Abstract 3930: Analysis of longitudinal relaxation time mapping for monitoring vascular permeability in brain tumors.

Abstract

Abstract Introduction: Vessel permeability measurement in tumors, as assessed by Ktrans exchange coefficient, may have diagnostic value for determining treatment effectiveness and are an emerging parameter for drug transport modeling. Transvascular exchange models have been the standard for deriving Ktrans in an imaging voxel based on the pharmacokinetic analysis of the longitudinal relaxation time (T1) measurements following systemic delivery of T1-shortening contrast reagent (CR) in tissue using dynamic contrast-enhanced (DCE)-MRI. Analysis requires a pre-contrast longitudinal relaxation time (T10) as a model input. This study investigates the effect of using T10 maps versus an average value for analyzing changes in Ktrans measurements in intracerebral lung carcinoma, glioma, and melanoma xenograft models before and after treatment with agents targeting tumor edema. Materials and Methods: Rats were inoculated in the right basal ganglia with LX-1 human lung carcinoma (n = 10), UW28 human glioma (n = 11), or A2058 human melanoma cells (n = 15). After baseline MRI scans, animals received glyburide (∼24 μg/kg oral), dexamethasone (∼1.8 mg/kg IV) or vehicle control. MRI study was performed using a horizontal bore 11.75 T magnet system. T10 in tumor tissue was measured using a spin-echo inversion recovery sequence. DCE-MRI was done with a fast-gradient-echo sequence. Ktrans was calculated by non-linear squared fitting to the T1 measurements; animal-specific arterial input functions were used. A Student's two-tailed, paired t-test was used to test statistical significance of change in Ktrans between using T10 maps versus ROI-based T10 value. Results and Discussion: On pre-treatment MRI, 50% of lung and melanoma metastases showed standard deviation ≥10% the average T10 value. For UW28 tumors, standard deviation was <10% of the average T10 for all animals. LX-1 metastases became more heterogeneous based on σ increase. Ktrans was assessed using an ROI-based average T10 value and T10 maps before and after treatment. T10 mapping reduced average Ktrans in over 88% of cases (including pre- and post- treatment). Change in Ktrans from T10 mapping did not change with treatment (p ≥ 0.07) and change in Ktrans resulting from treatment was not significantly affected by T10 mapping (p ≥ 0.09) the treatment groups. There was a trend towards significance observed with larger animal numbers. Conclusions: LX-1 and A2058 metastases were more heterogeneous than the UW28 model based on standard deviation of T10 values within tumor. Using an ROI-based average T10 value was likely to overpredict average Ktrans values, but not likely to significantly affect treatment outcomes of study for these tumors and therapies despite tumor heterogeneity. Overpredicted Ktrans hot spots could have more consequences on hot spot analysis to monitor treatment outcome rather than average Ktrans. Citation Format: Gregory L. Pishko, Eric M. Thompson, Edward A. Neuwelt. Analysis of longitudinal relaxation time mapping for monitoring vascular permeability in brain tumors. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3930. doi:10.1158/1538-7445.AM2013-3930