Diffusion-weighted imaging as predictor of therapy response in an animal model of Ewing sarcoma

To evaluate the potential of diffusion-weighted imaging (DWI) for monitoring dose-dependent tumor response in a mouse-xenograft model of Ewing sarcoma after administration of treosulfan in different dosages.

Ewing sarcoma implanted orthotopically into the left thigh of non-obese diabetic/severe combined immunodeficiency (NOD/SCID) mice were evaluated with a 9.4 T MR unit by using a specific mouse whole body coil. Transverse T2-weighted fast spin-echo sequences, T1-weighted spin-echo sequences, and transverse echo-planar DWI examinations were performed at baseline, 24 hours and 72 hours after intraperitoneal injection of treosulfan in 2 different doses. The apparent diffusion coefficient (ADC) in the viable parts of the tumor was automatically calculated from DWI imaging findings. These findings were correlated with histopathologic results at each time point. Volumetric measurements were performed by summing up the regions of interest in consecutive slices.

T1- and T2-weighted images before administration of treosulfan showed viable tumor tissue with small necrotic areas. At 24 hours after treosulfan injection, a significantly higher increase in ADC in the viable parts of the tumor could be observed in tumors of mice that had been injected with the higher dose of treosulfan compared with mice injected with the lower dose treosulfan, whereas no significant volumetric differences between the 2 different doses could be observed. Seventy-two hours after injection of treosulfan the ADC values in the viable parts of the tumor of mice treated with the high dose of treosulfan further increased and the tumor volume in the high-dose group was now significantly lower than in the low-dose group.

Compared with volumetric measurements, DWI of the viable tumor parts could be used to discriminate between the effects of 2 different dosages at an earlier time point than volumetric measurements in an animal model in vivo. This method could be especially useful for monitoring drug effects in phase I/II clinical trials.