Routes to diagnosis for hepatocellular carcinoma patients: predictors and associations with treatment and mortality.

BACKGROUND: Hepatocellular carcinoma (HCC) incidence has increased rapidly, and prognosis remains poor. We aimed to explore predictors of routes to diagnosis (RtD), and outcomes, in HCC cases. METHODS: HCC cases diagnosed 2006-2017 were identified from the National Cancer Registration Dataset and linked to Hospital Episode Statistics and the RtD metric. Multivariable logistic regression was used to explore associations between RtD, diagnosis year, 365-day mortality and receipt of potentially curative treatment. RESULTS: 23,555 HCC cases were identified; 36.1% via emergency presentation (EP), 30.2% GP referral (GP), 17.1% outpatient referral, 11.0% two-week wait and 4.6% other/unknown routes. Odds of 365-day mortality was >70% lower via GP or OP routes than EP, and odds of curative treatment 3-4 times higher. Further adjustment for cancer/cirrhosis stage attenuated the associations with curative treatment. People who were older, female, had alcohol-related liver disease, or were more deprived, were at increased risk of an EP. Over time, diagnoses via EP decreased, and via GP increased. CONCLUSIONS: HCC RtD is an important predictor of outcomes. Continuing to reduce EP and increase GP and OP presentations, for example by identifying and regularly monitoring patients at higher risk of HCC, may improve stage at diagnosis and survival.

Quantifying the benefit of nonselective beta-blockers in the prevention of hepatic decompensation: A Bayesian reanalysis of the PREDESCI trial.

BACKGROUND AND AIMS: Beta-blockers have been studied for the prevention of variceal bleeding and, more recently, for the prevention of all-cause decompensation. Some uncertainties regarding the benefit of beta-blockers for the prevention of decompensation remain. Bayesian analyses enhance the interpretation of trials. The purpose of this study was to provide clinically meaningful estimates of both the probability and magnitude of the benefit of beta-blocker treatment across a range of patient types. APPROACH AND RESULTS: We undertook a Bayesian reanalysis of PREDESCI incorporating 3 priors (moderate neutral, moderate optimistic, and weak pessimistic). The probability of clinical benefit was assessed considering the prevention of all-cause decompensation. Microsimulation analyses were done to determine the magnitude of the benefit. In the Bayesian analysis, the probability that beta-blockers reduce all-cause decompensation was >0.93 for all priors. The Bayesian posterior hazard ratios (HR) for decompensation ranged from 0.50 (optimistic prior, 95% credible interval 0.27-0.93) to 0.70 (neutral prior, 95% credible interval 0.44-1.12). Exploring the benefit of treatment using microsimulation highlights substantial treatment benefits. For the neutral prior derived posterior HR and a 5% annual incidence of decompensation, at 10 years, an average of 497 decompensation-free years per 1000 patients were gained with treatment. In contrast, at 10 years 1639 years per 1000 patients were gained from the optimistic prior derived posterior HR and a 10% incidence of decompensation. CONCLUSIONS: Beta-blocker treatment is associated with a high probability of clinical benefit. This likely translates to a substantial gain in decompensation-free life years at the population level.

Accurate prediction of all-cause mortality in patients with metabolic dysfunction-associated steatotic liver disease using electronic health records.

INTRODUCTION AND OBJECTIVES: Despite the huge clinical burden of MASLD, validated tools for early risk stratification are lacking, and heterogeneous disease expression and a highly variable rate of progression to clinical outcomes result in prognostic uncertainty. We aimed to investigate longitudinal electronic health record-based outcome prediction in MASLD using a state-of-the-art machine learning model. PATIENTS AND METHODS: n = 940 patients with histologically-defined MASLD were used to develop a deep-learning model for all-cause mortality prediction. Patient timelines, spanning 12 years, were fully-annotated with demographic/clinical characteristics, ICD-9 and -10 codes, blood test results, prescribing data, and secondary care activity. A Transformer neural network (TNN) was trained to output concomitant probabilities of 12-, 24-, and 36-month all-cause mortality. In-sample performance was assessed using 5-fold cross-validation. Out-of-sample performance was assessed in an independent set of n = 528 MASLD patients. RESULTS: In-sample model performance achieved AUROC curve 0.74-0.90 (95 % CI: 0.72-0.94), sensitivity 64 %-82 %, specificity 75 %-92 % and Positive Predictive Value (PPV) 94 %-98 %. Out-of-sample model validation had AUROC 0.70-0.86 (95 % CI: 0.67-0.90), sensitivity 69 %-70 %, specificity 96 %-97 % and PPV 75 %-77 %. Key predictive factors, identified using coefficients of determination, were age, presence of type 2 diabetes, and history of hospital admissions with length of stay >14 days. CONCLUSIONS: A TNN, applied to routinely-collected longitudinal electronic health records, achieved good performance in prediction of 12-, 24-, and 36-month all-cause mortality in patients with MASLD. Extrapolation of our technique to population-level data will enable scalable and accurate risk stratification to identify people most likely to benefit from anticipatory health care and personalized interventions.

