Received: 2 March 2021 | Revised: 18 June 2021 | Accepted: 5 July 2021 DOI: 10.1111/liv.15013 REVIEW Outcomes and potential surrogate markers for future clinical trials of non-­alcoholic steatohepatitis cirrhosis Jesús Rivera-­Esteban1,2 | Angelo Armandi3 | Salvador Augustin1,4 | Elisabetta Bugianesi3 1 Liver Unit, Department of Internal Medicine, Vall d’Hebron Hospital Abstract Universitari, Vall d’Hebron Institut de Non-­alcoholic steatohepatitis has emerged as a major public health problem, and the Recerca (VHIR), Vall d’Hebron Barcelona Hospital Campus, Barcelona, Spain burden of non-­alcoholic steatohepatitis cirrhosis is projected to increase by 64%-­ 2 Universitat Autònoma de Barcelona, 156% by 2030. The threat is aggravated by the fact that are currently no approved Bellaterra, Spain drugs for the treatment of non-­alcoholic steatohepatitis. In this paper, we review the 3 Department of Medical Sciences, Division of Gastroenterology and Hepatology, A.O. main challenges to drug development in patients with non-­alcoholic steatohepati- Città della Salute e della Scienza di Torino, tis cirrhosis, and describe the opportunities brought by the advances in the under- University of Turin, Turin, Italy 4 standing of the clinical and pathophysiological nuances of cirrhosis. The design of Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas therapeutic regimens for non-­alcoholic steatohepatitis cirrhosis will vary according (CIBERehd), Instituto de Salud Carlos III, to the specific cirrhosis substage (compensated vs decompensated), and the spe- Madrid, Spain cific mechanistic basis of therapy, targeted either at improving aetiology-­specific Correspondence pathways and/or at more general aetiology-­agnostic processes. The understanding Elisabetta Bugianesi, MD, PhD, Department of Medical Sciences, Division of of the probabilistic expectations for the whole range of potential outcomes, rooted Gastroenterology and Hepatology, A.O. at different mechanistic drivers at each specific substage, will be essential in order Città della Salute e della Scienza di Torino, Corso Dogliotti 14, 10126, University of to choose adequate estimands and therapeutic strategies for clinical trials and in- Turin, Turin, Italy. dividual patients with non-­alcoholic steatohepatitis cirrhosis. Finally, we provide a Email: elisabetta.bugianesi@unito.it summary of the main pitfalls and uncertainties in the design of clinical trials for non-­ Funding information alcoholic steatohepatitis cirrhosis and discuss potential biomarkers for use in trials Funding was provided by Boehringer Ingelheim. Boehringer Ingelheim did not and practice for these patients. have a role in the study design, analysis or interpretation of the data, the writing of the KEYWORDS manuscript or the decision to submit the manuscript for publication. biomarkers, clinical trials, liver cirrhosis, metabolic associated fatty liver disease (MAFLD), portal hypertension, portal pressure Editor: Salvatore Petta Abbreviations: ACLF, acute-­on-­chronic liver failure; CC, cryptogenic cirrhosis; CD, clinical decompensation; CSPH, clinically significant portal hypertension; CVD, cardiovascular disease; ELF, enhanced liver fibrosis; FDA, Food and Drug Administration; HCC, hepatocellular carcinoma; HE, hepatic encephalopathy; HVPG, hepatic venous pressure gradient; Hyperdynamic circ. Sd., hyperdynamic circulatory syndrome; IHVR, intrahepatic vascular resistance; MELD, model for end-­s tage liver disease; NAFLD, non-­alcoholic fatty liver disease; NASH, non-­alcoholic steatohepatitis; OLT, orthotopic liver transplantation; PBF, portal blood flow; PH, portal hypertension; PRO, patient reported outcome; RCTs, randomized controlled trials; T2DM, type 2 diabetes mellitus. Jesús Rivera-­E steban and Angelo Armandi are Joint first authors. Salvador Augustin and Elisabetta Bugianesi are Joint senior authors. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. Liver International published by John Wiley & Sons Ltd. Liver International. 2021;41:1999–2008.  wileyonlinelibrary.com/journal/liv | 1999 2000 | RIVERA-­E STEBAN et al 1 | TH E B U R D E N O F C I R R H OS I S D U E TO N O N -­A LCO H O LI C S TE ATO H E PATITI S Key points (NASH) • Non-­alcoholic steatohepatitis (NASH) represents a major public health problem, with the burden of NASH Non-­alcoholic fatty liver disease (NAFLD) has emerged as a major cirrhosis projected to increase 64%-­156% by 2030. public health threat, and the burden of end-­stage liver disease due • There are no Food and Drug Administration-­approved to NAFLD is projected to increase from 64% to 156% by 2030.1 All drugs for NASH cirrhosis, and clinical outcomes are the longitudinal studies conducted on biopsy-­proven NAFLD cohorts only recommended endpoints for market approval. over the past 2 decades have highlighted the number of patients • Therapeutic regimens will vary according to cirrhosis stage with cirrhosis. The prevalence of cirrhosis is currently estimated to (compensated vs decompensated) and the mechanistic range from 0.6% to 65%, with a median value of 9% (Table 1). 2-­16 basis of therapy (aetiology specific vs symptomatic). Overall mortality in patients with NAFLD is predicted to be approxi- • A clear definition of NASH cirrhosis, understanding of mately 20%, while liver-­related mortality is approximately 4% across liver disease biology and detailed patient risk stratifica- all stages of fibrosis, showing a stepwise progression from mild to tion are required for future clinical trials. severe fibrosis.17-­19 • Non-­invasive tests, eg enhanced liver fibrosis and liver By 2030, cirrhosis and end-­stage liver disease related to NASH stiffness, are promising biomarkers. are projected to increase worldwide.1 Models of future disease bur- den for incident decompensated cirrhosis have shown the highest percentage in France (164%) followed by the USA (150%).1 In addi- tion, projections from one Canadian study showed that cirrhosis due 2 | CU R R E NTLY A PPROV E D E N D P O I NT S to NASH will increase in all birth cohorts by 2040, accounting for I N N A S H C I R R H OS I S TR I A L S 73% of all new diagnoses of cirrhosis. 