Cornelia J. Verhoeven

Biomarkers to assess graft quality during conventional and machine preservation in liver transplantation

A global rising organ shortage necessitates the use of extended criteria donors (ECD) for liver transplantation (LT). However, poor preservation and extensive ischemic injury of ECD grafts have been recognized as important factors associated with primary non-function, early allograft dysfunction, and biliary complications after LT. In order to prevent for these ischemia-related complications, machine perfusion (MP) has gained interest as a technique to optimize preservation of grafts and to provide the opportunity to assess graft quality by screening for extensive ischemic injury. For this purpose, however, objective surrogate biomarkers are required which can be easily determined at time of graft preservation and the various techniques of MP. This review provides an overview and evaluation of biomarkers that have been investigated for the assessment of graft quality and viability testing during different types of MP. Moreover, studies regarding conventional graft preservation by static cold storage (SCS) were screened to identify biomarkers that correlated with either allograft dysfunction or biliary complications after LT and which could potentially be applied as predictive markers during MP. The pros and cons of the different biomaterials that are available for biomarker research during graft preservation are discussed, accompanied with suggestions for future research. Though many studies are currently still in the experimental setting or of low evidence level due to small cohort sizes, the biomarkers presented in this review provide a useful handle to monitor recovery of ECD grafts during clinical MP in the near future.

MicroRNA profiles in graft preservation solution are predictive of ischemic-type biliary lesions after liver transplantation

BACKGROUND & AIMS Ischemic-type biliary lesions (ITBL) are the second most common cause of graft loss after liver transplantation. Though the exact pathophysiology of ITBL is unknown, bile duct injury during graft preservation is considered to be a major cause. Here we investigated whether the release of cholangiocyte-derived microRNAs (CDmiRs) during graft preservation is predictive of the development of ITBL after liver transplantation. METHODS Graft preservation solutions (perfusates) and paired liver biopsies collected at the end of cold ischemia were analysed by RT-qPCR for CDmiR-30e, CDmiR-222, and CDmiR-296 and hepatocyte-derived miRNAs (HDmiRs) HDmiR-122 and HDmiR-148a. MicroRNAs in perfusates were evaluated on their stability by incubation and fractionation experiments. MicroRNA profiles in perfusates from grafts that developed ITBL (n=20) and grafts without biliary strictures (n=37) were compared. RESULTS MicroRNAs in perfusates were proven to be stable and protected against degradation by interacting proteins. Ratios between HDmiRs/CDmiRs were significantly higher in perfusates obtained from grafts that developed ITBL (p<0.01) and were identified as an independent risk factor by multivariate analysis (p<0.01, HR: 6.89). The discriminative power of HDmiRs/CDmiRs in perfusates was validated by analysis of separate brain death- (DBD) and cardiac death donors (DBD; p ≤ 0.016) and was superior to expression in liver biopsies (C=0.77 in perfusates vs. C<0.50 in biopsies). CONCLUSIONS This study demonstrates that differential release of CDmiRs during graft preservation is predictive of the development of ITBL after liver transplantation. This provides new evidence for the link between graft-related bile duct injury and the risk for later development of ITBL.

The ins and outs of microRNAs as biomarkers in liver disease and transplantation

Ongoing research is being conducted in the field of transplantation to discover novel, noninvasive biomarkers for assessment of graft quality before transplantation and monitoring of graft injury after transplantation. MicroRNAs (miRNAs) are among the most promising in this field. MiRNAs are small noncoding RNAs that function as important regulators of gene expression in response to cellular stress and disease. An advantage that makes miRNAs attractive candidates for biomarker research is their fast release from cells in response to stress and injury, which can occur via different routes. In the context of liver transplantation (LT), noninvasive measurement and stability of extracellular miRNAs in blood, bile, and graft perfusates has been linked to cell‐type specific injury and early graft outcome following LT. Furthermore, specific intrahepatic miRNA expression patterns have been associated with graft survival and recurrent disease, like hepatitis C virus‐related fibrosis and hepatocellular carcinoma. Therefore, miRNAs with strong predictive value and high sensitivity and specificity might be successfully applied to assess hepatic injury and to diagnose (recurrent) liver disease before, during and after LT. In this review, the current features and future prospects of miRNAs as biomarkers in and out of the liver are discussed.

