TH‐C‐141‐06: Estimating Cell Density Using Fractional Anisotropy From Postoperative Diffusion Tensor Imaging of High‐Grade Gliomas

Purpose: Fractional anisotropy (FA) from diffusion tensor imaging (DTI) has been suggested as a predictor of glioma cell density and proliferation activity. The purpose of this study is estimate the glioma cell density inside high‐grade glioma radiotherapy target volumes using FA images. Methods: Five patients with histologically‐confirmed glioma underwent radiotherapy planning with postoperative magnetic resonance imaging (MRI) and computed tomography. T1‐weighted images with gadolinium contrast enhancement and T2‐weighted fluid attenuated inversion recovery images used for treatment planning were obtained. DTI was obtained using echo planar imaging for 20 noncolinear directions with b=1000 s/mm2 and one additional image with b=0. Diffusion imaging was repeated four times for signal averaging. The gross tumor volume (GTV) was delineated using T1‐weighted and T2‐weighted images. A clinical target volume (CTV) was defined as a 2‐cm expansion of the GTV and a shell volume (CTV‐shell) between the GTV and CTV contours was created. The distribution of FA inside the GTV, CTV‐shell, and normal brain tissue was calculated. Cell density inside the CTV was estimated from FA values using a linear model. Results: The mean FA inside the GTV was 0.13±0.08 and inside CTV‐shell was 0.20±0.12. The mean FA in normal brain tissue (0.28±0.12) was significantly higher than the mean FA inside the GTV (p=0.003) and CTV‐shell (p=0.01). The estimated mean cell density inside CTV‐shell was 1.57±0.94 times the mean cell density inside the GTV (p=0.02). Conclusion: FA values in the GTV and CTV‐shell were significantly smaller than values in normal brain regions. FA and estimated cell density values approached those of normal brain tissue as the distance from the GTV increased, consistent with the expectation of a gradual and decreasing presence of tumor cells. Further research is warranted to determine if treatment planning using FA images will improve treatment outcome.