20 As shown in Table 2, the median proportion of cirrhotic There are currently no Food and Drug Administration (FDA)-­approved patients with liver decompensation in longitudinal studies of drugs for compensated NASH cirrhosis. Patients with cirrhosis are the biopsy-­p roven NAFLD cohorts is 13.4%, ranging from 3% to most likely to develop hard outcomes (eg death, HCC, liver transplan- 4-­6 ,8,10,12-­16 45% across the different cohorts. Clinical events tation). Current trials in patients with cirrhosis capture outcomes re- related to portal hypertension are the most common findings: lated to cirrhosis, such as hepatic encephalopathy, ascites and variceal ascites (median 8.2%, min–­m ax: 0.4%-­7 0%), variceal bleeding haemorrhage, as well as the laboratory components of the Child–­Pugh (median 8.6%, min–­m ax: 0%-­6 6.4%) and hepatic encephalopa- score. Improvement in histological fibrosis stage or hepatic collagen thy (median 4%, min–­m ax: 0%-­31.6%). Accordingly, the develop- content have been also used as primary endpoints of outcome events ment of liver decompensation constitutes a healthcare burden in trials focused on NASH cirrhosis. However, the reversal of cirrhosis that requires high resource utilisation, and is related to inpa- to lesser degrees of fibrosis is an ambitious goal, and the relationship tient and short-­t erm mortality. 21 Onset of hepatocellular car- between histological changes in cirrhosis and clinical outcomes has cinoma (HCC) represents another relevant clinical outcome, as not been characterised. Improvements in fibrosis stage may reflect well as occurrence of liver transplantation (median 0.5%, min–­ either true regression or sampling bias. max: 0%-­3 4.2%), that underlines the progressiveness of NAFLD. Currently, the only endpoints recommended by the FDA to sup- Otherwise, cardiovascular disease (CVD) involves a low rate of port marketing approval in compensated NASH cirrhosis are clinical events in cirrhotic patients compared with liver-­r elated out- outcomes.23 According to the FDA, Phase III trials in patients with com- comes and to the relative weight of morbimortality due to CVD pensated NASH cirrhosis should evaluate the effect of the investiga- in pre-­c irrothic stages.16 tional drug relative to placebo on the composite endpoint of time from One recent study in Wales22 conducted in nearly 70 000 in- randomisation to the first of any one of the following outcome events: dividuals affected by cirrhosis has shown that the incidence of NAFLD has increased 10-­f old over the past 10 years and has be- 1. Complication of ascites including any of the following: sponta- come the predominant cause of liver damage. As compared with neous bacterial peritonitis, diuretic-­resistant ascites (refractory other aetiologies, the course of NASH-­r elated liver disease ap- ascites), hepato-­pleural effusion. pears to be milder, with a smaller proportion of decompensated 2. Variceal haemorrhage. patients (8% of patients with NASH cirrhosis, vs 25% of patients 3. Hepatic encephalopathy. with alcohol-­r elated cirrhosis). However, clinical outcomes ex- 4. Worsening model for end-­stage liver disease (MELD) score to tracted from current longitudinal studies need to be related to greater than or equal to 15 (this endpoint approximates listing for the rapidly increasing burden of NASH cirrhosis, which will even- liver transplant). tually cause a considerable increase in liver-­r elated events and 5. Liver transplantation. mortality. 6. Death from any cause. RIVERA-­E STEBAN et al | 2001 TA B L E 1 Prevalence of cirrhosis, and overall and liver-­related mortality in longitudinal studies of biopsy-­proven NAFLD cohorts2-­16 Year of Number of Cirrhosis All-­cause Liver-­related Follow Study publication subjects prevalence (%) mortality (%) mortality (%) up (years) Matteoni et al2 1999 132 20.4 48.9 9.1 8.3 3 Dam-­L arsen et al 2004 109 0.6 24.8 0.9 16.7 Adams et al4 2005 420 5a 12.6 1.7 7.6 5 Ekstedt et al 2006 88 7.9 20.1 2.8 13.7 Sanyal et al6 2006 152 100a 19 4.6 10 7 Soderberg et al 2010 118 9 40 6.7 28 Bhala et al8 2010 247 54 13.4 5.6 7.4 9 Younossi et al 2011 210 10.5 26.6 8.5 12.1 Sebastiani et al10 2015 148 14.9 6 2 5 11 Ekstedt et al 2015 224 1.8 42.4 4 26.4 Angulo et al12 2015 619 2.9 31.1 4.2 12.6 13 Seko et al 2015 312 8 2.6 0.3 4.8 Leung et al14 2016 307 15.3 2.6 0.3 4.1 Hagström et al15 2017 646 3.1 33.1 7.9 20 Vilar-­Gomez et al16 2018 458 65.2 8 6.7 5.5 Median values 9 20.1 4.2 Abbreviation: NAFLD, non-­alcoholic fatty liver disease. a Diagnosed either by biopsy or clinical/imaging assessment. TA B L E 2 Proportions of cirrhotic patients with overt decompensation and most frequently reported liver-­related events in longitudinal studies of biopsy-­proven NAFLD cohorts Study Liver decompensation (%) Ascites (%) Variceal haemorrhage (%) HCC (%) HE (%) OLT (%) 4 Adams et al 3.1 2 1 0.5 2 0.2 Ekstedt et al 20065 5.7 4.5 1.1 2.3 0 2.3 6 Sanyal et al 45 42.8 66.4 1.2 31.6 34.2 Bhala et al8 19.4 7.7 10.5 2.4 7.7 NA 10 Sebastiani et al 16.2 8.7 6.7 0.7 0 1.3 Angulo et al12 13.4 34.6 46 11.5 23.1 0.5 13 Seko et al NA NA NA 1.9 NA 0 Leung et al14 1.6 0.4 0 0.9 0.4 0 15 Hagström et al 11.8 NA NA 1.9 NA 0 Vilar Gomez et al16 19 70 24 9 6 8 Median values 13.4 8.2 8.6 1.9 4 0.5 Abbreviations: HCC, hepatocellular carcinoma; HE, hepatic encephalopathy; NA, not applicable; OLT, orthotopic liver transplantation. 3 | W H I C H PATI E NT S W ITH N A S H aetiology-­specific pathways (such as the metabolic derangements C I R R H OS I S A R E E LI G I B LE FO R typical of NASH) or at more general aetiology-­agnostic processes PH A R M ACO LO G I C A L TH E R A PY A N D H OW (such as the intrahepatic or extrahepatic vascular dysregulation TO S E LEC T TH E M : TH E D I FFE R E NT S TAG E S common to all forms of portal hypertension). 24-­26 O F N A S H C I R R H OS I S The classical concept of cirrhosis is histological in nature, and has been traditionally regarded as a static, non-­reversible last stage The structure and focus of therapeutic regimens for NASH cirrho- for all forms of progressive liver diseases. 24 However, both the suc- sis will vary according to 2 main factors: (i) the specific substage of cess of antiviral therapies in cirrhosis, as well as the integration of cirrhosis at which the therapeutic strategy is directed, and (ii) the clinical and haemodynamic knowledge generated in the past 2 de- specific mechanistic basis of therapy–­either targeted at improving cades, have helped to develop a more comprehensive, dynamic and 2002 | RIVERA-­E STEBAN et al nuanced view of this last phase of chronic liver disease. 24 Nowadays, for earlier Phase II trials. 25 In order to assess the impact of hard clin- there are at least 4 distinct, well-­differentiated substages of cirrho- ical endpoints in this early NASH cirrhosis population, large sample sis, classified according to the mechanisms driving progression and sizes and a longer duration of follow up are likely to be required, as potential for regression of disease at each stage and the expected seen in current Phase III/IV trials.32,38 probabilities of a different range of outcomes. 