Evaluation of MARC1 Variants and Extracellular Vesicle Cytokeratin-18 as Predictive Biomarkers in a group of Egyptian MASLD and MASH patients

Background: Metabolic dysfunction-associated steatotic liver disease (MASLD) and its progressive inflammatory phenotype, metabolic dysfunction-associated steatohepatitis (MASH), represent an increasing global health challenge. Disease progression reflects a complex interaction between metabolic stress and genetic susceptibility. Variants in the mitochondrial amidoxime reducing component 1 (MARC1) gene have been implicated in hepatic lipid handling and hepatocellular injury, and hepatic outcomes. In parallel, hepatocyte-derived extracellular vesicles (EVs), particularly those carrying cytokeratin-18 (CK-18), have emerged as promising non-invasive indicators of liver cell damage. Aims: To investigate whether integrating MARC1 genetic variants with metabolic traits, conventional biochemical markers, and circulating EV-bound CK-18 improves the diagnostic and predictive performance for MASLD and MASH. Methodology: This case–control study comprised 450 participants, including 150 with fibroscan-confirmed MASH, and 150 healthy controls. TaqMan real-time PCR was used for genotyping of rs2642438 G>A in MARC1. While total, filtered, and EV-bound CK-18 levels were quantified using enzyme-linked immunosorbent assays. Multivariate regression and receiver operating characteristic (ROC) analyses were applied to evaluate genotype–phenotype associations and diagnostic performance. Results: Carriers of the MARC1 A allele exhibited a significantly lower risk of MASLD and MASH, with lower odds of MASLD diagnosis, suggesting a hepatoprotective genetic profile. Furthermore, circulating CK-18 levels, including the EV-bound fraction, increased progressively from controls to MASLD and were highest in MASH patients, correlating with disease severity. Conclusions: A multimodal approach that combines MARC1 genetic profiling with EV-bound CK-18 and conventional biochemical markers significantly improves non-invasive prediction and risk stratification across the MASLD–MASH spectrum.