Transfusional hemosiderosis is often associated with hepatic siderosis or infection with hepatotropic viruses, resulting in hepatocellular injury and progression to chronic liver disease. Liver biopsy is the method of choice for directly assessing damage; scales have been developed to measure necroinflammatory activity (grading) and tissue fibrosis (staging). Iron chelation therapy is historically known to decrease morbidity associated with hepatosiderosis.

Aim: To assess 1 year’s chelation therapy with the novel once-daily oral chelator, deferasirox (DSX), or the current standard deferoxamine (DFO), on pathology of liver tissue in chronically transfused patients.

Methods: Liver biopsy was performed at baseline and after 1 year in patients participating in DSX Studies 0107 (n=454) and 0108 (n=101). All patients, except 25 in 0107 and 3 in 0108, had liver tissue evaluated by pathology. In 0107, patients with β-thalassemia were randomized to DSX (5, 10, 20 or 30 mg/kg/day; n=224) or DFO (<25, 25–35, 35–50 and ≥ 50 mg/kg; n=230) according to baseline liver iron concentration (LIC). In 0108, patients with β-thalassemia unable to be treated with DFO (n=61) and patients with rare anemias (MDS, DBA and others) (n=40) were enrolled and received DSX only. Grading and staging were determined from biopsy by the Ishak method; LIC was measured in parallel by atomic absorption spectrometry.

Results: DSX and DFO dose-dependently affected grading, which mirrored effects on LIC and serum ferritin. DSX 5 and 10 mg/kg increased these parameters, while stabilization and decreases were seen with the highest doses of both chelators, regardless of hepatitis C status. In 0107, a decrease in mean ±SD necroinflammatory score was noted with DSX 30 mg/kg (2.5 ±1.6 to 1.7 ±1.3, n=95) and DFO ≥ 50 mg/kg 5 days/week (2.1 ±1.6 to 1.4 ±1.3, n=95). Similar results were observed in 0108 for β-thalassemia (2.4 ±1.7 to 1.7 ±1.6, n=58) and rare anemia patients (1.8 ±1.5 to 1.5 ±1.3, n=40). This decrease was accompanied by dose-dependent modification of available liver enzyme levels in 0107 (Table); with a similar trend in 0108. No obvious modification of staging was observed after 1 year of treatment, suggesting that longer time periods are needed to observe potential reversal of fibrosis.

Change in liver enzymes (ALT; U/L) by treatment (Study 0107)

DSX, mg/kgDFO, mg/kg
5102030<2525–3535–50≥ 50
n=8n=43n=64n=107n=6n=28n=88n=107
Mean ± SD 
34.9 21.9 3.6 −2.8 −13.9 −3.7 −2.8 −12.4 
±35.1 ±25.2 ±28.7 ±79.3 ±29.7 ±17.6 ±20.7 ±38.6 
Median (range) 
36.8 10.0 0.5 −8.0 −10.8 −1.8 −0.8 −9.5 
(−19.0, 101.5) (−7.5, 105.0) (−111.5, 91.0) (−143.0, 711.0) (−61.0, 27.0) (−55.5, 32.0) (−93.5, 70.5) (−171.0, 193.3) 
DSX, mg/kgDFO, mg/kg
5102030<2525–3535–50≥ 50
n=8n=43n=64n=107n=6n=28n=88n=107
Mean ± SD 
34.9 21.9 3.6 −2.8 −13.9 −3.7 −2.8 −12.4 
±35.1 ±25.2 ±28.7 ±79.3 ±29.7 ±17.6 ±20.7 ±38.6 
Median (range) 
36.8 10.0 0.5 −8.0 −10.8 −1.8 −0.8 −9.5 
(−19.0, 101.5) (−7.5, 105.0) (−111.5, 91.0) (−143.0, 711.0) (−61.0, 27.0) (−55.5, 32.0) (−93.5, 70.5) (−171.0, 193.3) 

Conclusions: Although the data show considerable variability, results suggest that chelation therapy with DSX or DFO is associated with reduced hepatocellular inflammation and improved liver function. These modifications appear to be linked with effects on LIC and serum ferritin levels.

Author notes

Corresponding author

Sign in via your Institution