24,26 Table 3 sum- Once the critical CSPH threshold is reached, the risk of CD in- marises a schematic view of these 4 substages and the potential im- creases considerably, 27,31,32 and therefore the prevention of CD plications for the treatment of patients with NASH cirrhosis. should become the main therapeutic goal and the ideal endpoint for From a clinical standpoint, the occurrence of a first clinical de- trials. 25 During this late compensated stage, structural intrahepatic compensation (CD) indicates a dramatic worsening of prognosis changes are characterised by the presence of thicker fibrotic septa, in patients with cirrhosis and enables division into 2 main stages: thus making the regression of fibrosis unlikely. 24 Also, extrahepatic 27 compensated and decompensated cirrhosis. Indeed, the ex- adaptations start to develop, with an increase in portal blood flow, the pected median survival of a patient with compensated cirrhosis is development of collateral circulation and systemic adaptations lead- 10-­12 years, compared with the 1.5-­2 years that could be expected ing to the initial phases of a mild hyperdynamic circulatory state.39,40 28,29 after a patient's first decompensating event (of note, these Once these adaptations appear, splanchnic vasoconstrictors start to rates are drawn from studies including patients with all aetiologies become effective at halting the progression of portal hypertension of cirrhosis–­the impact on survival of the transition from compen- and reducing the risk of decompensation.40,41 Extrapolating from ex- sated to decompensated stage has not been reported specifically for perience with antiviral therapies, it can be assumed that for NASH NASH-­). cirrhosis, there might also be a ‘point of no return’ somewhere after The compensated stage is the longest phase and can go undi- CSPH is established, for which the effects of aetiology-­specific ther- agnosed given its asymptomatic nature. Nonetheless, early identifi- apies are no longer able to reverse CSPH and eliminate the risk of CD cation of patients at this stage would be of paramount importance, without concomitant medication.34-­37 Under these pathophysiolog- since this is precisely the phases at which therapies might be still able ical assumptions, therapeutic efforts in this late compensated phase to prevent the transition to the decompensated stage, or even lead to should aim to target the same aetiology-­specific mechanistic path- a regression to pre-­cirrhotic stages. Within the compensated stage, ways as well as both intrahepatic and extrahepatic vascular tone. As the presence and degree of portal hypertension is a key predictor of a matter of fact, in patients already harbouring oesophageal vari- outcome. By integrating haemodynamic data, further distinction of ces, the use of non-­selective beta-­blockers is the standard of care25 prognostic relevance can be made into an early compensated stage and cannot be avoided as a comparator for trials. 26 The transition 24,26,27,30 and a late compensated stage. The early compensated stage of HVPG values to <10 mm Hg could be considered as a potentially is characterised by a mild degree of portal hypertension (defined by relevant surrogate endpoint in these patients, 25,26 but histological an hepatic pressure venous gradient (HVPG) <10 mm Hg), and a low changes in NASH and fibrosis become less likely32 and are thus not 31,32 risk of CD (<10% at 4 years). The late compensated stage is char- ideal endpoints. acterised by the presence of clinically significant portal hyperten- Finally, the occurrence of CD signals that intrahepatic and ex- sion (CSPH), defined by an HVPG ≥10 mm Hg, the threshold above trahepatic structural and functional adaptations are insufficient and which CD is up to 4 times more likely to occur.31,32 therefore that the patient has progressed to NASH cirrhosis. 24,30 The distinction between the early compensated stage and CSPH The risk of further decompensation, progression to end-­stage is not merely prognostic but also carries pathophysiological and ther- liver disease and death increases exponentially as decompensat- apeutic implications. In the early compensated stage, the main driv- ing events accumulate. 26,28 From a pathophysiological perspective, 25,26 ers of progression and outcomes are aetiology-­specific. There there is a progression of the hyperdynamic circulatory state, peaking seem to be intrahepatic vascular adaptations, both structural and in patients with refractory ascites, for whom there is a relative de- functional–­the latter being characterised by a moderate increase in crease in the cardiac index that leads to a further decompensated intrahepatic vascular resistance (IHVR)–­all of which translates into stage. 24,26 Mounting evidence also points to a deleterious hyperco- 30,33 mild portal hypertension. Thus, within this stage, therapeutic agulable state in decompensated cirrhosis, especially affecting the efforts should be directed towards arresting the driving mechanism micro-­circulation.42 All of these changes, along with the decrease in of NASH. The underlying assumption (extrapolated from the lessons synthetic liver function, lead to an increase in gut bacterial trans- learnt from antiviral therapies34-­37), is that the improvement of those location and systemic inflammation, which are key drivers of the aetiology-­specific mechanisms will carry over indirectly into the ar- decompensating inertia of this stage, eventually leading to a final rest (or even regression) of fibrosis and the functional improvement high-­mortality stage characterised by multiorgan failure. 24,27 The of IHVR, leading to a decrease of portal pressure and the risk of CD. role of aetiology-­specific therapies is likely secondary in these later Within this clinical and pathophysiological framework, the choice of stages and, for the specific case of NASH, might even be associ- outcomes in this specific early stage should focus on histological im- ated with deleterious effects, especially with weight loss-­directed provement of NASH features and fibrosis as the main endpoint (as approaches, which could lead to worsening of patients’ nutritional in current Phase III trials). Progression to CSPH and/or resolution of and performance status and an increased risk of complications and portal hypertension could be considered as secondary outcomes or death.43,44 Aetiology-­agnostic approaches in this phase might be TA B L E 3 The different stages of NASH cirrhosis: Clinical features, mechanistic drivers and ideal therapeutic goals and endpoints Compensated Decompensated NASH cirrhosis NASH cirrhosis RIVERA-­E STEBAN et al Late compensated Early compensated (compensated CSPH) Early decompensated Further