Utility of Serum Biomarker Indices for Staging of Hepatic Fibrosis Before and After Venesection in Patients With Hemochromatosis Caused by Variants in HFE.

BACKGROUND & AIMS: Hemochromatosis that is associated with variants in the homeostatic iron regulator gene (HFE) is characterized by intestinal absorption of iron and excessive body and hepatic iron stores; it can lead to hepatic fibrosis and cirrhosis. Fibrosis has been staged by analysis of liver biopsies, but non-invasive staging methods are available. We evaluated the ability of aspartate aminotransferase:platelet ratio index (APRI), the fibrosis-4 (FIB-4) index, and gamma-glutamyl transferase:platelet ratio (GPR) to assess hepatic fibrosis staging in subjects with HFE-associated hemochromatosis, using liver biopsy-staged fibrosis as the reference standard.

METHODS

We performed a retrospective, cross-sectional analysis of 181 subjects with HFE-associated hemochromatosis and hepatic fibrosis staged by biopsy analysis and available serum samples. We calculated APRI, FIB-4, and GPR at diagnosis for all 181 subjects and following venesection therapy in 64 of these subjects (7 subjects had follow-up biopsy analysis). We used area under the receiver operating characteristic curve (AUROC) analysis to assess the relationships between APRI score, FIB-4 score, and GPR and advanced (F3-F4) fibrosis and to select cut-off values.

RESULTS

Hepatic fibrosis stage correlated with APRI score (r = 0.54; P < .0001), FIB-4 score (r = 0.35; P < .0001), and GPR (r = 0.36, P < .0001). An APRI score above 0.44 identified patients with advanced fibrosis with an AUROC of 0.88, 79.4% sensitivity, 79.4% specificity, and 81% accuracy. A FIB-4 score above 1.1 identified patients with advanced fibrosis with an AUROC of 0.86, 80% sensitivity, 80.3% specificity, and 81% accuracy. A GPR above 0.27 identified patients with advanced fibrosis with an AUROC of 0.76, 67.7% sensitivity, 70.3% specificity, and 69% accuracy. APRI score was significantly more accurate than GPR (P = .05) in detecting advanced fibrosis; there was no difference between APRI and FIB-4. Venesection treatment was associated with significant reductions in APRI (P < .0001) and GPR (P < .001), paralleling fibrosis regression observed in available liver biopsies. Post-venesection APRI identified 87% of subjects with advanced fibrosis that decreased to levels that indicate stage F1-F2 fibrosis.

CONCLUSIONS

In a retrospective study of 181 subjects with HFE-associated hemochromatosis, we found that APRI and FIB-4 scores identified patients with advanced hepatic fibrosis with 81% accuracy. APRI scores might also be used to monitor fibrosis regression following venesection.