Erythroferrone in β-thalassemia: Systematic review and meta-analysis.
Source: PubMed, NCBI / U.S. National Library of Medicine
Erythroferrone suppresses hepcidin and has emerged as a candidate biomarker of ineffective erythropoiesis in β-thalassemia. No prior meta-analysis has quantitatively synthesised circulating ERFE data. A PRISMA 2020-compliant systematic review and meta-analysis (PROSPERO: CRD420261347288) searched PubMed/MEDLINE, Scopus, and Web of Science up to March 2026. Hedges' g was pooled using random-effects REML; correlation outcomes were Fisher z-transformed. Seven pre-specified sensitivity analyses evaluated robustness. Thirteen studies (975 patients; 350 controls; 2019-2026) were included. Six met primary meta-analysis criteria (k = 6; n = 541), yielding a large pooled effect (g = 2.700; 95%CI: 0.906-4.495; p = 0.003; I = 96.1%), with all seven sensitivity analyses remaining significant (g range: 1.882-3.028). The ERFE-hepcidin pool demonstrated a significant inverse association with zero heterogeneity (r = -0.342; 95%CI: -0.472, -0.197; p < 0.001; I = 0.0%). A significant positive ERFE-ferritin correlation was identified (r = +0.439; p = 0.0002); ERFE-hemoglobin was non-significant (r = -0.205; 95%CI: -0.515, +0.153; p = 0.260). Four studies reported formal ROC analyses yielding a weighted pooled AUC of 0.872 (range: 0.766-0.940). One study documented a 51% ERFE increase following luspatercept (p < 0.0001). ERFE is markedly elevated in β-thal
Abstract
Erythroferrone suppresses hepcidin and has emerged as a candidate biomarker of ineffective erythropoiesis in β-thalassemia. No prior meta-analysis has quantitatively synthesised circulating ERFE data. A PRISMA 2020-compliant systematic review and meta-analysis (PROSPERO: CRD420261347288) searched PubMed/MEDLINE, Scopus, and Web of Science up to March 2026. Hedges' g was pooled using random-effects REML; correlation outcomes were Fisher z-transformed. Seven pre-specified sensitivity analyses evaluated robustness. Thirteen studies (975 patients; 350 controls; 2019-2026) were included. Six met primary meta-analysis criteria (k = 6; n = 541), yielding a large pooled effect (g = 2.700; 95%CI: 0.906-4.495; p = 0.003; I = 96.1%), with all seven sensitivity analyses remaining significant (g range: 1.882-3.028). The ERFE-hepcidin pool demonstrated a significant inverse association with zero heterogeneity (r = -0.342; 95%CI: -0.472, -0.197; p < 0.001; I = 0.0%). A significant positive ERFE-ferritin correlation was identified (r = +0.439; p = 0.0002); ERFE-hemoglobin was non-significant (r = -0.205; 95%CI: -0.515, +0.153; p = 0.260). Four studies reported formal ROC analyses yielding a weighted pooled AUC of 0.872 (range: 0.766-0.940). One study documented a 51% ERFE increase following luspatercept (p < 0.0001). ERFE is markedly elevated in β-thalassemia and inversely correlates with hepcidin in a homogeneous and reproducible manner. Assay standardisation and prospective validation are required before clinical implementation.