Beyond the baseline: Longitudinal surveillance with noninvasive tests for fibrosis in MASLD

Metabolic dysfunction–associated steatotic liver disease (MASLD) is the most prevalent cause of chronic liver disease worldwide. The clinical presentation and progression of MASLD are highly variable. While MASLD generally carries a lower risk of progression to cirrhosis compared to other liver diseases, its health care burden remains substantial. It is estimated that MASLD accounts for up to 38% of HCC cases, with projections suggesting that MASLD could become the leading cause of HCC, potentially leading to as many as 135,000 cases in the United States between 2015 and 2030.1,2 In this context, the need for precise tools to identify individuals at heightened risk for liver-related morbidity and mortality is critical. Currently, clinical practice is hindered by the absence of robust, validated methods to effectively stratify risk, particularly among patients with advanced fibrosis. This gap in risk stratification is crucial, as advanced fibrosis is a well-recognized surrogate for the development of metabolic-associated liver outcomes (MALOs), including cirrhosis and HCC. One of the most pressing challenges in MASLD management is the early and accurate identification of patients who are at significant risk of progressing to advanced fibrosis. In patients with MASLD, the stage of liver fibrosis has been strongly linked to all-cause mortality, including MALOs.3 This is particularly important in light of recent FDA approval of therapy specifically targeting patients with MASLD and stage 2–3 fibrosis.4 With anticipation of advancement in treatment options ahead, their successful application in clinical practice hinges on the ability to accurately identify appropriate candidates with advanced fibrosis. Given the impracticality of routine liver biopsy in real-world practice, noninvasive tests (NITs) for liver fibrosis are increasingly being integrated into screening and surveillance algorithms for patients with MASLD. NITs for liver fibrosis, such as the FIB-4 index, play a crucial role in risk stratification for HCC surveillance in patients with MASLD. Elevated FIB-4 scores have been consistently associated with an increased risk of HCC and other liver-related adverse events.3 Current guidelines endorse the use of serum and imaging-based NITs, including FIB-4, for assessing fibrosis in patients with MASLD.3 The American Association for the Study of Liver Diseases recommends longitudinal assessment of liver fibrosis using a tiered approach to capture those with advanced fibrosis.3 However, comprehensive studies assessing the efficacy of these risk stratification tools for MALO are lacking. In this issue of Hepatology, Shang et al5 studied a cohort of patients without cirrhosis and with MASLD and evaluated whether longitudinal changes in FIB-4 were predictive of MALO. Among the 1260 patients, 111 (8.8%) developed MALOs, with 85.7% of those having an intermediate (FIB-4 1.3–2.67) or high-risk FIB-4 (FIB-4>2.67) at any point. The risk of developing MALO was significantly elevated in patients with baseline intermediate (HR: 2.89, CI: 1.9–4.41) and high-risk FIB-4 (HR: 6.74, CI: 4.03–11.3). Longitudinal analysis revealed that a doubling of FIB-4 at any point increased the risk of MALO by 2.60-fold (CI: 1.89–3.50) despite adjusting for confounders such as age, sex, BMI, type 2 diabetes, hyperlipidemia, and fibrosis stage. Similar associations were noted with a doubling of AST (adjusted hazard ratio: 2.69, CI: 2.57–3.05) and a 10 Giga/L reduction in platelets (adjusted hazard ratio: 0.93, CI: 0.90–0.97). However, the rate of FIB-4 change itself did not significantly influence MALO risk (adjusted hazard ratio: 1.04; CI: 0.67–1.61). The authors conclude that an elevated FIB-4 at any time point is a strong predictor of MALO, while the velocity of change in FIB-4 is less relevant. This study offers several strengths, including a relatively large sample size of patients without cirrhosis and with MASLD with up to 20 years of longitudinal follow-up. The use of the Swedish Cancer and Causes of Death Registrars enhanced the capture of MALO incidents. However, generalizability is limited, as the population was restricted to three Swedish university hospitals. Additionally, while the sample size is substantial, only 8.8% (n=111) developed MALO, with chronic liver failure documented in 29.7% (n=33) of those cases—an outcome that was not clearly defined, potentially leading to misclassification and information bias. HCC was diagnosed in only 14.4% (n=16) of MALO cases. Despite these limitations, Shang and colleagues provide valuable data on the utility of FIB-4 monitoring in patients with MASLD. Their findings contribute to the growing evidence supporting FIB-4 surveillance for risk stratification in MASLD. Approximately one-third of patients with initially low FIB-4 developed MALO, and an elevated FIB-4 at any time was associated with an increased risk of MALO.5 This aligns with our prior study involving 202,319 Veterans Administration patients with MASLD, where progression from low-risk to high-risk FIB-4 resulted in a 17.1-fold increased risk of cirrhosis or HCC (HR: 17.09, CI: 13.77–21.20).6 Furthermore, although regression from high-risk to low-risk FIB-4 reduced the incidence of cirrhosis and HCC, the risk remained elevated compared to those consistently at low risk. These findings underscore that singular FIB-4 screening is inadequate for monitoring patients with MASLD, supporting the need for longitudinal NIT surveillance to identify those at risk for advanced fibrosis and MALO. Current guidelines recommend repeating FIB-4 testing every 1–3 years for those at low risk of advanced fibrosis.3 Given the accumulating evidence that elevated FIB-4 levels are associated with MALO, more frequent monitoring may be warranted for patients with a history of elevated FIB-4. Since patients with FIB-4 values below 1.3 are often managed in primary care, this approach could guide timely hepatology referrals and tailor screening protocols. Additionally, the use of FIB-4 could help reduce costs by identifying patients who truly need further imaging or biopsy. While NITs offer significant potential, they remain emerging tools and are not yet fully validated or calibrated as risk predictors in MASLD. In general populations, FIB-4 can have a false-positive rate of up to 35%, particularly in individuals over 65, where age is a known confounder.7–9 Consequently, higher cutoffs have been proposed to differentiate between low and intermediate risk in this age group.7,8 Beyond FIB-4, serum NITs, such as enhanced liver fibrosis tests, as well as imaging NITs, including vibration-controlled transient elastography (VCTE) and magnetic resonance elastography, have shown promise in conjunction with FIB-4. One study involving 3378 patients found that using FIB-4 followed by enhanced liver fibrosis in indeterminate cases before transient elastography correctly classified 88% of cases and reduced false-negative rates compared to using FIB-4 or enhanced liver fibrosis alone.7 Another multicenter study of 3202 patients found that FIB-4, followed by VCTE, reduced indeterminate results from 43% to 20%.9 In a US study evaluating the cost-effectiveness of sequential FIB-4 with VCTE, magnetic resonance elastography, and liver biopsy, the combination of FIB-4 with VCTE had the best cost per correct diagnosis.10 Further cost-effective analyses are needed to assess the most appropriate tiered approach. While these emerging tools and algorithms offer promise, there remains a need for long-term data to validate their use in clinical practice. As NITs are increasingly adopted as secondary endpoints in clinical trials for MASLD, there will be opportunities to gather these data to provide insights into the long-term outcomes associated with different levels of fibrosis as measured by NITs, including the impact of therapeutic interventions on disease progression. Currently, clinicians must continue to rely on the best available evidence to guide their use of NITs in MASLD management. This includes being cautious in interpreting NIT results and using a multimodal approach to risk stratification.