Abstract

Objective: Patients with iron overload suffer from different organ damage due to increased iron concentration. Iron overload in the bone marrow and the influence of iron chelation therapy on bone marrow iron content may play an additional role in these patients and is until now not well examined.

Material and Methods: We performed MRI-R2* measurements in the vertebral bone marrow using water-fat chemical shift relaxometry for estimation of iron and fat content in comparison to hepatic and splenic iron concentrations and serum ferritin in patients with iron overload due to hereditary hemochromatosis (HHC) and patients with siderosis due to red blood cell transfusions and /or iron loading anemia.

112 patients with iron overload, mean age: 32 y (transfusion dependent thalassemia major (TM) n=65, Diamond-Blackfan anemia (DBA) n=12, HHC n=10, iron loading anemia (EA) n=7, transfusion siderosis n=12 and stem cell transplantation n=6) and 14 control subjects underwent MRI for determination of the transverse relaxation rate R2*assessed from ROI based signal intensities of one transversal slice (10mm) through the liver, spleen, and mid-vertebral bone marrow.

Breathhold water-fat relaxometry (12 echo times, TE=1.3-26ms, FA=20°, bandwidth=1955Hz/px) was performed to determine apparent fat contents (aFC) and bone marrow R2*. Additionally, serum ferritin values were assessed.

In 67 patients with TM under chelation therapy with Deferasirox (DFX) we compared R2* bone marrow iron content with the ratio of the chelator dose rate (Deferasirox, [mmol/d]) to the total liver iron (LivFe = LIC*volume [g-Fe]).

Results: Relative to controls (n=14, R2* = 95s-1) and HHC (n=10, R2* = 95s-1), median bone marrow R2* rates were significantly increased in patients with TM (n=65, R2* = 398s-1, p<10-4) or DBA (n=12, R2* = 252s-1, p = 0.005). R2* of the bone marrow significantly correlated with serum ferritin (rS=0.52, p<10-4), splenic R2* (rS=0.43, p<10-4), cardiac R2* (rS=0.43, p<10-4), and hepatic R2* (rS=0.37, p<10-4). No significant correlation of aFC with marrow R2* could be obtained.

The bone marrow iron content was higher in patients with a low Deferasirox (DFX) dose. A DFX dose > 150 (mmol/d)/g-Fe was a negativ predictor for increased bone marrow iron content (R2* > 700 1/s), but a significant correlation couldn't be found (r = 0,077), probably due to the low number of patients. The efficacy dose of DFX for a low bone marrow iron content seems to lay between a dose of 1mmol/d (18mg/d) and 2,5 mmol/d (45mg/d). A DFX dose > 2,5 mmol/d didn't seem to increase the efficacy.

Table 1.
Diagnosis (n) Age
Median (IQR) 
KM R2* [s-1 ]
Median (IQR) 
KM R2*
95 % range 
Control group (14) 29.8 (26.0) 95 (43) 62 - 134 
TM ( 65) 30.6 (16.2) 398 (455) 73 - 1213 
HH (10) 56.4 (9.3) 95 (40) 61 - 270 
EA (7) 46.6 (24.8) 191 (93) 159 - 490 
DBA (12) 27,6 (14.6) 252 (250) 62 - 652 
TS(12) 32.3 (46.2) 322 (304) 116 - 607 
BMT (6) 20 (20.8) 285 (342) 105 - 470 
Diagnosis (n) Age
Median (IQR) 
KM R2* [s-1 ]
Median (IQR) 
KM R2*
95 % range 
Control group (14) 29.8 (26.0) 95 (43) 62 - 134 
TM ( 65) 30.6 (16.2) 398 (455) 73 - 1213 
HH (10) 56.4 (9.3) 95 (40) 61 - 270 
EA (7) 46.6 (24.8) 191 (93) 159 - 490 
DBA (12) 27,6 (14.6) 252 (250) 62 - 652 
TS(12) 32.3 (46.2) 322 (304) 116 - 607 
BMT (6) 20 (20.8) 285 (342) 105 - 470 

Conclusion: Water-fat chemical shift relaxometry allows precise determination of bone marrow R2* rates and estimation of the apparent fat content, which may add additional information to these patients. Patients with transfusion related iron overload have significant higher bone marrow iron content than patients with iron overload due to hereditary hemochromatosis. DFX in an adäquat dose seem to prefent bone marrow iron overload.

Disclosures

Grosse:Swedish Orphan Biovitrum: Honoraria; Novartis Oncology: Honoraria, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.