Skip to main content



Reporting period: 2020-04-01 to 2021-09-30

In 2013, cirrhosis was responsible for 1.2 million deaths worldwide. This mortality is mainly due to cirrhosis decompensation, i.e. development of ascites, hepatic encephalopathy, and/or gastrointestinal hemorrhage, and its progression to acute-on-chronic liver failure (ACLF). Patients with decompensated cirrhosis receive many treatments such as intravenous and oral absorbable antibiotics, oral non-absorbable antibiotics, albumin, proton-pump inhibitors, laxatives, diuretics, betablockers, vasoconstrictors, statins, anticoagulants, steroids and antiviral agents. Despite these multiple treatments, ACLF or mortality in patients with decompensation of cirrhosis remains high (15% at day 28, 28% at day 90) because of large interindividual variability in precipitating events, in clinical presentation and in response to treatment. This heterogeneity calls for treatment personalization according to underlying mechanisms.
The objective of DECISION is to enhance our understanding, at systems level, of the pathophysiology of decompensation of cirrhosis leading to ACLF or death to decrease patients’ mortality at day 28. First, DECISION will improve our knowledge of the pathophysiology of decompensation of cirrhosis by integrating results of high-throughput multi-omic profiling with comprehensive clinical data from large cohorts of fully characterized patients (more than 8,600 time points) with available standardized biological samples.
Second, we will identify novel combinatorial therapies for patients with decompensation of cirrhosis to prevent death. We will refine these therapies in new and/or optimized animal models and then test the best combination in high risk patients in a phase II clinical trial built in DECISION.
Third, we will develop 2 tests: one predicting outcome of patients with decompensation of cirrhosis when treated with standard treatment (prognostic test); and the other identifying patients who will respond to the novel combinatorial therapy (test for response).
To achieve a better understanding of acute decompensation of cirrhosis and to identify novel combinatorial therapies, DECISION builds on 3 large available prospective cohorts of patients with acute decompensation of cirrhosis, namely CANONIC (from Europe), PREDICT (from Europe) and ACLARA (from South America). From the beginning of DECISION, many advances have been made on the characterization and the analysis of these cohorts.
First, specialty laboratories of DECISION performed epigenomics, whole-blood transcriptomics, serum microRNA profiling, serum metabolomics analysis, plasma inflammatory protein (cytokines) and lipid (eicosanoids) mediators characterization and plasma extracellular vesicle concentration measurement in samples from ~1355 patients from the PREDICT and ACLARA studies, using state-of-the-art high-throughput technologies. These results will be merged with those of the CANONIC cohort, obtained outside DECISION. On top of that, single cell transcriptomics of blood cells, untargeted proteomic analysis of plasma extracellular vesicles and circulating miRNA next generation sequencing has been performed on blood samples from selected patients with acute decompensation of cirrhosis (n=23 to 65) to broaden our view of changes occurring in those patients.
In parallel, detailed clinical data of the patients included in CANONIC, PREDICT and ACLARA including medical history, clinical presentation, drugs taken and outcome, have been harmonized and merged into a unique database. Omics data are being integrated with these clinical data in this unique database. Once done, by the end of 2021, systems approaches will be initiated to better understand patient heterogeneity in acute decompensation of cirrhosis.
At the same time, DECISION’s teams started the analysis of the merged harmonized clinical databases to identify new combinatorial therapies. Teams identified the appropriate level of granularity for classification of drug used in DECISION cohorts, and the prevalence of each combination of drug in DECISION cohorts. Then, several approaches have been developed: safety-efficacy approach; machine learning approach; and comparison of expected vs. observed survival. These approaches will be refined with the inclusion of omics data in the next months.
Reliable animal models being key to test new drugs, strong efforts have been conducted by DECISION’s teams to refine existing and develop new rat models of decompensation of cirrhosis leading to ACLF and to characterize these models. Four different models of cirrhosis and 4 different precipitants have been compared back to back. Five out of these 16 combinations have been identified as particularly relevant to mimic the human situation and have been selected for very detailed characterization in the next months.
In order to disseminate of our findings to the widest possible audience, DECISION’s teams -with the help of the European Association for the Study of the Liver (EASL) and the European Liver Patients Association (ELPA)- conducted many dissemination actions towards scientists and physicians (e.g. organization of conferences, newsletters), patients (2 meetings with patients) and general audience (e.g. press releases, website, publication in journals for laymen).
The harmonized database we are obtaining including 2000 patients with acute decompensation of cirrhosis with detailed clinical data and with extensive multiomics characterization is unique worldwide. This will allow, in the next months, systems medicine approaches to provide an unprecedented understanding of the pathophysiology of acute decompensation of cirrhosis.
This database is also the basis for identification of novel combinatorial therapies for patients with decompensation of cirrhosis to prevent death and of novel tests predicting outcome of patients with decompensation of cirrhosis.
The extensive and detailed analysis of rat models of acute decompensation of cirrhosis we performed, and will continue to analyze, will be also a step forward in the field of cirrhosis since relevant models are lacking so far.
New pathophysiological concepts brought by the DECISION project