decompensated Clinical features Unspecific symptoms Unspecific symptoms Few/mild decompensating events • Recurrent/refractory decompensating (± varices) events • ACLF (multiorgan failure) Biological features Histology Scar, X-­linking Acellular scar, nodules (thick Insoluble scar Insoluble scar (thin septa) septa) Haemodynamic 5-­10 mm Hg >10 mm Hg >12 mm Hg >16 mm Hg (HVPG) (mild PH) (CSPH) Mechanistic drivers Primary NASH mechanisms • NASH mechanisms • Increased portal pressure (PBF+IHVR) • Failed hyperdynamic compensation (metabolic/ (fibrosis) • Hyperdynamic circ. Sd. • Decreased liver function inflammation/ • Increased IHVR fibrosis) Secondary Increased IHVR • Increased PBF • Decreased liver function • Increased inflammation • Collaterals development • Increased inflammation • Increased bacterial translocation • Increased bacterial translocation • Hypercoagulability • Hypercoagulability • Increased portal pressure Therapeutic goals/ endpoints Primary • Improvement in NASH histology • Prevention of liver clinical • Transplant-­free survival • Overall survival • (NASH resolution and/or fibrosis events (CD) • Overall survival • PROs regression?) Secondary • Prevention of liver +non-­liver clinical • Improvement in NASH • Recurrence/control of CD events • Recurrence/control of CD events events histology • Mechanistic/proof-­of-­concept: specific • Mechanistic/proof-­of-­concept: specific to • Resolution of PH • Prevention of non-­liver to type of CD type of CD • Decreased progression to CSPH clinical events • Changes in MELD/Child-­Pugh scores • Post-­transplant outcomes • New fibrosis biomarkers? • Decrease in HVPG (>20% • PROs and/or <10 mm Hg) • New fibrosis/liver function biomarkers? Abbreviations: ACLF, acute-­on-­chronic liver failure; CD, clinical decompensation; CSPH, clinically significant portal hypertension; HVPG, hepatic venous pressure gradient; Hyperdynamic circ. Sd., hyperdynamic circulatory syndrome; IHVR, intrahepatic vascular resistance; MELD, model for end-­stage liver disease; PBF, portal blood flow; PH, portal hypertension; PRO, patient reported outcome. | 2003 2004 | RIVERA-­E STEBAN et al more generally directed to the modification of intra and extrahe- NASH cirrhosis, with respect to those with unexplained (‘crypto- patic vascular tone dysregulation or can be more specifically tar- genic’) cirrhosis (CC).50 Data suggest that patients with NASH cir- geted to the specific complications (control of fluid retention for rhosis and CC form different parts of the same spectrum of chronic ascites, nitrogen metabolism for hepatic encephalopathy [HE], etc). liver disease that originates in the setting of metabolic abnormalities, Therapeutic efforts to control infection and systemic inflammation, as these patients have similar metabolic and clinical features, but are and even to counterbalance the described hypercoagulable state, not currently enrolled in clinical trials because a threshold of ste- should also have a role in this stage. atosis (>5%) is required as an inclusion criterion. In a recent study Considering the poor prognosis in the decompensated stage, among CC patients with <5% fat at baseline biopsy who had another transplant-­free survival is recommended as the primary endpoint biopsy during follow-­up, 40.4% had >5% fat on their subsequent for Phase III trials. 25,45 Either surrogate or secondary endpoints for biopsy, and, therefore, were no longer ‘cryptogenic’. Furthermore, Phase III trials (such as development of multiorgan involvement as in patients with CC seem to have a more aggressive disease than those acute-­on-­chronic liver failure), or primary endpoints for Phase II tri- with NASH cirrhosis, as indicated by greater hepatic collagen con- als focusing on specific decompensating events (recurrence of var- tent and α-­SMA expression on biopsy, higher serum fibrosis markers iceal haemorrhage, hepatic encephalopathy, control of ascites, etc), and MELD scores and a greater risk of liver-­related clinical events should be chosen according to the mechanism of action of the drug during follow-­up.51 under investigation and the specific targeted patient subpopulation. The recent proposal to substitute the term NASH by Metabolic Inclusion of patient reported outcomes should become standard (dysfunction) Associated Fatty Liver Disease (MAFLD)52 stems in practice in this symptomatic, late stage of NASH cirrhosis. Three ex- part from these challenges with the disease definition. The new cellent recent reviews provide a comprehensive view of the different term brings the focus to the metabolic substrate at the mechanistic tools and approaches for reporting quality of life in patients with and clinical basis of the disease. For the particular of NASH cirrhosis, advanced liver disease.46-­48 this recognition may help at clarifying in part the conundrum of CC In summary, the understanding of the probabilistic expecta- described above, since most patients with CC share the same met- tions for the whole range of potential outcomes, rooted at different abolic substrate of NASH and have overlapping clinical progression. mechanistic drivers at each specific substage of NASH cirrhosis, is The new term also provides an enabling framework from where to essential in order to choose adequate estimands and therapeutic reinterpret the coexistence in real practice of metabolic and alco- strategies for clinical trials, and tailor therapy to individual patients holic liver disease, with the recognition that the two conditions are with NASH cirrhosis. not mutually exclusive and can coexist together. Also, the removal of specific thresholds of alcohol intake from the definition, forces the revision of the assumption that slight to moderate alcohol intake 4 | PITFA LL S A N D U N C E RTA I NTI E S I N have no deleterious hepatic effects, which is implicit in the classical C LI N I C A L TR I A L S FO R N A S H C I R R H OS I S definition of NASH. This is also relevant in patients with NASH cir- rhosis, in whom the potential impact of alcohol in the progression of 4.1 | Definition of NASH cirrhosis the disease becomes especially critical. Nonetheless, adoption of the new term is still a matter of debate, An accurate case definition of NASH cirrhosis is critical for enroll- and proof of it is that the whole MAFLD concept has not been ap- ing appropriate patients into clinical trials, but currently there are no plied in clinical trials yet, where precise and well-­characterised crite- published criteria for defining NASH cirrhosis. NASH is a histological ria are still needed to avoid bias and confounders and for regulatory diagnosis, but in clinical practice, the term NASH cirrhosis