The Natural History of Advanced Chronic Liver Disease Defined by Transient Elastography.

BACKGROUND & AIMS: The clinical course of cirrhosis does not follow a predictable trajectory. Transient elastography (TE) is commonly used in clinical practice to diagnose liver fibrosis and increasingly to risk stratify patients. The aim of this study was to assess the natural history of advanced chronic liver disease (ACLD) defined by TE using electronic health record (EHR) data in a multistate framework. METHODS: TE data were collected between 2008 and 2019. Patients with a liver stiffness measurement (LSM) >10 kPa were included. Disease and procedure code information held in EHR was analyzed. Clinical events including decompensation, hepatocellular carcinoma (HCC), and death were identified. Outcomes were described in a multistate model using flexible parametric survival methods including LSM and the albumin bilirubin (ALBI) score. RESULTS: Three thousand and twenty eight patients were included. Median follow up was 3.1 years. LSM and ALBI were associated with the development of varices and decompensation, and ALBI, age, sex, and viral liver disease were associated with the development of HCC from the compensated state. The cumulative incidence of HCC before decompensation was low for patients with alcohol-related liver disease (3.8%) and nonalcoholic fatty liver disease (1.3%) at 5 years after TE. Importantly, death was predicted to occur before decompensation or HCC in most cases. CONCLUSIONS: Liver stiffness, ALBI score, and disease etiology are each associated with outcomes in a large contemporary cohort with ACLD. EHR data can be used to define clinical progression in these patients, facilitating large clinical effectiveness trials and cost-effectiveness analyses.

Extra-hepatic morbidity and mortality in alcohol-related liver disease: Systematic review and meta-analysis.

BACKGROUND: Alcohol use increases the risk of many conditions in addition to liver disease; patients with alcohol-related liver disease (ALD) are therefore at risk from both extra-hepatic and hepatic disease. AIMS: This review synthesises information about non-liver-related mortality in persons with ALD. METHODS: A systematic literature review was performed to identify studies describing non-liver outcomes in ALD. Information about overall non-liver mortality was extracted from included studies and sub-categorised into major causes: cardiovascular disease (CVD), non-liver cancer and infection. Single-proportion meta-analysis was done to calculate incidence rates (events/1000 patient-years) and relative risks (RR) compared with control populations. RESULTS: Thirty-seven studies describing 50 302 individuals with 155 820 patient-years of follow-up were included. Diabetes, CVD and obesity were highly prevalent amongst included patients (5.4%, 10.4% and 20.8% respectively). Outcomes varied across the spectrum of ALD: in alcohol-related fatty liver the rate of non-liver mortality was 43.4/1000 patient-years, whereas in alcoholic hepatitis the rate of non-liver mortality was 22.5/1000 patient-years. The risk of all studied outcomes was higher in ALD compared with control populations: The RR of death from CVD was 2.4 (1.6-3.8), from non-hepatic cancer 2.2 (1.6-2.9) and from infection 8.2 (4.7-14.3). CONCLUSION: Persons with ALD are at high risk of death from non-liver causes such as cardiovascular disease and non-hepatic cancer.

Regional variations in hepatocellular carcinoma incidence, routes to diagnosis, treatment and survival in England.

BACKGROUND: Hepatocellular carcinoma (HCC) incidence, management and survival across England were examined to determine if geographical inequalities exist. METHOD: 15,468 HCC cases diagnosed 2010-2016 were included. Age-standardised incidence rates, net survival and proportions receiving potentially curative treatment and presenting through each route to diagnosis adjusted for age at diagnosis, sex and area-based deprivation quintile, were calculated overall and by Cancer Alliance. RESULTS: HCC incidence rates increased in men from 6.2 per 100,000 in 2010 to 8.8 in 2016, and in women from 1.5 to 2.2. The highest incidence rates, found in parts of the North of England and London, were nearly double the lowest. The adjusted proportion presenting as an emergency ranged 27-41% across Cancer Alliances. Odds increased with increasing deprivation quintile and age. Only one in five patients received potentially curative treatment (range 15-28%) and odds decreased with increasing deprivation and age. One-year survival in 2013-2016 ranged 38-53%. CONCLUSION: This population-based, nationwide analysis demonstrates clear differences in HCC incidence, management and survival across England. It highlights socioeconomic-associated variation and the need for improvement in early diagnosis and curative treatment of HCC. This research should assist policymakers, service providers and clinicians to identify regions where additional training, services and resources would be best directed.

