Editorial: Silent But Detectable—High‐Throughput Proteomics to Identify Clinically Significant Liver Disease

A major challenge in hepatology is that progressive liver disease is often silent until the onset of major adverse liver outcomes (MALO) such as ascites, variceal haemorrhage and hepatocellular carcinoma—events that carry substantial morbidity and mortality [1]. In this issue, the authors utilise proximity extension assay (PEA)-based high-throughput proteomics to risk-stratify patients for clinically significant liver disease and evaluate whether a novel plasma-derived proteomic score can perform more effectively than widely used serum scores [2]. The investigators analysed > 2900 plasma proteins in ~52,000 UK Biobank participants with > 10 years of follow-up and ultimately derived a five-protein score (ITGBL1, ADAMTSL2, IGFBP7, KRT18 and ALDH1A1) that predicted incident MALOs with strong discrimination (AUROC ~0.83 in the test set). This five-protein score (‘PEA score’) outperformed widely used serum indices AST-to-platelet ratio index (APRI, AUROC ~0.72) and Fibrosis-4 (FIB-4, AUROC ~0.73). The PEA score also remained stable in higher-risk subgroups of participants with obesity and type 2 diabetes mellitus. A key strength of this study is external validation in two clinically distinct populations: severe alpha-1 antitrypsin deficiency [3] (n = 287) and people living with HIV (n = 960), using transient elastography-derived liver stiffness measurement (LSM) as a fibrosis surrogate, where all five components increased across fibrosis stages and the PEA score continued to outperform APRI and FIB-4 for prediction of MALOs. The three independent cohorts had strong overlap in their proteomic signature, which included three markers linked to hepatic stellate cell activation (fibrogenesis) and two markers of epithelial injury. This mechanistic coherence—emerging from an unbiased high-throughput survey—strengthens confidence that the proteomics signal reflects true fibrosis biology and supports the potential for broad clinical applicability across heterogeneous at-risk patient populations. Important caveats remain. The cohorts are predominantly of European ancestry; performance appears less robust in minimal fibrosis stages; applicability to cholestatic and non-hepatocellular diseases is untested; and cost considerations will be an important factor for real-world scalability. Incorporating liver biopsy as a reference standard, as opposed to LSM as a surrogate marker, in future studies could further strengthen the validation of this prognostic proteomic score, albeit we certainly acknowledge the broader shift within the field of hepatology toward biopsy-free approaches [4, 5]. In summary, we commend the authors on this persuasive proof-of-concept study which demonstrates that standardised affinity proteomics can yield mechanism-anchored biomarkers for detection of clinically significant liver disease, thus facilitating the potential for timely clinical interventions in those patients at highest risk for MALOs [1]. These results support future efforts to prospectively validate proteomic technologies in diverse populations and clinical care pathways. Suzanne Sharpton: supervision, writing – review and editing, conceptualization. Emily Stephenson: writing – original draft. The authors have nothing to report. Suzanne Sharpton serves on an advisory board for Takeda Pharmaceuticals and her institution has received research funding from Takeda, AstraZeneca and Boehringer Ingelheim. This article is linked to Balasubramani et al. paper. To view this article, visit https://doi.org/10.1111/apt.70656. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.