has been reassurance. In this regard, the Liver Forum recently developed a increasingly used by physicians to address the forms of end-­stage consensus case definition of NASH-­related cirrhosis for inclusion liver diseases in the presence of known risk factors for metabolic in clinical trials that may be qualitatively categorised according to derangement, mainly obesity and type 2 diabetes mellitus. However, the degree of certainty of NASH as the cause of cirrhosis: definitive, the progression of NASH is associated with a loss of intrahepatic fat probable and possible.53 When histological evidence is absent, the 49 and diminished inflammatory activity. Therefore, in the absence presence of concomitant metabolic risk factors strengthens the like- of a prior diagnosis of NASH, end-­stage liver histology cannot be lihood that NASH is the cause of cirrhosis (Box 1). linked with a specific aetiology, and the term NASH cirrhosis remains hypothetical. In this line, the FDA requires fulfilment of full NASH histological criteria for enrolment in phase 3 trials in the F4 popula- 4.2 | Biology of liver disease in cirrhosis tion. This certainly represents a significant hurdle for recruitment, since aproximatelhy, half of F4 patients do not show full NASH fea- The drugs currently in development for NASH can be divided into 49 tures, which explain in part the high screen failure rate common in 3 broad categories: metabolic, anti-­inflammatory and antifibrotic. NASH trials. Favourable metabolic effects from a drug are desirable in patients The presence of a suggestive medical history and/or hallmarks with NASH, but to what extent the partial correction of these meta- of metabolic syndrome may help to characterise patients with true bolic abnormalities leads to a more favourable outcome in cirrhosis RIVERA-­E STEBAN et al | 2005 BOX 1 NASH Cirrhosis: Liver Forum consensus definitions for clinical trials48 1. Definite NASH 1a. Patients with current liver biopsy showing cirrhosis with steatohepatitis. cirrhosis 1b. Patients with a previous biopsy showing steatohepatitis, but now with evidence of cirrhosis, either by clinical history or current features, imaging, non-­invasive tests or biopsy. 1c. Patients with a current biopsy showing cirrhosisa with steatosis (but no findings of active steatohepatitis) together with at least two coexisting or historical metabolic comorbidities including obesity and/or T2DM to corroborate a diagnosis of NASH as the cause of cirrhosis. 2. Probable NASH 2a. Patients with a previous biopsy with steatosis but not steatohepatitis, and current cirrhosisa . At least cirrhosis two coexisting or historical metabolic comorbidities including obesity and/or T2DM to support NASH as an underlying cause of cirrhosis. 2b. Patients with cirrhosisa with current or previous imaging showing evidence of steatosis. At least two coexisting or historical metabolic comorbidities including obesity and/or T2DM to support NASH as an underlying cause of cirrhosis. 2c. Patients with ‘cryptogenic cirrhosis’ without current or previous evidence of steatosis by imaging or steatosis/steatohepatitis by histology. At least two coexisting or historical metabolic comorbidities including obesity and/or T2DM to support NASH as an underlying cause of cirrhosis. 3. Possible NASH 3a. Patients with ‘cryptogenic cirrhosis’ without current or previous evidence of steatosis by imaging or cirrhosis steatosis/steatohepatitis by histology. At least one coexisting or historical metabolic comorbidity including obesity and/or T2DM. a Either by a clinical history or current features, imaging, non-­invasive tests or biopsy. Abbreviations: NASH: non-­alcoholic steatohepatitis; T2DM: type 2 diabetes mellitus. is unknown. The biology of clinical decompensation in patients with a clinically meaningful outcome. The expected risk of each outcome previously compensated cirrhosis is not fully understood. For exam- changes markedly from stage to stage, and sample size and follow- ple, diabetes arising in liver cirrhosis (so-­called hepatogenous dia- ­up times should vary accordingly. This nuanced view of the different betes) has a profound impact on the pathology and natural history stages of cirrhosis is often overlooked in the design of large clinical of the liver disease.54 Although peripheral insulin resistance and trials in NASH cirrhosis. From an operational point of view, strati- impairment of the hepatocellular function are two potential major fication by direct or indirect evidence of CSPH (HPVG>10 mm Hg, causes, the beta-­cell capacity plays a critical role. Recent evidence esophageal varices at endoscopy or, hopefully in the near future, that the failing liver exerts an independent ‘toxic’ effect on beta-­ non-­invasive biomarkers of CSPH) and/or a decreased liver function cells suggests that individuals with hepatogenous diabetes might (Child-­Pugh score ≥6 points) could be used to fine-­tune the assump- benefit from interventions aimed at improving beta-­cell function, tions around expected event rates informing sample size, trial dura- such as thiazolidinediones and incretins.55 On the other hand, cir- tion, etc A more granular and detailed sub-­stratification of patients rhosis is a condition of intense muscle protein wasting leading to sar- with NASH cirrhosis and the corresponding choice of outcomes is copenia even in individuals affected by obesity (so-­called sarcopenic needed to stimulate therapeutic development in this population. obesity).43 Malnutrition and sarcopenia are associated with higher complication rates in patients with cirrhosis.43 Moreover, they are associated with increased mortality in hospitalised patients with 4.4 | Placebo reponse and Hawthorne effect cirrhosis and those waiting for liver transplantation.43 In this set- ting, drugs promoting weight loss may have a detrimental impact on Placebo response represents a major challenge in NASH clinical tri- the progression of cirrhosis. Therefore, it remains to be determined als. Patients allocated to placebo arms in therapeutic trials show whether treatment of the underlying steatohepatitis as opposed to significant histologic, radiologic and biochemical responses56,57 that the fibrosis will prevent clinical decompensation. These possibilities must be taken into account as they can interfere with the overall trial are currently under active investigation. design and an adequate interpretation of results. Hawthorne effect is particulary relevant in a lifestyle-­driven disease such as NASH, as subjects may conciosuly or unconsciously change their behaviour 4.3 | Risk stratification after enrolment, directly affecting NASH biology and outcomes.58 To control