2018; 49. Watson C, Hunt F, Butler A, Sutherland A, Upponi S, Currie I, et al. Normothermic regional perfusion (NRP) for DCD liver transplantation in the UK: better graft survival with no cholangiopathy. ILTS Conf Lisbon Port. 2018; 50. Czigany Z, Lurje I, Tolba RH, Neumann UP, Tacke F, Lurje G. Machine perfusion for liver transplantation in the era of marginal organs New kids on the block. Liver International. 2018; 51. Selten J, Schlegel A, de Jonge J, Dutkowski P. Hypo- and normothermic perfusion of the liver: Which way to go? Best Practice and Research: Clinical Gastroenterology Author names in bold designate shared co-first authorship 12

14 Table 1: Current observational studies and randomized trials (RCT) on human machine liver perfusion Completed studies Year RCT Graft type Technique Endpoints Clinical relevance Bral et al(44) 2017 no DBD+DCD NMP (Organox) Graft function and survival yes De Carlis et al(45) 2018 no DCD NRP+HOPE Graft function, biliary complications, 1y survival yes Dutkowski et al(46) 2015 no DCD HOPE Graft function, 1y graft survival, biliary complications yes Guarrera et al(6) 2010 no DBD HMP PNF, EAD; 1y graft and patient survival, biliary yes complications Guarrera et al(47) 2015 no DBD HMP PNF, EAD; 1y graft and patient survival, biliary yes complications Hessheimer et al(48) 2018 no DCD NRP Graft function, biliary complications, 1 y graft survival yes Nasralla et al(5) 2018 yes DBD+DCD NMP (Organox) Peak AST, EAD, Graft utilization no Ravikumar et al(4) 2016 no DBD+DCD NMP Peak AST, EAD no Liver Revive trial 2016 yes DCD NMP (OCS) Number of donor livers preserved in a near physiological no (18) state, serious adverse events Schlegel et al(11) 2018 no DCD HOPE 5 year graft survival, biliary complications yes Van Rijn et al(9) 2017 no DCD D-HOPE Histology of bile ducts during implantation yes Van Rijn et al(10) 2018 no DCD D-HOPE Liver function, liver ATP, biliary complications, yes 1 y graft and patient survival Watson et al(49) 2018 no DCD NRP Peak ALT, graft function, 90 day survival no Watson et al(14) 2018 no DBD+DCD NMP (Liver Assist) Post reperfusion syndrome, PNF, bile duct histology, biliary complications, graft survival yes Ongoing RCTs Dutkowski et al(18) ongoing yes DBD HOPE 1-year cumulative Clavien grade >=III, CCI * Porte et al(18) ongoing yes DCD D-HOPE 6 month biliary complications * Lurje et al (50) ongoing yes ECD DBD HOPE Peak ALT no Liver Protect trial (18) ongoing yes DBD+ DCD NMP (OCS) EAD, serious adverse events no * clinical relevance not yet available 13

15 Figure 1: Machine liver perfusion strategies Figure legend: A. During cold or warm ischemia, liver grafts suffer from lack of oxygen supply, leading to major metabolic alterations at the mitochondrial level(32). First, electron carriers such as NADH 1 are highly reduced, and second, due to a block in the electron transport chain (32), the citric acid metabolite succinate increases 2, while ATP and ADP decrease 3, and purine metabolites increase 4 (MODE 1 of mitochondrial operation according to Murphy)(33). B. Upon normothermic reperfusion, rapid oxidation of the accumulated succinate 1 together with low adenine nucleotide levels 2and a reduced Coenzyme Q pool trigger reverse instead of forward electron transfer (RET) 3 through the mitochondrial respiratory chain (32), which has been recognized as major generator of reactive oxygen species (ROS) at complex I 4(33). Increased ROS release can also occur at the flavin mononucleotide site of complex I by an increased NADH/NAD ratio. Subsequently, ROS release will initiate an inflammatory cascade (34) during normothermic machine perfusion (MODE 1 or MODE 2 of mitochondrial operation according to Murphy (33). C. In contrast, during hypothermic/subnormothermic perfusion, mitochondrial respiratory function is significantly slowed down(51), preventing accumulation of electron overload at complex I and II. At the same time, ATP and ADP are 14

16 resynthesized by purine salvage pathway(38)1, leading to an efficient energy charge reload together with succinate metabolism within 1-2 hours (41). Mitochondria are therefore prepared to function normal (forward electron flow 2) by hypothermic/subnormothermic perfusion with oxidized NADH pool and sufficient ATP/ADP before implantation (MODE 3 of mitochondrial operation according to Murphy (33). 15

17 Table 1: Current observational studies and randomized trials (RCT) on human machine liver perfusion Completed studies Year RCT Graft type Technique Endpoints Clinical relevance Bral et al(44) 2017 no DBD+DCD NMP (Organox) Graft function and survival yes De Carlis et al(45) 2018 no DCD NRP+HOPE Graft function, biliary complications, 1y survival yes Dutkowski et al(46) 2015 no DCD HOPE Graft function, 1y graft survival, biliary complications yes Guarrera et al(6) 2010 no DBD HMP PNF, EAD; 1y graft and patient survival, biliary yes complications Guarrera et al(47) 2015 no DBD HMP PNF, EAD; 1y graft and patient survival, biliary yes complications Hessheimer et al(48) 2018 no DCD NRP Graft function, biliary complications, 1 y graft survival yes Nasralla et al(5) 2018 yes DBD+DCD NMP (Organox) Peak AST, EAD, Graft utilization no Ravikumar et al(4) 2016 no DBD+DCD NMP Peak AST, EAD no Liver Revive trial 2016 yes DCD NMP (OCS) Number of donor livers preserved in a near physiological no (18) state, serious adverse events Schlegel et al(11) 2018 no DCD HOPE 5 year graft survival, biliary complications yes Van Rijn et al(9) 2017 no DCD D-HOPE Histology of bile ducts during implantation yes Van Rijn et al(10) 2018 no DCD D-HOPE Liver function, liver ATP, biliary complications, yes 1 y graft and patient survival Watson et al(49) 2018 no DCD NRP Peak ALT, graft function, 90 day survival no Watson et al(14) 2018 no DBD+DCD NMP (Liver Assist) Post reperfusion syndrome, PNF, bile duct histology, biliary complications, graft survival yes Ongoing RCTs Dutkowski et al(18) ongoing yes DBD HOPE 1-year cumulative Clavien grade >=III, CCI * Porte et al(18) ongoing yes DCD D-HOPE 6 month biliary complications * Lurje et al (50) ongoing yes ECD DBD HOPE Peak ALT no Liver Protect trial (18) ongoing yes DBD+ DCD NMP (OCS) EAD, serious adverse events no * clinical relevance not yet available 1

18