Reply to: “Are FIB-4 and liver stiffness measurement interchangeable for HCC risk stratification in MASLD?” Authors: Georg Semmler, Jan Embacher, Mattias Mandorfer Published: 2026 To the Editor: We thank Drs Zhou and Meng for their interest in our study and fully agree that metabolic dysfunction-associated steatotic liver disease (MASLD) deserves particular attention when considering risk-based hepatocellular carcinoma (HCC) surveillance.1,2 They argue that our study cannot provide a definitive conclusion on the interchangeability of FIB-4 and liver stiffness measurement (LSM) for guiding HCC surveillance in MASLD due to the limited sample size (n = 1,790) and low event rate (n = 6) in this subgroup, as well as the absence of LSM data in the UK biobank, which we fully acknowledge. Furthermore, they discuss the discordance between FIB-4 and LSM and the known impact of age on the accuracy of FIB-4 for fibrosis assessment, as well as the ongoing controversy regarding HCC in patients with MASLD without cirrhosis. Notably, most of their reasoning is based on the reliance on fibrosis as an endpoint. However, it is increasingly acknowledged that fibrosis is merely a surrogate of risk. Notably, the paradigm is currently shifting from diagnosing cirrhosis to predicting clinical events (e.g. hepatic decompensation or HCC), and thus, on what matters to patients. Our study is meant to facilitate this transition, which we strongly encourage. Accordingly, our objective was not to establish FIB-4 as a direct substitute for LSM for diagnosing advanced fibrosis/compensated advanced chronic liver disease (cACLD) in MASLD. Even though data from tertiary care supports approaches to define cACLD (i.e. increased relative risk of complications) by FIB-4,3 the higher diagnostic accuracy of LSM vs. FIB-4 for fibrosis due to MASLD (e.g.4) remains undebated. Thus, FIB-4 is not meant to be used as a stand-alone NIT to diagnose advanced fibrosis in MASLD, but is applied as a triage tool within referral pathways.5 Indeed, our study addressed a profoundly different question: Whether NITs can predict the risk of HCC. While diagnostic performance for fibrosis and prognostic performance for endpoints such as HCC are related, they are not identical. An imperfect NIT for diagnosing histology-based or LSM-based fibrosis may still provide a reasonable prognostic estimate for future liver-related events including HCC. Notably, FIB-4 and LSM demonstrated comparable discrimination for HCC in our cohorts, which aligns with a considerable body of evidence for the endpoint ‘liver-related events’.6,7 However, this observation should not be misinterpreted as evidence that both tests are interchangeable across diagnostic or prognostic settings. The age component of FIB-4 reduces its diagnostic utility for fibrosis, in particular for the threshold of 1.3, which is lower than the 1.75/2.67 thresholds evaluated in our study.8,9 However, as extensively discussed in our manuscript, age is also one of the strongest determinants of HCC risk (e.g.10). Thus, the overestimation of fibrosis by surrogate markers may be of limited relevance when the endpoint is HCC. Importantly, our study does not imply that FIB-4 should be interpreted uncritically, but it supports its value for HCC prediction without requiring confirmation of advanced fibrosis/cACLD by LSM. Of note, the UK Biobank analysis provided a broader, non-specialist population enriched for metabolic risk, but lacked LSM. Therefore, the UK Biobank data cannot be interpreted as evidence of FIB-4/LSM equivalence in MASLD. Instead, the subgroup with metabolic risk, alcohol consumption, or pre-existing liver disease was intentionally analysed to examine whether FIB-4 identifies strata of HCC risk outside tertiary care, which was confirmed in our analysis. However, we agree with Zhou and Meng that FIB-4 should not be used as a stand-alone trigger for HCC surveillance in unselected MASLD populations. In these individuals, the low incidence of HCC limits the clinical utility and cost-effectiveness of any surveillance strategy. In summary, we concur that our findings should not be overinterpreted to claim interchangeability of FIB-4/LSM in MASLD. However, the limitations of FIB-4 as a diagnostic fibrosis test do not negate its potential prognostic value for HCC. Our study supports the concept that FIB-4 can provide clinically useful HCC risk stratification, particularly in chronic liver disease populations and settings where LSM is unavailable, while dedicated MASLD-specific event-based validation remains an important next step. Finally, we strongly encourage the removal of surrogate fibrosis from equations used to estimate HCC risk using NITs, as this adds a layer of complexity for purely historical reasons, without benefiting patients. Authors' contributions Drafting of the manuscript (all authors), critical revision of the manuscript for important intellectual content (all authors). Declaration of generative AI and AI-assisted technologies During the preparation of this work the authors used ChatGPT (OpenAI) for language editing and rephrasing, without contributing to the scientific content or generation of ideas. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. Financial support No financial support was received to produce this manuscript. Conflicts of interest G.S. received speaker fees from DiaSorin. J.E. has nothing to disclose. M.M. received grant support from Echosens, served as a consultant and/or advisory board member and/or speaker for Albrireo, AstraZeneca, Boehringer Ingelheim, Chiesi, Echosens, Eli Lilly, Ipsen, Madrigal, Takeda, and W. L. Gore & Associates and received travel support from Ipsen.