Unencrypted login | home

Program Information

BEST IN PHYSICS (JOINT IMAGING-THERAPY): Correlation of 18F-DOPA-PET Uptake and MR-Diffusion Tensor Imaging Maps with Glioma Tumor Pathology

no image available
M Zakhary

M Zakhary*, J Sedlar , P Korfiatis , B Erickson , C Giannini , I Parney , D Pafundi , N Laack , D Brinkmann , Mayo Clinic, Rochester, MN


TH-AB-FS4-12 (Thursday, August 3, 2017) 7:30 AM - 9:30 AM Room: Four Seasons 4

Purpose: To determine correlations of 18F-DOPA-PET and MR-Diffusion Tensor Imaging (MR-DTI) with tumor pathology across multiple biopsy specimens.

Methods: Patients suspicious for newly diagnosed or recurrent glioma (N=13 patients, n=30 biopsy samples) were imaged with conventional MRI, 18F-DOPA-PET and MR-DTI prior to surgical resection or biopsy. Stereotactic biopsies were then taken in the regions of maximum 18F-DOPA-PET uptake using Stealth stereotactic biopsy planning software, and in up to 4 concordant and discordant conventional MR/18F-DOPA-PET regions of interest. MR-DTI maps were computed, and the MR and PET images rigidly registered using in-house software. Tumor pathology for each biopsy sample was assessed by an expert pathologist. Tumor pathology correlations were evaluated for 18F-DOPA-PET uptake and several MR-DTI maps: Average Diffusion Coefficient, Fractional Anisotropy, Anisotropic Diffusion Coefficient, and Isotropic Diffusion Coefficient. Relative quantities for each imaging modality were calculated by normalizing to the mean value in a contralateral, tumor-free region of the brain (T/N). For each modality, statistical significance using the Student’s t-test (α=0.05) was evaluated based on four metrics (mean, median, minimum, and maximum).

Results: Consistent with previous findings in the literature, 18F-DOPA-PET SUVmax T/N was strongly correlated with glioma tumor pathology. We observed a statistically significant difference between grade II and grade IV disease (p<0.0001) and between grade III and grade IV disease (p=0.0014). No statistically significant difference was observed between grade II and grade III tumors (p=0.0734).. No statistically significant difference was determined for any of the MR-DTI metrics with tumor pathology (0.15< p <1). For all modalities, data quantification using T/N ratios showed a stronger correlation with tumor pathology than absolute measurements.

Conclusion: Across imaging modalities, only 18F-DOPA-PET SUVmax shows a statistically significant correlation to tumor pathology. This suggests that 18F-DOPA-PET SUVmax T/N is preferable to MR-DTI as a tool for accurate biopsy targeting and eventual radiotherapy planning.

Add this talk to vcal | ical | Contact Email: