Noninvasive Detection of Clinically Significant Fibrosis in Biopsy-Proven MASLD: A Retrospective Cross-Sectional Diagnostic Accuracy Study

Background and Objectives: Metabolic dysfunction-associated steatotic liver disease (MASLD) is associated with substantial morbidity, and hepatic fibrosis stage is the strongest predictor of long-term outcomes. Although liver biopsy remains the reference standard for fibrosis assessment, its invasiveness has driven the development of noninvasive diagnostic tools, including transient elastography (TE) and serum-based fibrosis scores. To evaluate and compare the diagnostic performance of TE and commonly used serum-based fibrosis scores for detecting clinically significant fibrosis (≥F2) in patients with biopsy-proven MASLD. Materials and Methods: This retrospective cross-sectional exploratory diagnostic accuracy study included 30 adults with biopsy-confirmed MASLD. Patients were classified into clinically significant fibrosis (CSF, ≥F2; n = 9) and no/mild fibrosis (NMF, F0–F1; n = 21) groups. Among the nine patients with CSF, six had stage F2 fibrosis and three had stage F3 fibrosis. TE, FIB-4, APRI, NAFLD Fibrosis Score (NFS), BARD score, and HOMA-IR were evaluated. Diagnostic performance was assessed using receiver operating characteristic (ROC) analysis, optimal cutoff values, dual-cutoff strategies, and confusion matrix analysis. Results: TE, FIB-4, and APRI demonstrated the highest diagnostic performance (AUROC: 0.93 for each). TE showed the most balanced diagnostic accuracy (sensitivity 88.9%, specificity 90.5%), whereas APRI provided the highest specificity (95.2%) and the strongest positive likelihood ratio (+LR: 18.67). FIB-4 achieved 100% sensitivity with lower specificity (76.2%). NFS demonstrated moderate diagnostic performance (AUROC: 0.83) with a relatively wide gray zone (56.7%), whereas BARD and HOMA-IR showed limited discriminative ability. When fibrosis categories automatically assigned by the TE device were evaluated, the overall diagnostic accuracy was 93.3%. Conclusions: TE, FIB-4, and APRI demonstrated promising diagnostic performance as noninvasive tools for identifying clinically significant fibrosis in biopsy-proven MASLD. These findings generate the hypothesis that a sequential diagnostic strategy using serum-based fibrosis scores for initial risk stratification followed by TE for confirmation may reduce the need for liver biopsy. However, this hypothesis requires confirmation in larger prospective validation studies.