this effect adequately, trials should include a standard- As described above, cirrhosis represents a broader spectrum of dis- ised approach to lifestyle interventions and an objective assessment ease compared with its usual histological classification (F4), and the of lifestyle, including physical activity, dietary and alcohol intakes main challenge is the stratification of NASH patients with compen- questionnaires. In this regard, the Liver Forum has recently issued sated cirrhosis to identify strata for decompensation that represent a comprehensive review and position paper on this important topic, 2006 | RIVERA-­E STEBAN et al with specific recommendations.59 Trial duration may also affect the changes in fibrosis (even in the F4 stage) with clinical outcomes, al- impact of the Hawthorne effect. Long trail durations, as currently though the number of events in those clinical trials was very small required for phase 3 trials in NASH cirrhosis, might mitigate the po- and further studies are required to increase the confidence in this tential impact of a Hawthorne effect (since lifestyle modifications novel and exciting techniques. are harder to maintain in the long term), but this is yet to be proven. Other factors such as sample size or geographic location must be also considered when interpreting the results of placebo arms in 5.3 | New biomarkers in NASH cirrhosis RCTs. A major challenge in drug development is to identify and validate surrogate markers that predict a reduction in progression to hard 5 | S U G G E S TE D S U R RO G ATE S A N D N E W outcomes. Different exploratory efficacy endpoints have been BIOMARKERS studied in clinical trials in patients with NASH cirrhosis, including changes in liver biochemistry tests; non-­invasive tests (eg enhanced 5.1 | HVPG liver fibrosis [ELF], liver stiffness by transient elastography); mark- ers of apoptosis and necrosis and other histological measures in- As discussed above, the HVPG remains the most robust and accu- cluding hepatic collagen and fat content and α-­SMA expression.32,61 rate surrogate predictor of outcomes in patients with compensated The demonstration that ELF score and liver stiffness had the ability and early decompensated cirrhosis of different aetiologies, including to identify those most likely to progress from F3 to cirrhosis, in ad- NASH. 25,31 In patients with cirrhosis of viral aetiology, improvement dition to predicting which patients with cirrhosis at baseline were of HVPG after antiviral therapy translates into a clear improvement most likely to have a liver-­related event,61 was of particular interest. in outcomes.34-­37 In the case of NASH cirrhosis, observations from The authors defined a change in ELF score of 0.5 points as the value the simtuzumab programme suggest that reductions in HVPG (at that correlated with clinical liver events. This value also correlated least 20% and/or below 10 mm Hg) is associated with a significant with other non-­invasive tests, liver biochemistry tests, glycaemic pa- decrease in the risk of clinical decompensation.32,38 However, HVPG rameters, CK-­18 values, serum bile acid values and body weight, but measurement is technically challenging, invasive and expensive, and not with histological features. These findings give hope that in the is still regarded as logistically demanding for use in large clinical tri- future, reliance on histological endpoints will be a historical concept als. Furthermore, in the case of earlier compensated stages (before for NASH clinical trials. Validation from such biomarkers will hope- CSPH), which are the focus of most current programmes in NASH fully be a reality as clinical events accumulate in ongoing Phase III cirrhosis, more studies are required to demonstrate that improve- trial efforts. ment in HVPG (due to reduced fibrotic remodelling of the liver or Functional testing is a novel approach to assess the severity of other intrahepatic structural or functional mechanisms) translates liver disease and changes in actual liver function in the context of into improved clinical outcomes. On this basis, the most recent treatment trials, especially in patients with cirrhosis. The HepQuant FDA guidance on clinical trials in NASH cirrhosis does not discuss test simultaneously measure clearance from portal and systemic cir- HVPG as a valid endpoint for marketing approval in Phase III stud- culation as well as portal-­systemic shunting.62 Preliminary data in- 53 ies. However, in the case of earlier development (Phases II and IIb), dicate that this assessment of liver impairment correlates with hard HVPG remains an attractive surrogate endpoint in compensated endpoints, but more extensive validation is needed. Another test is NASH cirrhosis, due to the large numbers of patients and long du- the methacetin breath test, which is being evaluated in several ongo- ration of follow up that would be required in trials with a primary ing clinical trials for NASH,63,64 and longitudinal data that correlate endpoint of clinical outcomes, and the strong association of HVPG this assessment with histological changes may be forthcoming. How with clinical outcomes in this setting. the HepQuant, methacetin breath test and other tests of specific metabolic functions of the liver compare with clinical outcomes re- mains to be determined. 5.2 | Improved histological readings In an attempt to mitigate the inter and intra-­observer variability in 6 | CO N C LU S I O N S NASH histology readings discussed in the first section, improved machine-­driven methods to quantify basal amount of fibrosis and NASH cirrhosis has emerged as a major healthcare problem and changes in time are being increasingly explored in the last few years. finding effective therapies for patients with NASH has become one A recent paper evaluating a machine-­learning (ML) based-­approach of the main unmet clinical needs in hepatology. Drug development with paired biopsies from 3 large NASH RCTs including patients with in this field will face important challenges, due to the time needed advanced fibrosis (STELLAR-­3, STELLAR-­4 and ATLAS)60 showed to gather hard outcomes and the unreliable nature of fibrosis as an that these artificial intelligence-­driven techniques are sensitive and endpoint. A deeper understanding of the clinical nuances of NASH reliable, and represent promising approaches to correlate dynamic cirrhosis, and the mechanistic processes driving disease progression RIVERA-­E STEBAN et al | 2007 at each substage, will hopefully guide therapeutic efforts and ap- 11. Ekstedt M, Hagström H, Nasr P, et al. Fibrosis stage is the strongest predictor for disease-­specific mortality in NAFLD after up to 33 propriate patient selection, informative biomarkers and clinically years of follow-­up. Hepatology. 