The overall objective of this research is to improve the safety of iron-chelating therapy (ICT) in patients with transfusional iron overload by developing accurate non-invasive measurement of the liver iron concentration (LIC), the best measure of the body iron burden in all forms of systemic iron overload. Our scientific premise is that MRI using quantitative susceptibility mapping (QSM) will be free of the inherent interfering factors, particularly fibrosis, that distort current LIC measurements based on R2 (=1/T2) and R2* (=R2+R2') estimates. Transfusional iron overload progressively develops in patients with refractory anemia who undergo regular red blood cell (RBC) transfusion (thalassemia major, sickle-cell disease, and other disorders), because the body lacks any effective means to excrete excess iron. Excess iron from transfused RBCs eventually leads to the formation of circulating non-transferrin-bound iron that is then progressively deposited in the liver, pancreas, heart and other organs, causing cirrhosis, diabetes, heart failure, and other disorders. ICT, which we have developed the practice guideline, can remove excess iron from cells, clear circulating non–transferrin-bound iron, and maintain or return body iron to safe levels. Safe therapy requires careful adjustment of the dose of iron-chelating agents to the body iron burden to optimize iron excretion while avoiding chelator toxicity, including gastrointestinal disorders, auditory and visual impairment, agranulocytosis and neutropenia, arthropathy, growth retardation, and potentially fatal hepatic failure, renal failure, and gastrointestinal hemorrhage. LIC at present is primarily estimated by noninvasive MRI using R2 (=1/T2) and R2* (=R2+R2') techniques that depend upon the contribution of iron to relaxation (R2) and intravoxel dephasing (R2'). A fundamental limitation of the R2 and R2* approaches is that intravoxel contents other than iron, including fibrosis, steatosis and necroinflammation, also alter relaxation. We have the biophysical insight to eliminate the R2 and R2* interfering effects using QSM, which we have developed to measure tissue magnetic properties. QSM is generated from processing the phase, while R2* is determined from the magnitude, of the same gradient echo MRI data without additional scans. Magnetic susceptibility measured by QSM has a simple linear relationship with intravoxel contents in accordance with chemical decomposition, allowing iron quantification without interfering errors from fibrosis, steatosis, necroinflammation and other intravoxel contents. Hence, we conservatively anticipate a >5 fold improvement in the accuracy of LIC measurements using hepatic QSM (hQSM) compared to the current R2 and R2* method. Our research plan has 3 specific aims: Aim 1. Develop hQSM for accurate measurement of LIC without interfering errors. Aim 2. Validate hQSM using histology and chemical measurement of LIC in liver explants. Aim 3. Evaluate hQSM in patients wit...