PURPOSE: To clarify the change in the fluorodeoxyglucose (FDG) uptake by the bone marrow over time after administration of granulocyte colony-stimulating factor (G-CSF), we evaluated the correlation between the interval from the last day of administration of G-CSF to positron emission tomography/computed tomography (PET/CT) study and spinal bone marrow accumulation in patients with non-Hodgkin's lymphoma. METHODS: A total of 127 patients with confirmed non-Hodgkin's lymphoma who underwent FDG PET within 60 days from the last administration of G-CSF were retrospectively reviewed. Thirty age-matched and sex-matched healthy controls were also included to evaluate physiological FDG uptake. PET/CT examinations were retrospectively reviewed, and maximum standardized uptake value (SUVmax) was measured by placing volumetric regions of interest over each thoracic and lumbar vertebra on PET images referring to CT images. Bone marrow SUV was defined as the mean SUVmax of the vertebra. The correlation between the interval after G-CSF and the bone marrow SUV was plotted and analyzed with polynomial approximation. RESULTS: In controls, physiological bone marrow SUV of the spine was determined. In patients with lymphoma, bone marrow SUV decreased over time and reached a plateau at about 14 days after G-CSF administration, and this was higher by 5% than the plateau at 10 days. SUV declined to the 'physiological range', that is, mean+1 standard deviation of patients, at about 7 days. CONCLUSION: For a PET/CT study, an interval of 10 days after G-CSF administration is recommended to minimize the influence of G-CSF on the bone marrow when evaluating treatment response in patients with non-Hodgkin's lymphoma.
PURPOSE: To clarify the change in the fluorodeoxyglucose (FDG) uptake by the bone marrow over time after administration of granulocyte colony-stimulating factor (G-CSF), we evaluated the correlation between the interval from the last day of administration of G-CSF to positron emission tomography/computed tomography (PET/CT) study and spinal bone marrow accumulation in patients with non-Hodgkin's lymphoma. METHODS: A total of 127 patients with confirmed non-Hodgkin's lymphoma who underwent FDG PET within 60 days from the last administration of G-CSF were retrospectively reviewed. Thirty age-matched and sex-matched healthy controls were also included to evaluate physiological FDG uptake. PET/CT examinations were retrospectively reviewed, and maximum standardized uptake value (SUVmax) was measured by placing volumetric regions of interest over each thoracic and lumbar vertebra on PET images referring to CT images. Bone marrow SUV was defined as the mean SUVmax of the vertebra. The correlation between the interval after G-CSF and the bone marrow SUV was plotted and analyzed with polynomial approximation. RESULTS: In controls, physiological bone marrow SUV of the spine was determined. In patients with lymphoma, bone marrow SUV decreased over time and reached a plateau at about 14 days after G-CSF administration, and this was higher by 5% than the plateau at 10 days. SUV declined to the 'physiological range', that is, mean+1 standard deviation of patients, at about 7 days. CONCLUSION: For a PET/CT study, an interval of 10 days after G-CSF administration is recommended to minimize the influence of G-CSF on the bone marrow when evaluating treatment response in patients with non-Hodgkin's lymphoma.