PARIS – Clinicians were the target audience during the first day of presentations at the International Liver Congress (ILC), hosted by the European Association for the Study of the Liver (EASL). On Wednesday morning, EASL interactive research think tanks – a new format at this year's ILC – examined emerging studies in eight areas of importance to treating physicians. Topics ranged from identification of targets and pharmacotherapy in nonalcoholic steatohepatitis (NASH) to options and trends in managing advanced hepatocellular carcinoma (HCC), new directions in treating liver disease and the integration of care for HCV infection and substance use among people who use drugs.
As a lively session on treating chronic liver failure made clear, one challenge for biopharmas seeking to address the spectrum of liver diseases is that clinicians don't even agree on the best biomarkers to predict patient outcomes, let alone the role that new biomarkers might play in facilitating patient selection for clinical trials or, more broadly, assessing the potential for drug candidates to address unmet needs in the field.
Jonel Trebicka, head of the laboratory for fibrosis and portal hypertension in the department of internal medicine I at the University of Bonn, examined the value of existing biomarkers in stratifying patients with acute-on-chronic liver failure (ACLF) and predicting their outcome. Because the condition has a dynamic clinical course, standard measurements – the Child-Pugh score and Mayo Clinic Model for End-Stage Liver Disease (MELD), for instance – may offer little prognostic value in assessing the risk of death among patients with ACLF, with or without the presence of single organ failure, he pointed out.
The CLIF-C ACLF score for ACLF patients and CLIF-C AD score for non-ACLF patients with acute decompensation, developed by the EASL-CLIF Consortium, represent improved tools. Still, clinicians may gain more insight from monitoring individuals with ACLF for "relevant events" – intrahepatic insults such as the presence of hepatitis B virus and alcoholism as well as extrahepatic insults such infections, for example. Individuals who are younger, have ascites and alcoholic liver cirrhosis are more likely to develop ACLF, he said, yet nearly half of individuals with the disease present without a precipitating event, confounding researchers.
A growing body of evidence suggests that the presence of biomarkers such as anemia – which correlates with higher mortality in individuals with ACLF – hyponatremia and hepatic encephalopathy may offer directions for future research, according to Trebicka, especially when added to biomarkers that are becoming part of standard monitoring, such as increased cardiac output, increased hepatic resistance, increased portal pressure, reduction in effective blood volume and decreased renal perfusion.
Although existing tools may improve the predictability of survival for patients with ACLF, for now, none of them can predict the development of the disease, he added. Even surrogate markers of systemic inflammation, such as C-reactive protein and liver stiffness, though highly activated in individuals with ACLF, are less important as individual indicators than "the big picture" in those patients, Trebicka said.
Javier Fernández, a physician and researcher at the Hospital Clínic de Barcelona, observed in his presentation that similar limitations exist in the diagnosis and predictive nature of bacterial infections in chronic liver failure. More than half of individuals with decompensated cirrhosis will develop those infections, he said, but – while bacterial infections generally are associated with poorer prognosis, especially in patients with more severe ACLF – outcomes vary widely.
Improved predictability in such patients is important, according to Fernández, since his research suggests the presence of bacterial infection produces a fourfold increase in one-month mortality and a higher rate of death for those admitted to intensive care units, compared with chronic liver failure patients without those infections. The most prominent types of bacterial infections among hospitalized chronic liver failure patients include spontaneous bacterial peritonitis, urinary tract infection, pneumonia, bacteremia and skin and soft-tissue infection, he said, with the prevalence of multidrug-resistant organisms now exceeding 30 percent of infections in some regions of western Europe.
Although new tests are emerging to supplement serum CRP as the main biomarker of bacterial infection, all have limitations, Fernández said. He suggested the need for clinical studies on new biomarkers and early microbiological tests to guide antibiotic prescription and development.
'We have to work with patterns'
The situation isn't markedly different in seeking to identify the phenotypes of acute kidney injury (AKI) in advanced cirrhosis, according to Elsa Solà, a colleague of Fernández at the Hospital Clínic de Barcelona. Solà's research suggests that AKI is present in more than half of individuals with cirrhosis who are hospitalized, although one-third of those patients present with AKI, which also is associated with a diminished prognosis.
One refinement may help to elucidate more predictability in outcomes for these patients. A proposed modification in the definition of AKI stage 1 in patients with cirrhosis will create additional stratification into subcategories known as AKI stage 1A (serum creatine, or SCr <1.5 mg/dL) and 1B (SCr >1.5 mg/dL). Because patients with advanced cirrhosis may develop AKI due to different causes, differential diagnosis is essential, Solà said.
Finally, Sara Montagnese of the University of Padova, reported on potential biomarkers to improve clinical and testing strategies in hepatic encephalopathy (HE). Montagnese acknowledged that biomarkers in HE represent a "major unmet need," with more weaknesses than strengths in the existing tools, including "no clarity in outcomes, no consistent or widespread use of the definition and classification and a syndromic rather than pathophysiological approach."
In HE, Montagnese said, biomarkers are associated with physiological or pathophysiological processes, such as encephalitis and septic delirium, but marking a syndrome such as a mental confusion "is a much more difficult task."
Montagnese predicted prospects for the increased use of oral amino acid challenges, imaging biomarkers, microbiota and metabolomic studies in assessing HE, though she cautioned against disregarding well-known risk factors such as hepatic dysfunction, age, history of HE, suspicion of covert HE and ammonia levels.
The presentations prompted EASL program chair Mauro Bernardi, professor of internal medicine at Bologna University, to wonder aloud about prospects for the "perfect" biomarker.
Trebicka, while wishing for a marker that "doesn't cost anything and has 100 percent effectiveness," said the more realistic approach is to work with omics and machine learning to establish mechanisms and signature patterns that predict of disease.
"ACLF is very heterogeneous, with different factors and etiologies, so we have to work with patterns rather than single biomarkers," he said.
"I've given up on" a single biomarker, Montagnese agreed, though she expressed interest in seeing combinations tested that would both offer specificity for and exclude the diagnosis of HE.
For Fernández, a "cheap bedside biomarker" in infectious disease would enable clinicians to diagnose fungal infections and offer a more prognostic approach for bacterial infections.
But as to the application of such emerging biomarkers for drug development or even patient stratification into clinical trials, the panel was stumped.
"The role of biomarkers is difficult to predict," Trebicka told BioWorld. "There will be a role, but it's hard to know where it will go."
The ILC continues through Sunday.