The American Association for the Study of Liver Diseases (AASLD) virtual meeting featured new, innovative and interesting presentations on a wide breadth of hepatology topics that were covered. I would like to highlight some of the key research and presentations pertaining to Nonalcoholic Fatty Liver Disease (NAFLD) and Nonalcoholic Steatohepatitis (NASH) that were discussed during the Liver Meeting 2020. However, with the overwhelming number of presentations that were given on the topic, this discussion is by no means exhaustive.

First and foremost, NAFLD is widely known as the most prevalent form of chronic liver disease in the United States and worldwide. Recently, it has had a proposed name change to metabolic-dysfunction associated liver disease (MAFLD) to help emphasize its association with metabolic syndrome and alcohol use in the presence of hepatic steatosis. There was variable adaptation to the terminology throughout the meeting, and it will be interesting to see whether it picks up future momentum.  

A hot topic of debate during the meeting was whether clinical trial endpoints should target NASH resolution or fibrosis improvement. In a seminal lecture, Arun Sanyal, MD, argued that targeting steatosis in NASH trials is crucial, as NASH ultimately impacts fibrosis and improves clinical outcomes. As activity of NASH increases, fibrosis increases. However, it is known that improvement of fibrosis takes time. His worthy adversary, Rohit Loomba, MD, argued that NASH activity is considered a less precise measure requiring larger sample sizes compared to fibrosis measurement, which is more reliable, precise, and needs only a modest sample size. Additionally, the definitions of NASH resolution are in flux, making it difficult to interpret and compare amongst other trials.

Experts widely discussed noninvasive tests that are alternative measures to liver biopsy, and can actually be reliably used to monitor disease progression and assess response to therapies in clinical drug trials. Stephen Harrison, MD, discussed the variability of liver biopsy findings, and how this precludes us from finding a consistent measure for NASH related trial endpoints such as NASH activity or fibrosis. This necessitates more reliable measures, including artificial intelligence and machine learning, some of which use comprehensive NASH panels. This is an exciting area of research that is growing, and may allow for more accurate standardized assessment of liver histology in clinical trials, and maybe clinical practice, in the future.

In addition, experts highlighted several evolving noninvasive fibrosis scores including, but not limited to, FIB-4 index score, enhanced liver fibrosis (ELF) score, as well as newer scores such as FAST, NIS4, PRO-C3 and ADAPT, all of which accurately identify advanced fibrosis. FAST score in particular, which is a combination of fibroscan plus aspartate aminotransferase (AST) level, has particularly performed well in derivation and several external validation cohorts. It can easily be calculated using an App called MyFibroScan (Echosens). Well known nonserologic tests such as transient elastography and magenetic resonance elastography also continue to evolve in their efficacy. Similarly, new sequential algorithms are becoming more popular in clinical trial settings and show promise, particularly when serologic tests are in the ‘indeterminate’ zones of measurement. Lastly, MRI-PDFF response in NASH may start being used in trials to monitor response to drug therapies as a noninvasive surrogate given its increasing utility in monitoring liver fat.

Treatments for NASH are also promising. Therapies attacking liver fat are particularly showing promise. Drugs that are currently ongoing investigation include Cenicriviroc, obeticholic acid, Resmetirom, and Aramchol. Semaglutide showed a 67% response for NASH resolution compared to placebo, which was 23%, and had a 48% improvement in fibrosis compared to a 34% placebo. Interestingly, the placebo response rates varied in these trials. This may signify the observer variability in path specimen. This variability increasingly highlights the need for noninvasive methods, such as artificial intelligence, to provide more consistent standardized scores in measuring important clinical endpoints in NASH. The future of this remains exciting. Lastly, combination therapies targeting different mechanisms of action are also likely to surface in the future landscape of NASH therapies.