Liver grafts procured from donors after circulatory death have no increased risk of microthrombi formation.

Microthrombi formation provoked by warm ischemia and vascular stasis is thought to increase the risk of nonanastomotic strictures (NAS) in liver grafts obtained by donation after circulatory death (DCD). Therefore, potentially harmful intraoperative thrombolytic therapy has been suggested as a preventive strategy against NAS. Here, we investigated whether there is histological evidence of microthrombi formation during graft preservation or directly after reperfusion in DCD livers and the development of NAS. Liver biopsies collected at different time points during graft preservation and after reperfusion were triple‐stained with hematoxylin‐eosin (H & E), von Willebrand factor VIII (VWF), and Fibrin Lendrum (FL) to evaluate the presence of microthrombi. In a first series of 282 sections obtained from multiple liver segments of discarded DCD grafts, microthrombi were only present in 1%‐3% of the VWF stainings, without evidence of thrombus formation in paired H & E and FL stainings. Additionally, analysis of 132 sections obtained from matched, transplanted donation after brain death and DCD grafts showed no difference in microthrombi formation (11.3% versus 3.3% respectively; P = 0.082), and no relation to the development of NAS (P = 0.73). Furthermore, no microthrombi were present in perioperative biopsies in recipients who developed early hepatic artery thrombosis. Finally, the presence of microthrombi did not differ before or after additional flushing of the graft with preservation solution. In conclusion, the results of our study derogate from the hypothesis that DCD livers have an increased tendency to form microthrombi. It weakens the explanation that microthrombi formation is a main causal factor in the development of NAS in DCD and that recipients could benefit from intraoperative thrombolytic therapy to prevent NAS following liver transplantation. Liver Transplantation 22 1676–1687 2016 AASLD.

The release of microRNA-122 during liver preservation is associated with early allograft dysfunction and graft survival after transplantation

Early allograft dysfunction (EAD) after liver transplantation (LT) is associated with inferior graft survival. EAD is more prevalent in grafts from donation after circulatory death (DCD). However, accurate prediction of liver function remains difficult because of the lack of specific biomarkers. Recent experimental and clinical studies highlight the potential of hepatocyte‐derived microRNAs (miRNAs) as sensitive, stable, and specific biomarkers of liver injury. The aim of this study was to determine whether miRNAs in graft preservation fluid are predictive for EAD after clinical LT and in an experimental DCD model. Graft preservation solutions of 83 liver grafts at the end of cold ischemia were analyzed for miRNAs by reverse transcription polymerase chain reaction. Of these grafts, 42% developed EAD after transplantation. Results were verified in pig livers (n = 36) exposed to different lengths of warm ischemia time (WIT). The absolute miR‐122 levels and miR‐122/miR‐222 ratios in preservation fluids were significantly higher in DCD grafts (P = 0.001) and grafts developing EAD (P = 0.004). In concordance, the miR‐122/miR‐222 ratios in perfusion fluid correlate with serum transaminase levels within the first 24 hours after transplantation. Longterm graft survival was significantly diminished in grafts with high miR‐122/miR‐222 ratios (P = 0.02). In the porcine DCD model, increased WIT lead to higher absolute miR‐122 levels and relative miR‐122/miR‐222 ratios in graft perfusion fluid (P = 0.01 and P = 0.02, respectively). High miR‐122/miR‐222 ratios in pig livers were also associated with high aspartate aminotransferase levels after warm oxygenated reperfusion. In conclusion, both absolute and relative miR‐122 levels in graft preservation solution are associated with DCD, EAD, and early graft loss after LT. As shown in a porcine DCD model, miRNA release correlated with the length of WITs. Liver Transplantation 23 946–956 2017 AASLD.

