A novel multi-parameter MRI-based model for identification of high-risk metabolic dysfunction-associated steatohepatitis.

OBJECTIVES

To develop and evaluate a novel multi-parameter MRI-based model (EFT1) for identifying high-risk metabolic dysfunction-associated steatohepatitis (MASH) to improve diagnostic accuracy and efficiency.

MATERIALS AND

METHODS

A prospective study included 118 patients (55 male; 48 ± 13 years) with hepatic steatosis and metabolic risk factors. Among these, 80 patients were classified as having high-risk MASH. Magnetic resonance elastography (MRE), T1 mapping, chemical-shift encoded MRI for quantification of proton density fat fraction (PDFF) and R2* were performed, followed by liver biopsy. MRI parameters were analyzed and correlated with histological features. The EFT1 model was developed using logistic regression and full subset regression analysis on a training cohort (70%) and validated on a test cohort (30%). The performance of the model was compared with traditional scoring systems.

RESULTS

Significant differences were observed in MRE, PDFF, and T1 between high-risk MASH and non-high-risk MASH groups. The EFT1 model, combining MRE, PDFF, and T1 showed strong diagnostic performance in both training (AUC 0.995, 95% CI 0.985-1.000) and test cohorts (AUC 0.995, 95% CI 0.979-1.000). At the optimal cut-off value of -0.431, the model achieved high sensitivity (98.2% training, 95.7% test) and specificity (96.3% training, 100% test). The EFT1 model outperformed traditional scoring systems (FIB-4, APRI, GPR) and showed comparable performance to the MAST score in identifying high-risk MASH.

CONCLUSION

The novel EFT1 model demonstrates reasonable performance in non-invasive identification of high-risk MASH patients compared to other models, achieving an appropriate balance between sensitivity and specificity.

CLINICAL TRIAL REGISTRATION: This study is registered with Chictr.org.cn (ChiCTR2400094017).

KEY POINTS: Question Current non-invasive scoring systems for high-risk MASH have limitations. A novel multi-parameter MRI-based model is proposed to improve diagnostic accuracy and efficiency. Findings A novel multi-parameter MRI-based model (EFT1), incorporating MRE, PDFF, and T1 mapping, demonstrated higher accuracy in identifying high-risk MASH with an AUC of 0.995. Clinical relevance The EFT1 model provides a highly accurate, non-invasive tool for early identification of high-risk MASH, facilitating timely intervention and personalized treatment strategies, potentially reducing disease progression and improving patient outcomes.