The Placebo Response in Randomized Trials in Nonalcoholic Steatohepatitis Simply Explained.

BACKGROUND & AIMS: Liver histology is the primary endpoint in phase III trials in nonalcoholic steatohepatitis (NASH). There is an appreciable response to placebo that confounds endpoint assessment. The aim of this study was to quantify contributors to the placebo response and its impact on liver fibrosis improvement. METHODS: Estimates of fibrosis improvement in placebo-treated participants were made using probabilistic simulation. Each simulated trial included 120 participants. Parameters considered in the model included sampling and observer variability, regression to the mean, and net fibrosis progression calibrated to reported trial outcomes. RESULTS: In large phase IIb and III trials, 22% of placebo-treated participants with fibrosis stage 2 or 3 NASH at baseline improved by at least 1 fibrosis stage with minimal net disease progression. Estimates of sampling and observer variability in simultaneous biopsy studies highlighted an imbalance where apparent fibrosis improvement was more likely than worsening. Using these estimates and known trial outcomes, net fibrosis progression was estimated at 0.05 stages per year. Simulations of the placebo response rate showed a rate of 22% with 80% of trials falling between 15 and 30%, in keeping with trials reported to date. Additional increases in observer variability further increased the placebo response. CONCLUSIONS: The analyses presented simply define the placebo response in liver fibrosis in trials in NASH in terms of sampling and observer variability, regression to the mean, and fibrosis progression. Factors relating to liver biopsy are largely unmodifiable, and the variation in placebo response rates, both simulated and observed, challenges the role of biopsy in trial endpoint assessment.

Treatment Candidacy for Pharmacologic Therapies for NASH.

Nonalcoholic steatohepatitis (NASH) has emerged as one of the important causes of cirrhosis and hepatocellular carcinoma, and over 50 therapeutic agents are in various phases of clinical development. Recently, obeticholic acid has achieved the interim histological endpoint of fibrosis improvement with no worsening of NASH in the phase 3 REGENERATE study, and now patients are being followed for long-term clinical outcomes. Several drugs are in Phase 3 trials with a goal to achieve conditional registration under the subpart H pathway by the United States Food and Drug Administration (FDA). It is thus timely to consider the current situation and the way ahead in the management of NASH. In this article, we review the natural history of nonalcoholic fatty liver disease, upcoming treatments for NASH and various assessments. Based on the current knowledge, we discuss what should be the target treatment population and whether noninvasive tests are ready to guide NASH treatments both for patient selection and evaluation of treatment response.

Trajectory of Serum Bilirubin Predicts Spontaneous Recovery in a Real-World Cohort of Patients With Alcoholic Hepatitis.

BACKGROUND AND AIMS: Alcoholic hepatitis (AH) is a severe condition with poor short-term prognosis. Specific treatment with corticosteroids slightly improves short-term survival but is associated with infection and is not used in many centers. A reliable method to identify patients who will recover spontaneously will minimise the numbers of patients who experience side effects of available treatments. METHODS: We analysed the trajectory of serum bilirubin concentration over the course of hospital admissions in patients with AH to predict spontaneous survival and the need for treatment. RESULTS: data from 426 patients were analysed. Based on bilirubin trajectory, patients were categorized into three groups: 'fast fallers' (bilirubin <0.8 x admission value at day 7), 'static' (bilirubin of >0.9 - <1.2 x admission value) and 'rapid risers' (bilirubin of ≥1.2 x admission bilirubin). Fast fallers had significantly better 90-day survival compared to other groups (log rank p < .001), and showed no benefit of corticosteroid therapy (OR for survival at 28 days of treatment, 0.94, 95% CI 0.06 - 8.41). These findings remained even amongst patients with severe disease based on initial DF, GAHS or MELD scores. CONCLUSIONS: We present an intuitive method of classifying patients with AH based on the trajectory of bilirubin over the first week of admission. It is complimentary to existing scores that identify candidates for corticosteroid treatment or assess response to treatment. This method identifies a group of patients with AH who recover spontaneously and can avoid corticosteroid therapy.

A Systematic Review of Animal Models of NAFLD Finds High-Fat, High-Fructose Diets Most Closely Resemble Human NAFLD.

