Density Changes Of Vertebra With Spinal Metastasis Following Radiation Therapy

While palliative radiation therapy (RT) is a well-established treatment for bone metastases, its effects on bone healing are not. Altered fractionation has generated interest in comparing regimen outcomes. Currently, the scoring of bone healing is purely subjective and non-reproducible. Bone density is the standard measure of bone integrity. We have undertaken a detailed analysis to find if bone healing can be measured by changes in bone density. This study included 117 spinal metastases treated with RT with follow-up CT scans >9 months following a CT simulation. Volumes for the metastasis and the whole involved vertebra were created on the pre-treatment scan and fused to the follow-up scan. As a control, a volume was drawn for the whole vertebra of the nearest uninvolved and unirradiated bone. A Hounsfield-density calibration curve was used to measure the density of these volumes before and after treatment. None of these patients had evidence of local progression. Based on the baseline density of the metastasis, patients were divided into low (mostly lytic), medium (mixed lesions) and high (mostly blastic) clusters. This correlated well with our visual scoring of lytic versus blastic. A limited number of high-density lesions (n = 3) led to the exclusion of this group from further analysis. After treatment, the GTV density increased in both groups; 6.9% (p<0.0001) in the low density and 4.5% (p = 0.0241) in the medium density lesions. Median density of the lesions in the low-density group went from 1.14 to 1.20 and in the medium density group from 1.34 to 1.43. The density of the involved vertebra increased in both groups; 4.1% in the low density and 3.7% in the medium density group. Overall, the ratio of involved vertebra to control vertebra density increased by 3.9%. After RT, most bone metastases appear to heal with an increase in bone density, which is the hallmark of bone integrity. Not surprisingly, those with a large lytic component had a larger increase in density. We seek to validate these healing results in bone metastases at other sites and, ultimately, to create an objective system for grading earlier bone response to RT.