Noninvasive detection of early pathophysiological changes of irradiated bone marrow in mouse radiation carcinogenesis by diffusion-weighted imaging

Ionizing radiation is capable of causing various tumors in both mouse and human and thus is a well-known environmental carcinogen. Mouse models have been invaluable to elucidate the mechanism(s) of radiation carcinogenesis. To date, many rodent models have been developed for this purpose. Fractionated whole-body X-ray irradiation (FX) efficiently induces thymic lymphoma (TL) in C57BL/6 mice, which are susceptible to radiation-induced thymic lymphoma (RITL) development, at 3 to 10 months after the last irradiation. This model has long been used as a classical method to study radiation carcinogenesis. In this model, although TL is arising from the thymus, many previous studies have shown that cellular damages of the bone marrow (BM) by FX are critically involved in TL development. BM changes after FX has been examined by histological analyses so far. However, analyzing spatiotemporal dynamics of pathophysiological BM changes induced by FX has been hampered by the lack of a reliable method to monitor them noninvasively. The aim of this study is to explore methods for noninvasive monitoring of the pathophysiological changes of the X-ray irradiated BM in mouse radiation carcinogenesis. Here we report that diffusion-weighted imaging (DWI) noninvasively detects early pathophysiological changes of FX-treated BM in mouse radiation carcinogenesis. Four-weeks-old C57BL/6 mice were X-irradiated with 4 doses of 1.6 Gy (total dose 6.4 Gy) each given at 7-days intervals and the pathophysiological changes of their BM were evaluated by MRI and histopathological analyses 3 days after each irradiation. T2 values were not significantly different between FX-treated and unirradiated normal BM. Interestingly, the apparent diffusion coefficient (ADC) measured by DWIs, which quantitatively represents the diffusivity of protons in tissue, significantly increased in FX-treated BM compared to those in unirradiated normal BM from the first to third irradiation (p < 0.001). These ADC changes were correlated with and well accounted for by the pathological findings, likely reflecting reduced cellularity and altered vascularity in the irradiated BM. These data indicate that DWI is useful in quantitatively evaluating dynamic changes of FX-treated BM in living subjects and may help elucidate the pathogenesis of RITL. In addition, this study further supports the notion that DWI is able to detect altered cellularity and thus may be applied to evaluate BM suppression which often is the limiting factor in clinical fractionated radiotherapy.2012 World Molecular Imaging Congres