BACKGROUND AND AIMS: Animal models of human disease are a key component of translational hepatology research, yet there is no consensus on which model is optimal for NAFLD. APPROACH AND RESULTS: We generated a database of 3,920 rodent models of NAFLD. Study designs were highly heterogeneous, and therefore, few models had been cited more than once. Analysis of genetic models supported the current evidence for the role of adipose dysfunction and suggested a role for innate immunity in the progression of NAFLD. We identified that high-fat, high-fructose diets most closely recapitulate the human phenotype of NAFLD. There was substantial variability in the nomenclature of animal models: a consensus on terminology of specialist diets is needed. More broadly, this analysis demonstrates the variability in preclinical study design, which has wider implications for the reproducibility of in vivo experiments both in the field of hepatology and beyond. CONCLUSIONS: This systematic analysis provides a framework for phenotypic assessment of NAFLD models and highlights the need for increased standardization and replication.

Administrative Coding in Electronic Health Care Record-Based Research of NAFLD: An Expert Panel Consensus Statement.

BACKGROUND AND AIMS: Electronic health record (EHR)-based research allows the capture of large amounts of data, which is necessary in NAFLD, where the risk of clinical liver outcomes is generally low. The lack of consensus on which International Classification of Diseases (ICD) codes should be used as exposures and outcomes limits comparability and generalizability of results across studies. We aimed to establish consensus among a panel of experts on ICD codes that could become the reference standard and provide guidance around common methodological issues. APPROACH AND RESULTS: Researchers with an interest in EHR-based NAFLD research were invited to collectively define which administrative codes are most appropriate for documenting exposures and outcomes. We used a modified Delphi approach to reach consensus on several commonly encountered methodological challenges in the field. After two rounds of revision, a high level of agreement (>67%) was reached on all items considered. Full consensus was achieved on a comprehensive list of administrative codes to be considered for inclusion and exclusion criteria in defining exposures and outcomes in EHR-based NAFLD research. We also provide suggestions on how to approach commonly encountered methodological issues and identify areas for future research. CONCLUSIONS: This expert panel consensus statement can help harmonize and improve generalizability of EHR-based NAFLD research.

Assessing the Time-Dependent Impact of Performance Status on Outcomes After Liver Transplantation.

BACKGROUND AND AIMS: Identifying how the prognostic impact of performance status (PS) differs according to indication, era, and time period ("epoch") after liver transplantation (LT) could have implications for selection and treatment of patients on the waitlist. We used national data from the United Kingdom and Ireland to assess impact of PS on mortality separately for HCC and non-HCC recipients. APPROACH AND RESULTS: We assessed pre-LT PS using the 5-point modified Eastern Cooperative Oncology Group scale and used Cox regression methods to estimate hazard ratios (HRs) that compared posttransplantation mortality in different epochs of follow-up (0-90 days and 90 days to 1 year) and in different eras of transplantation (1995-2005 and 2006-2016). 2107 HCC and 10,693 non-HCC patients were included. One-year survival decreased with worsening PS in non-HCC recipients where 1-year survival was 91.9% (95% confidence interval [CI], 88.3-94.4) in those able to carry out normal activity (PS1) compared to 78.7% (95% CI, 76.7-80.5) in those completely reliant on care (PS5). For HCC patients, these estimates were 89.9% (95% CI, 85.4-93.2) and 83.1% (95% CI, 61.0-93.3), respectively. Reduction in survival in non-HCC patients with poorer PS was in the first 90 days after transplant, with no major effect observed between 90 days and 1 year. Adjustment for donor and recipient characteristics did not change the findings. Comparing era, post-LT mortality improved for HCC (adjusted HR, 0.55; 95% CI, 0.40-0.74) and non-HCC recipients (0.48; 95% CI, 0.42-0.55), but this did not differ according to PS score (P = 0.39 and 0.61, respectively). CONCLUSIONS: Impact on mortality of the recipient's pretransplant PS is principally limited to the first 3 months after LT. Over time, mortality has improved for both HCC and non-HCC recipients and across the full range of PS.

Fibrosis progression rate in a systematic review of placebo-treated nonalcoholic steatohepatitis.