2015;61:1547-­1554. meaningful endpoints. 12. Angulo P, Kleiner DE, Dam-­L arsen S, et al. Liver fibrosis, but no other histologic features, is associated with long-­term outcomes AC K N OW L E D G E M E N T S of patients with nonalcoholic fatty liver disease. Gastroenterology. 2015;149:389-­397.e310. Funding was provided by Boehringer Ingelheim. Boehringer 13. Seko Y, Sumida Y, Tanaka S, et al. Predictors of malignancies and Ingelheim did not have a role in the study design, analysis or inter- overall mortality in Japanese patients with biopsy-­proven non-­ pretation of the data, the writing of the manuscript or the decision alcoholic fatty liver disease. Hepatol Res. 2015;45:728-­738. to submit the manuscript for publication. 14. Leung J-­F, Loong T-­W, Wei JL, et al. Histological severity and clinical outcomes of nonalcoholic fatty liver disease in nonobese patients. Hepatology. 2017;65:54-­6 4. C O N FL I C T S O F I N T E R E S T 15. Hagström H, Nasr P, Ekstedt M, et al. Fibrosis stage but not NASH Jesús Rivera-­Esteban reports no conflicts of interest. Angelo predicts mortality and time to development of severe liver disease Armandi reports no conflicts of interest. Salvador Augustin is now in biopsy-­proven NAFLD. J Hepatol. 2017;67:1265-­1273. an employee of Boehringer Ingelheim but was not when the manu- 16. Vilar-­Gomez E, Calzadilla-­Bertot L, Wai-­Sun Wong V, et al. Fibrosis severity as a determinant of cause-­specific mortality in patients script was written. Elisabetta Bugianesi reports advisory board ac- with advanced nonalcoholic fatty liver disease: a multi-­national co- tivity for Boehringer Ingelheim, Bristol Myers Squibb, Gilead, Innova, hort study. Gastroenterology. 2018;155:443-­457.e417. Intercept, Novo Nordisk. 17. Dulai PS, Singh S, Patel J, et al. Increased risk of mortality by fibro- sis stage in nonalcoholic fatty liver disease: systematic review and meta-­analysis. Hepatology. 2017;65:1557-­1565. R E S E A R C H E T H I C S A N D PAT I E N T C O N S E N T 18. Marengo A, Jouness RI, Bugianesi E. Progression and natural his- Not applicable. tory of nonalcoholic fatty liver disease in adults. Clin Liver Dis. 2016;20:313-­324. ORCID 19. Taylor RS, Taylor RJ, Bayliss S, et al. Association between fibrosis stage and outcomes of patients with nonalcoholic fatty liver dis- Jesús Rivera-­Esteban https://orcid.org/0000-0003-4357-8817 ease: a systematic review and meta-­analysis. Gastroenterology. 2020;158:1611-­1625.e1612. REFERENCES 20. Flemming J, Djerboua M, Booth C, et al. Increasing burden of cir- 1. Estes C, Anstee QM, Arias-­Loste MT, et al. Modeling NAFLD dis- rhosis projected to 2040 in Canada: implications for prevention, ease burden in China, France, Germany, Italy, Japan, Spain, United screening, and management. J Hepatol. 2020;73:S45. Kingdom, and United States for the period 2016–­2030. J Hepatol. 21. Axley P, Ahmed Z, Arora S, et al. NASH Is the most rapidly growing 2018;69:896-­904. etiology for acute-­on-­chronic liver failure-­related hospitalization 2. Matteoni C, Younossi Z, Gramlich T, Boparai N, Liu Y, Mccullough A. and disease burden in the United States: a population-­based study. Nonalcoholic fatty liver disease: a spectrum of clinical and patho- Liver Transpl. 2019;25:695-­705. logical severity. Gastroenterology. 1999;116:1413-­1419. 22. Pembroke T, John G, Michalak O, et al. The progression of liver dis- 3. Dam-­L arsen S, Franzmann M, Andersen IB, et al. Long term prog- ease in 69,290 individuals in Wales from 1999–­2019: tracking the nosis of fatty liver: risk of chronic liver disease and death. Gut. evolution of liver disease. J Hepatol. 2020;73:S46. 2004;53:750-­755. 23. U.S. Department of Health and Human Services Food and Drug 4. Adams LA, Lymp JF, St Sauver J, et al. The natural history of non- Administration Center for Drug Evaluation and Research (CDER). alcoholic fatty liver disease: a population-­based cohort study. Nonalcoholic steatohepatitis with compensated cirrhosis: devel- Gastroenterology. 2005;129:113-­121. oping drugs for treatment guidance for industry. https://www. 5. Ekstedt M, Franzén LE, Mathiesen UL, et al. Long-­term follow-­up fda.gov/media/​12773​8/download. Published June 2019. Accessed of patients with NAFLD and elevated liver enzymes. Hepatology. June 5, 2019. 2006;44:865-­873. 24. Garcia-­Tsao G, Friedman S, Iredale J, Pinzani M. Now there are many 6. Sanyal AJ, Banas C, Sargeant C, et al. Similarities and differences in (stages) where before there was one: in search of a pathophysiolog- outcomes of cirrhosis due to nonalcoholic steatohepatitis and hep- ical classification of cirrhosis. Hepatology. 2010;51:1445-­1449. atitis C. Hepatology. 2006;43:682-­689. 25. de Franchis R. Expanding consensus in portal hypertension: report 7. Söderberg C, Stål P, Askling J, et al. Decreased survival of sub- of the Baveno VI consensus workshop: stratifying risk and individu- jects with elevated liver function tests during a 28-­year follow-­up. alizing care for portal hypertension. J Hepatol. 2015;63:743-­752. Hepatology. 2010;51:595-­602. 26. Abraldes JG, Garcia-­Tsao G. The design of clinical trials in portal 8. Bhala N, Angulo P, van der Poorten D, et al. The natural his- hypertension. Semin Liver Dis. 2017;37:73-­8 4. tory of nonalcoholic fatty liver disease with advanced fibrosis 27. D'Amico G, Morabito A, D'Amico M, et al. Clinical states of cirrhosis or cirrhosis: an international collaborative study. Hepatology. and competing risks. J Hepatol. 2018;68:563-­576. 2011;54:1208-­1216. 28. D'Amico G, Garcia-­Tsao G, Pagliaro L. Natural history and prog- 9. Younossi ZM, Stepanova M, Rafiq N, et al. Pathologic criteria for nostic indicators of survival in cirrhosis: a systematic review of 118 nonalcoholic steatohepatitis: interprotocol agreement and ability to studies. J Hepatol. 2006;44:217-­231. predict liver-­related mortality. Hepatology. 2011;53:1874-­1882. 29. Ripoll C, Bari K, Garcia-­Tsao G. Serum albumin can identify pa- 10. Sebastiani G, Alshaalan R, Wong P, et al. Prognostic value of non-­ tients with compensated cirrhosis with a good prognosis. J Clin invasive fibrosis and steatosis tools, hepatic venous pressure gradi- Gastroenterol. 2015;49:613-­619. ent (HVPG) and histology in nonalcoholic steatohepatitis. PLoS One. 30. Bosch J, Groszmann RJ, Shah VH. Evolution in the understanding of 2015;10:e0128774. the pathophysiological basis of portal hypertension: how changes 2008 | RIVERA-­E STEBAN et al in paradigm are leading to successful new treatments. J Hepatol. 