Polarized release of hepatic microRNAs into bile and serum in response to cellular injury and impaired liver function

Extracellular microRNAs (miRNAs) in serum and bile are currently under intense investigation for biomarker purposes in liver disease. However, the directions and pathways by which miRNAs are released from hepatic cells remains largely unknown. Here, we investigated the release of hepatocyte and cholangiocyte‐derived miRNAs (HDmiRs and CDmiRs) into blood and bile during various (patho)physiological hepatic conditions.

MicroRNAs in bile vesicles: Finding a trade-off for biomarker discovery

1. Yang JD, Abdelmalek MF, Pang H, Guy CD, Smith AD, Diehl AM, et al. Gender and menopause impact severity of fibrosis among patients with nonalcoholic steatohepatitis. HEPATOLOGY 2014;59:1406-1414. 2. Shimizu I, Kohno N, Tamaki K, Shono M, Huang HW, He JH, et al. Female hepatology: favorable role of estrogen in chronic liver disease with hepatitis B virus infection. World J Gastroenterol 2007;13:4925-4305. 3. Standish RA, Cholongitas E, Dhillon A, Burroughs AK, Dhillon AP. An appraisal of the histopathological assessment of liver fibrosis. Gut 2006; 55:569-578. 4. Coperchot C, El Naggar A, Poupon R. Gender and liver: is the liver stiffness weaker in weaker sex? HEPATOLOGY 2006;44:513-514. 5. Roulot D, Czernichow S, Cl esiau H, Costes JL, Vergnaud AC, Beaugrand M. Liver stiffness values in apparently healthy subjects: influence of gender and metabolic syndrome. J Hepatol 2008;48:606-613.

Barking up the wrong tree: MicroRNAs in bile as markers for biliary complications

Ischemic-type biliary lesions (ITBL) are a major threat to graft survival and a major cause of recipient morbidity after liver transplantation (LT). The nonanastomotic strictures that are typical of ITBL can cause severe biliary obstructions that require interventions such as biliary stent placement and even retransplantation in up to 15% of LT recipients. Donation after circulatory death (DCD) has been identified as a major risk factor for the development of ITBL in recipients, with the incidence rising up to 30%. It is thought that the prolonged period of warm ischemia in DCD underlies the unrepairable injury to the biliary tree that predisposes patients to ITBL. However, the recognition of immune-related risk factors such as ABO-incompatible LT and recurring primary sclerosing cholangitis suggests a multifactorial etiology and complicates our understanding of the pathophysiology of ITBL. Several histological studies have described the association of cholangiocyte injury and peribiliary vascular injury at the time of graft preservation with the development of ITBL after LT. Necrosis of the bile duct arterioles is believed to cause an insufficient regenerative capacity for damaged cholangiocytes by the peribiliary glands. In addition, a reduced blood supply from the portal vein, which is responsible for approximately 40% of the blood flow through the common bile duct, has been shown to contribute to ITBL. However, the invasive character and the risk of interobserver variability in histological grading make tissue biopsies less ideal for a fast, noninvasive, and objective assessment of the diffusely spread biliary injury seen in ITBL. Therefore, developing early noninvasive biomarkers for the diagnosis and prognosis of ITBL is of increasing interest. At the time of LT, fast and noninvasive measurements of sensitive and specific biomarkers could be helpful in quantitatively assessing the degree of biliary injury in marginal grafts susceptible to ITBL. Recent work from our research group has shown that cholangiocyte-derived microRNAs (miRNAs) in graft preservation solutions before LT are predictive of ITBL. These findings suggest that the release of miRNAs by cholangiocytes is adjusted in response to biliary injury at the time of graft preservation, even before the liver is actually implanted into the recipient. Indeed, miRNAs are emerging as novel biomarkers that possess several useful features, including cell-type specificity, sensitivity, and stability. In particular, extracellular miRNAs, whose presence has been demonstrated in a wide variety of body fluids, have become attractive candidate biomarkers for minimally invasive or noninvasive detection of cellular injury. miRNAs derived from hepatocytes, for instance, are already elevated in the serum of recipients during episodes of acute cellular rejection well before a rise in transaminases can be detected. Moreover, extracellular miRNAs are more resistant against ribonuclease activity than longer messenger RNAs at room temperature or after repeated freezethaw cycles.