Purpose: To assess the prevalence and severity of transfusion-associated iron overload in survivors of childhood cancer. Patients and Methods: Serum iron, total iron binding capacity, percent iron saturation and ferritin were measured in 75 survivors of childhood cancer. In addition, blood bank records were reviewed to determine the volume of packed red blood cells (pRBCs) administered during cancer therapy. Patients who received > 120ml/kg of pRBCs or had a ferritin > 1,000mcg/L underwent hepatic R2 and cardiac T2* MRI for iron quantification, echocardiogram, assessment of liver and endocrine function, and genetic analysis for hereditary hemochromatosis. Results: Forty-nine patients qualified for second level studies. Of these, 35 completed the MRI scans. Fifteen patients had a liver iron concentration (LIC) >3mg iron/g (moderate hepatic iron overload), including eight patients who had an LIC greater than 7 mg iron/g (severe hepatic iron overload), with a mean LIC of 4.3 mg iron/g (0-15.6mg iron/g). LIC correlated with both total volume of pRBCs and ferritin. No patient had cardiac iron loading by MRI. Eleven patients were heterozygous and one was homozygous for mutations associated with hereditary hemochromatosis. There was no correlation between iron overload and hereditary hemochromatosis gene status. Conclusion: There is a high prevalence of transfusion-associated iron overload among survivors of childhood cancer. This is concerning given the overlap between organ toxicities associated with cancer treatment and those known to be associated with iron overload. The tight correlation between LIC and ferritin suggests ferritin may be a reliable indicator of iron load in this patient population.