BACKGROUND & AIMS: Liver fibrosis is the critical determinant of liver-related outcomes in persons with nonalcoholic fatty liver disease. The rate that fibrosis develops determines the time taken to reach cirrhosis and consequent clinical outcomes. Estimates of the fibrosis progression rate (FPR) are uncertain having been defined in small observational series that rely largely on nonstandardised repeat biopsy in selected patients. The aim of this study was to evaluate the FPR in placebo-treated participants with nonalcoholic steatohepatitis (NASH) in randomised controlled trials (RCTs). METHODS: Systematic review and meta-analysis of RCTs in NASH with data on fibrosis change extracted. Calculated fibrosis progression rates were pooled in meta-analysis. The pooled estimate was then used to model the proportion of hypothetical cohorts starting with no fibrosis at the age of 30 who develop cirrhosis. RESULTS: A total of 35 trials including 1419 placebo-treated participants who underwent repeat liver biopsy were evaluated. Considering all trials, the overall FPR was 0.00 stages per year, increasing to 0.03 stages per year in both trials at low risk of bias and trials including >50 placebo-treated participants. This estimate was markedly lower than the value derived from previously pooled analyses of observational data. Using a FPR of 0.03 resulted in a substantial reduction in the proportion of patients developing cirrhosis compared with the FPR derived from observational studies (13% vs 28%). CONCLUSIONS: The FPR in placebo-treated participants in RCTs is lower than that described from observational data. Slower fibrosis progression predicts fewer persons with NASH will progress to cirrhosis than previously estimated.

Transjugular intrahepatic portosystemic stent-shunt in the management of portal hypertension.

These guidelines on transjugular intrahepatic portosystemic stent-shunt (TIPSS) in the management of portal hypertension have been commissioned by the Clinical Services and Standards Committee (CSSC) of the British Society of Gastroenterology (BSG) under the auspices of the Liver Section of the BSG. The guidelines are new and have been produced in collaboration with the British Society of Interventional Radiology (BSIR) and British Association of the Study of the Liver (BASL). The guidelines development group comprises elected members of the BSG Liver Section, representation from BASL, a nursing representative and two patient representatives. The quality of evidence and grading of recommendations was appraised using the GRADE system. These guidelines are aimed at healthcare professionals considering referring a patient for a TIPSS. They comprise the following subheadings: indications; patient selection; procedural details; complications; and research agenda. They are not designed to address: the management of the underlying liver disease; the role of TIPSS in children; or complex technical and procedural aspects of TIPSS.

Histological Endpoints for Nonalcoholic Steatohepatitis Trials: Lights and Shadows.

Nonalcoholic steatohepatitis (NASH) with liver fibrosis is an increasingly important cause of liver-related morbidity and mortality. A diagnosis of NASH can only be made using liver biopsy. Liver histology also forms the endpoint for the expedited licensing strategies that have been approved by regulators to allow patients with NASH access to treatment before the impact of these on clinical outcomes is known. Validation of these histological surrogate endpoints is critical for the ongoing development of new therapies for NASH. The use of liver biopsy to define both trial entry and endpoints raises questions about the use of treatments for NASH in practice when the effectiveness of treatment will likely depend, at least in part, on the use of histology for patient selection in the real world.

Natural history of histologically proven alcohol-related liver disease: A systematic review.

BACKGROUND & AIMS: To date, studies into the natural history of alcohol-related liver disease (ALD) have lacked long-term follow-up, large numbers of participants, or both. We performed a systematic review to summarise studies that describe the natural history of histologically proven ALD. METHODS: PubMed and Medline were searched for relevant studies according to pre-specified criteria. Data were extracted to describe the prevalence of ALD, histological progression of disease and mortality. Single-proportion meta-analysis was used to combine data from studies regarding rates of progression or mortality. RESULTS: Thirty-seven studies were included, reporting data from 7,528 participants. Amongst cohorts of hazardous drinkers, on average 15% had normal histological appearance, 27% had hepatic steatosis, 24% had steatohepatitis and 26% had cirrhosis. The annualised rates of progression of pre-cirrhotic disease to cirrhosis were 1% (0-8%) for patients with normal histology, 3% (2-4%) for hepatic steatosis, 10% (6-17%) for steatohepatitis and 8% (3-19%) for fibrosis. Annualised mortality was 6% (4-7%) in patients with steatosis and 8% (5-13%) in cirrhosis. In patients with steatohepatitis on biopsy a marked difference was seen between inpatient cohorts (annual mortality 15%, 8-26%) and mixed cohorts of inpatients and outpatients (annual mortality 5%, 2-10%). Only in steatosis did non-liver-related mortality exceed liver-specific causes of mortality (5% per year vs. 1% per year). CONCLUSIONS: These data confirm the observation that alcohol-related hepatic steatohepatitis requiring admission to hospital is the most dangerous subtype of ALD. Alcohol-related steatosis is not a benign condition as it is associated with significant risk of mortality. LAY SUMMARY: Knowledge of the natural history of a disease allows clinicians and patients to understand the risks that are associated with a medical condition. In this study we systematically gathered all the published data regarding the natural history of alcohol-related liver disease in people who had a liver biopsy. We used this data to define the prevalence of the disease, the annual risk of progression to cirrhosis and the annual risk of death at each stage of the disease.