49. Ampuero J, Aller R, Gallego-­Durán R, et al. Definite and indeterminate 2015;62:S121-­130. nonalcoholic steatohepatitis share similar clinical features and progno- 31. Ripoll C, Groszmann R, Garcia–­Tsao G, et al. Hepatic venous pres- sis: a longitudinal study of 1893 biopsy-­proven nonalcoholic fatty liver sure gradient predicts clinical decompensation in patients with disease subjects. Liver Int. 2021. https://doi.org/10.1111/liv.14898 compensated cirrhosis. Gastroenterology. 2007;133:481-­488. 50. Thuluvath PJ, Kantsevoy S, Thuluvath AJ, et al. Is cryptogenic cir- 32. Sanyal AJ, Harrison SA, Ratziu V, et al. The natural history of ad- rhosis different from NASH cirrhosis? J Hepatol. 2018;68:519-­525. vanced fibrosis due to nonalcoholic steatohepatitis: data from the 51. Younossi Z, Stepanova M, Sanyal AJ, et al. The conundrum of cryp- simtuzumab trials. Hepatology. 2019;70:1913-­1927. togenic cirrhosis: adverse outcomes without treatment options. J 33. Iwakiri Y, Groszmann RJ. Vascular endothelial dysfunction in cirrho- Hepatol. 2018;69:1365-­1370. sis. J Hepatol. 2007;46:927-­934. 52. Eslam M, Sanyal AJ, George J, et al. International Consensus Panel. 34. Lens S, Rincón D, García-­Retortillo M, et al. Association between MAFLD: a consensus-­driven proposed nomenclature for metabolic severe portal hypertension and risk of liver decompensation in pa- associated fatty liver disease. Gastroenterology. 2020;158(7):1999-­ tients with hepatitis C, regardless of response to antiviral therapy. 2014.e1. Clin Gastroenterol Hepatol. 2015;13:1846-­1853.e1841. 53. Noureddin M, Chan JL, Barradas K, et al. Attribution of nonalco- 35. Mandorfer M, Kozbial K, Schwabl P, et al. Sustained virologic re- holic steatohepatitis as an etiology of cirrhosis for clinical trials el- sponse to interferon-­free therapies ameliorates HCV-­induced por- igibility: recommendations from the multi-­stakeholder liver forum. tal hypertension. J Hepatol. 2016;65:692-­699. Gastroenterology. 2020;159:422-­427.e421. 36. Mandorfer M, Kozbial K, Schwabl P, et al. Changes in hepatic ve- 54. Elkrief L, Rautou PE, Sarin S, Valla D, Paradis V, Moreau R. Diabetes nous pressure gradient predict hepatic decompensation in patients mellitus in patients with cirrhosis: clinical implications and manage- who achieved sustained virologic response to interferon-­free ther- ment. Liver Int. 2016;36:936-­948. apy. Hepatology. 2020;71:1023-­1036. 55. Orsi E, Grancini V, Menini S, Aghemo A, Pugliese G. Hepatogenous 37. Lens S, Baiges A, Alvarado-­Tapias E, et al. Clinical outcome and he- diabetes: is it time to separate it from type 2 diabetes? Liver Int. modynamic changes following HCV eradication with oral antiviral 2017;37:950-­962. therapy in patients with clinically significant portal hypertension. J 56. Han MAT, Altayar O, Hamdeh S, et al. Rates of and factors associ- Hepatol. 2020;73:1415-­1424. ated with placebo response in trials of pharmacotherapies for non- 38. Harrison SA, Abdelmalek MF, Caldwell S, et al. Simtuzumab is in- alcoholic steatohepatitis: systematic review and meta-­analysis. Clin effective for patients with bridging fibrosis or compensated cir- Gastroenterol Hepatol. 2019;17(4):616-­629.e26. rhosis caused by nonalcoholic steatohepatitis. Gastroenterology. 57. Ampuero J, Romero-­Gomez M. Stratification of patients in NASH 2018;155:1140-­1153. clinical trials: a pitfall for trial success. JHEP Rep. 2020;2(5):100148. 39. Iwakiri Y, Groszmann RJ. The hyperdynamic circulation of chronic 58. Fontaine KR, Williams MS, Hoenemeyer TW, Kaptchuk TJ, Dutton liver diseases: from the patient to the molecule. Hepatology. GR. Placebo effects in obesity research. Obesity (Silver Spring). 2006;43:S121-­S131. 2016;24(4):769-­771. 40. Villanueva C, Albillos A, Genescà J, et al. Development of hyperdy- 59. Glass O, Filozof C, Noureddin M, et al. Liver Forum Standard of namic circulation and response to β-­blockers in compensated cir- Care Working Group. Standardisation of diet and exercise in clinical rhosis with portal hypertension. Hepatology. 2016;63:197-­206. trials of NAFLD-­NASH: recommendations from the liver forum. J 41. Villanueva C, Albillos A, Genescà J, et al. β blockers to prevent Hepatol. 2020;73(3):680-­693. decompensation of cirrhosis in patients with clinically significant 60. Taylor-­Weiner A, Pokkalla H, Han L, et al. A machine learning approach portal hypertension (PREDESCI): a randomised, double-­blind, enables quantitative measurement of liver histology and disease moni- placebo-­controlled, multicentre trial. Lancet. 2019;393:1597-­1608. toring in NASH. Hepatology. 2021.https://doi.org/10.1002/hep.31750 42. Tripodi A, Anstee QM, Sogaard KK, et al. Hypercoagulability 61. Anstee QM, Lawitz EJ, Alkhouri N, et al. Noninvasive tests accu- in cirrhosis: causes and consequences. J Thromb Haemost. rately identify advanced fibrosis due to NASH: baseline data from 2011;9:1713-­1723. the STELLAR trials. Hepatology. 2019;70:1521-­1530. 43. Merli M. Nutrition in cirrhosis: dos and don'ts. J Hepatol. 62. Helmke S, Colmenero J, Everson GT. Noninvasive assessment of 2020;73:1563-­1565. liver function. Curr Opin Gastroenterol. 2015;31:199-­208. 44. Montano-­Loza AJ, Duarte-­Rojo A, Meza-­Junco J, et al. Inclusion 63. Gorowska-­Kowolik K, Chobot A, et al. (13)C Methacetin breath test of sarcopenia within MELD (MELD-­Sarcopenia) and the predic- for assessment of microsomal liver function: methodology and clin- tion of mortality in patients with cirrhosis. Clin Transl Gastroenterol. ical application. Gastroenterol Res Pract. 2017;2017:7397840. 2015;6:e102. 64. Razlan H, Marzuki NM, Tai M-­L S, et al. Diagnostic value of the 13C 45. Solà E, Pose E, Campion D, et al. Endpoints and design of clinical methacetin breath test in various stages of chronic liver disease. trials in patients with decompensated cirrhosis: position paper of Gastroenterol Res Pract. 2011;2011. the LiverHope Consortium. J Hepatol. 2021;74(1):200-­219. 46. Orr JG, Homer T, Ternent L, et al. Health related quality of life in people with advanced chronic liver disease. J Hepatol. How to cite this article: Rivera-­Esteban J, Armandi A, Augustin 2014;61(5):1158-­1165. S, Bugianesi E. Outcomes and potential surrogate markers for 47. Younossi ZM, Golabi P, Henry L. A comprehensive review of patient-­reported outcomes in patients with chronic liver diseases. future clinical trials of non-­alcoholic steatohepatitis cirrhosis. J Clin Gastroenterol. 2019;53(5):331-­3 41. Liver Int. 2021;41:1999–­2008. https://doi.org/10.1111/ 48. Tandon P, Walling A, Patton H, Taddei T. AGA clinical practice up- liv.15013 date on palliative care management in cirrhosis: expert review. Clin Gastroenterol Hepatol. 2021;19(4):646-­656.e3.