Background/Objectives: We aim to determine if cytokeratin-18 fragment (CK-18F) could be used to diagnose metabolic dysfunction-associated steatohepatitis (MASH). Methods: A total of 289 patients with metabolic dysfunction-associated steatotic liver disease (MASLD) were enrolled in the analysis. To evaluate the association between CK-18F levels and the histological features of MASH, weighted receiver operating characteristic (ROC) curve analyses were performed. The diagnostic utility of CK-18F was compared with that of the Mac-2 binding protein glycan isomer (M2BPGi). Additionally, we assessed the predictive performance of combining CK-18F with either the FIB-4 index or the FIB-3 index for diagnosing MASH and investigated predictors of future progression to cirrhosis. Results: CK-18F was more useful for MASH diagnosis than M2BPGi and the FIB-4 index in the multivariate analysis, with a sensitivity of 47% and specificity of 80% at a CK-18F cutoff value of 750 U/L. Because CK-18F decreases with advanced liver fibrosis, the combination of the FIB-4 or FIB-3 index with CK-18F was examined to identify cases with cirrhosis. The combination of the CK-18F level and the FIB-3 index better predicted MASH than the combination of the CK-18F level and the FIB-4 index. The FIB-3 index was the most useful predictor of cirrhosis on imaging five years after diagnosis with F2 or less disease. Conclusions: CK-18F is useful for MASH diagnosis, and the diagnostic algorithm combining CK-18F with the FIB-3 index may be more useful than the previously reported MASH diagnostic algorithm that combined it with the FIB-4 index.
Keywords: cytokeratin-18 fragment; fibrosis-3 index; fibrosis-4 index; metabolic dysfunction-associated steatohepatitis; metabolic dysfunction-associated steatotic liver disease.