We are learning that fibrosis is not that simple. There are genetic influences in NASH, such as PNPLA3. And we now understand that fibrosis is a very dynamic process, a function of both fibrogenesis and fibrosis resolution. The scar tissue that you see under the microscope today is not going to be the same scar tissue that you will see in the future. PRO-C3 is likely the best-known biomarker in terms of fibrogenesis, or the fibrogenic drive of the disease.

Numerous markers of fibrosis are more static, reflecting the amount of scar tissue that is present at a specific point in time. We have always used fibrosis as a surrogate for the risk of developing clinical decompensation, so the more scar tissue the patient has, the more likely it is that they will develop a liver event. Of these static markers, among the best known are the AST to Platelet Ratio Index (APRI) and the Fibrosis-4 (FIB-4) index. The FIB-4 performs well across different causes of liver disease as a marker of liver fibrosis, and it is easy to use. It can play a very important role in clinical practice in ruling out significant scarring because, if a patient’s FIB-4 is low (eg, <1.3), then they have a very low probability of having significant fibrosis, and a focus primarily on weight management may be reasonable. If a patient’s FIB-4 is higher than 2.67, then you might want to refer them elsewhere for further assessment (ie, to someone with more invasive tools for confirmation) because they are likely to have advanced fibrosis and to experience a clinical outcome. When FIB-4 results fall in the middle, or when a vibration-controlled transient elastography (eg, FibroScan [Echosens]) and the FIB-4 give you discrepant results, a consideration of risk factors and other tests may be useful.

As relates to the use of tests in the real-world setting, I believe that we have a great opportunity to develop common practice guidelines with other nonhepatology stakeholders, much like what was done in the past with viral hepatitis. The American Association for the Study of Liver Diseases is beginning to move in that direction, which is a healthy sign for the field.

We also need regulatory approvals. There are a lot of biomarkers that are under development, but very few have been approved by the US Food and Drug Administration, although the Enhanced Liver Fibrosis (ELF) test (Siemens Healthineers), which is the extended liver fibrosis test, has recently been approved as a prognostic biomarker. For fibrosis biomarkers to meet the regulatory threshold for approval, we need to generate the data, and that is the goal of the Non-Invasive Biomarkers of Metabolic Liver Disease (NIMBLE) program. We think that in the next 2 to 3 years, we will have addressed and generated the evidence to provide a slew of biomarkers for liver fibrosis, both circulating biomarkers and imaging biomarkers that use ultrasound and magnetic resonance–based tools.

I emphatically agree that noninvasive markers are the future. We are clearly moving away from liver biopsy, which we once held as the gold standard. However, it is inconvenient, there is a risk involved with performing the procedure, and both patients and providers are increasingly reluctant to travel this path. Additionally, a small portion of the liver sampled may not tell us the whole story because of the patchy nature of fibrosis and steatohepatitis. For these reasons, I think of the liver biopsy as being more of a bronze rather than a gold standard.

There is the potential for accurate biomarkers, whether they are serum based or imaging based. Compared with liver biopsy, by being more integrative, these tests may even improve upon our assessment and accuracy of staging metabolic fatty liver disease and the extent of associated fibrosis. I think that these tools will be liberating in the settings of primary care, internal medicine, endocrinology, and cardiology insomuch as they will allow clinicians to identify the patients who should receive more specialized hepatology care.

Through the concerted efforts of the NIMBLE consortium and others like it, I have confidence that we will have not only excellent staging tools but also, potentially, markers that can dynamically reflect responses to inhibitors and other agents that are designed to slow fibrogenesis or to enhance fibrolysis. Dynamic markers could tell us about the effects of antifibrotic agents on those markers of fibrogenesis. Through the use of such easily obtainable biomarkers, we could tailor our interventions to the disease process and to assess the state of inhibition of the drivers of disease progression. Calibrating therapy for our patients would be a very attractive concept.

There has been a huge amount of data generated in the last decade on several noninvasive markers to detect liver fibrosis stage, monitor patients longitudinally, and predict outcomes in patients with NAFLD. Liver biopsies are fraught with sampling errors, are costly, and have the potential for complications. Thus, biopsy is not optimal, and these noninvasive tests therefore meet an important clinical unmet need.

Tests that incorporate markers of fibrosis can be categorized as patented or nonpatented tests. Nonpatented tests do not require laboratory assessment other than what we typically use in the care of patients with NAFLD. Examples of these tests include the FIB-4, the NAFLD fibrosis score, and the BARD score. Patented tests require additional laboratory testing. Of several such tests, the most commonly used are FibroMeter (Echosens), FibroSure (Labcorp), and the ELF test.  

It is well recognized that fibrosis is the most important determinant for liver disease and even non–liver disease–related outcomes in NASH and NAFLD. Studies have demonstrated that many noninvasive markers are associated with developing liver-related outcomes and mortality, and even cardiovascular-related disease outcomes in patients with NAFLD. Therefore, when we see a patient with metabolic-associated fatty liver disease, our first aim is to determine the risk of liver fibrosis, typically using a nonpatented test such as the FIB-4, which has a very good negative predictive value. When in doubt, or when screening results are indeterminant, additional tests such as FibroScan are used.