The mechanism(s) determining the progression from fatty liver to steatohepatitis is currently unknown. Our goal was to define the relative impact of iron overload, genetic mutations of HFE, and insulin resistance on the severity of liver fibrosis in a population of subjects with nonalcoholic fatty liver disease (NAFLD) who had low prevalence of obesity and no overt symptoms of diabetes. In a cohort of 263 prospectively enrolled patients with NAFLD, 7.4% of patients had signs of peripheral iron overload and 9% had signs of hepatic iron overload, but 21.1% had hyperferritinemia. The prevalence of C282Y and H63D HFE mutations was similar to the general population and mutations were not associated with iron overload. Although subjects were on average only moderately overweight, insulin sensitivity, measured both in the fasting state and in response to oral glucose, was lower. Univariate analysis demonstrated that the presence of severe fibrosis was independently associated with older age, female sex, overweight, aspartate/alanine aminotransferase ratio, serum ferritin level, fasting glucose and insulin levels, decreased insulin sensitivity, and with histologic features (degree of necroinflammation and steatosis). After adjustment for body mass index (BMI), age, sex, and degree of steatosis, ferritin levels (odds ratio [OR] = 1.77; 95% CI = 1.21- 2.58; P = .0032) and the oral glucose insulin sensitivity (OR = 0.53; CI = 0.33-0.87; P = .0113) were independent predictors of severe fibrosis. In conclusion, the current study indicates that insulin resistance is a major, independent risk factor for advanced fibrosis in patients with NAFLD. Increased ferritin levels are markers of severe histologic damage, but not of iron overload. Iron burden and HFE mutations do not contribute significantly to hepatic fibrosis in the majority of patients with NAFLD. (Hepatology 2004;39:179-187.)

Hereditary hemochromatosis usually results from C282Y homozygosity in the HFE gene on chromosome 6p. Recently, a new type of hemochromatosis (HFE3) has been characterized in 2 unrelated Italian families with a disorder linked to 7q. Patients with HFE3 have transferrin receptor 2 (TFR2) inactivated by a homozygous nonsense mutation (Y250X). Here the identification of 2 newTFR2 mutations is reported. In a large inbred family from Campania, a frameshift mutation (84-88 insC) in exon 2 that causes a premature stop codon (E60X) is identified. In a single patient with nonfamilial hemochromatosis, a T→A transversion (T515A), which causes a Methionine→Lysine substitution at position 172 of the protein (M172K), has been characterized. TFR2 gene gives origin to 2 alternatively spliced transcripts—the α-transcript, which may encode a transmembrane protein, and the β-transcript, a shorter, possibly intracellular variant. Based on their positions, the effects of the identified mutations on the 2 TFR2 forms are expected to differ. Y250X inactivates both transcripts, whereas E60X inactivates only the α-form. M172K has a complex effect: it causes a missense in the α-form, but it may also prevent the β-form production because it affects its putative initiation codon. Analysis of the clinical phenotype of 13 HFE3 homozygotes characterized at the molecular level has shown a variable severity, from nonexpressing patients to severe clinical complications. The identification of new mutations of TFR2 confirms that this gene is associated with iron overload and offers a tool for molecular diagnosis in patients without HFE mutations.

Although the presence of nontransferrin bound plasma iron (NTBI) in transfusional iron overload is well documented, knowledge about its clinical significance is limited. We assessed NTBI levels in a large and homogeneous series of thalassemia patients on regular transfusion and chelation and explored the hypothesis that NTBI levels may be associated with relevant clinical outcomes: in particular, heart disease. Among 174 patients with thalassemia major and intermedia, we showed the presence NTBI in 145 of 174 or 83.3% of cases. NTBI levels correlated with transferrin saturation, age, and ALT, and not with serum ferritin or liver iron concentrations. At a multiple regression analysis, transferrin saturation and heart disease but not age was independent predictors of NTBI. Patients with heart disease had NTBI levels significantly higher than those without. All patients with heart disease had transferrin saturation above 70%, and all were NTBI positive. Conversely, none of the patients without NTBI and/or with transferrin saturation less than 70% had preclinical or clinical heart disease. To our knowledge, this is the first documentation of a link between the presence of NTBI in thalassemic patients with transfusional iron overload and heart disease. Further investigation from these preliminary findings may clarify whether NTBI assessment may have a role in evaluating the risks and optimizing treatment for transfusion‐dependent patients. Am. J. Hematol., 2009.

β-thalassemia is a hereditary disorder with limited approved treatment options; patients experience anemia and its complications, including iron overload. The study aim was to determine whether luspatercept could improve anemia and disease complications in patients with β-thalassemia. This open-label, nonrandomized, uncontrolled study consisted of a 24-week dose-finding and expansion stage (initial stage) and a 5-year extension stage, currently ongoing. Sixty-four patients were enrolled; 33 were non–transfusion dependent (mean hemoglobin, <10.0 g/dL; <4 red blood cell [RBC] units transfused per 8 weeks), and 31 were transfusion dependent (≥4 RBC units per 8 weeks). Patients received 0.2 to 1.25 mg/kg luspatercept subcutaneously every 21 days for ≥5 cycles (dose-finding stage) and 0.8 to 1.25 mg/kg (expansion cohort and 5-year extension). The primary end point was erythroid response, defined as hemoglobin increase of ≥1.5 g/dL from baseline for ≥14 consecutive days (without RBC transfusions) for non–transfusion-dependent patients or RBC transfusion burden reduction ≥20% over a 12-week period vs the 12 weeks before treatment for transfusion-dependent patients. Eighteen non–transfusion-dependent patients (58%) receiving higher dose levels of luspatercept (0.6-1.25 mg/kg) achieved mean hemoglobin increase ≥1.5 g/dL over ≥14 days vs baseline. Twenty-six (81%) transfusion-dependent patients achieved ≥20% reduction in RBC transfusion burden. The most common grade 1 to 2 adverse events were bone pain, headache, and myalgia. As of the cutoff, 33 patients remain on study. In this study, a high percentage of β-thalassemia patients receiving luspatercept had hemoglobin or transfusion burden improvements. These findings support a randomized clinical trial to assess efficacy and safety. This study was registered at www.clinicaltrials.gov as #NCT01749540 and #NCT02268409.