The noninvasive assessment of NAFLD and hepatic fibrosis is likely the greatest unmet need driving liver disease research right now. We have a huge population of patients—perhaps more than 100 million Americans—with NAFLD, but only approximately 12 to 18 million of these individuals have NASH and are thus at risk for progression to advanced fibrosis. 

Currently, the identification of NASH is based on its histologic features and requires a liver biopsy. For someone like me, who specializes in hepatology and is involved in NASH-related research, it is second nature to order, view, and interpret liver biopsies. However, biopsy is not widely used by or accessible to other physicians, nor is it appealing to patients.

We want to empower clinicians across multiple specialties to be comfortable with diagnosing and treating the disease while making the process acceptable to patients. Liver biopsy is invasive and unpalatable to many patients. Additionally, biopsy is an imperfect tool, with issues such as sampling variability and biologic variability. Different stages of disease may be observed in different parts of the same liver. Yet another limitation is interobserver variability, where 2 skilled and experienced pathologists may look at the same biopsy and reach different conclusions. US Food and Drug Administration–approved therapies for NASH are likely to emerge, and, when they do, we will need to think about how the diagnosis of NASH with fibrosis will be made in the community setting.

In summary, NASH is defined by biopsy, but the biopsy itself is not a good tool for the reasons previously outlined. Therefore, there is a need for noninvasive tools to differentiate probable NASH from not NASH, and, more importantly, to differentiate advanced fibrosis from earlier stages of the disease. At The Liver Meeting 2021, the Non-Invasive Biomarkers of Metabolic Liver Disease (NIMBLE) abstract examined different proprietary and routine tools for the noninvasive assessment of NASH and fibrosis (abstract LO1). I had the privilege of being one of the coauthors of this abstract, and it was presented by Arun J. Sanyal, MD. The goal of this study was to assess various noninvasive “liquid” assays with rigor and reproducibility to determine which tests may be the most appropriate to answer clinical questions. 

The Fibrosis-4 (FIB-4) index has performed quite well and is particularly suitable to automation and integration into electronic medical records. Discussion has also centered around the Enhanced Liver Fibrosis (ELF) test, which was recently approved in the United States. The ELF test has been shown to be effective in correlating with advanced fibrosis and in potentially predicting clinical events in the future. 

We also have a combination of imaging-based and “liquid” biomarker tests. Imaging-based tools include vibration-controlled transient elastography and magnetic resonance elastography (abstract 75). There are other technologies that are currently being developed to assess fat and fibrosis in the liver in different ways, including novel approaches using protease-based liquid biopsy (abstract LO3) and thrombospondin-2 as biomarkers (abstract 73). I think that there will be a role for both imaging-based and liquid-based biomarkers in the future. 

When considering who to test for possible NASH with fibrosis, simple clinical heuristics are quite useful, including the consideration of ethnic or racial backgrounds associated with higher risk, the presence of type 2 diabetes, and elevated liver enzymes. However, use of a single marker may be misleading. For example, while elevated serum alanine aminotransferase is predictive of NAFLD in patients with diabetes, this test may not be elevated in all patients with diabetes and NAFLD. So, if we only pursue labs that are abnormal, we may neglect the patients who have disease but do not have laboratory abnormalities. The combination of a clinical heuristic and some type of biomarker that can stratify patients into lower or higher risk categories is clearly needed and will be beneficial.

Whether you are using the ELF test, PRO-C3, or FIB-4, I would say that you will have a greater likelihood of accomplishing the task of identifying patients who are at risk. If you have a patient who has features that suggest NAFLD and who has multiple risk factors for NASH, some type of diagnostic tool is helpful. Transient elastography and a liquid test such as FIB-4 or the ELF test are the tests that are performing fairly well thus far in the NIMBLE cohort.

I think that there is a great deal of interest within the liver community in the development of noninvasive tools to detect advanced fibrosis in NASH. Although we are still working through the regulatory process to bring new therapies to the clinic, from a societal and public health perspective, we need to focus on the fact that a multidisciplinary approach is required.