Abstract 2681: Combined Dynamic 18F-FDG PET Imaging and DCE MRI prediction of treatment response in an orthotopic model of glioblastoma multiforme.

Abstract

Abstract BACKGROUND: Gliobastomas (GBM) are highly aggressive tumors. When patients are treated with the standard of care of combined radiation therapy and temozolomide (TMZ), median survival is less than 15 months. An earlier prediction of treatment response could allow alternative salvage therapies to be applied sooner if efficacy is not evident. Non-invasive medical imaging provides the opportunity to link imaging-based biomarkers to treatment response. It was hypothesized that a multimodality imaging protocol (heretofore lacking) could allow more accurate prediction of treatment response than conventional means. MATERIALS & METHODS: 1x106 U87MG-luc cells were intracranially injected into 6-8 week old female Nude mice using a stereotactic surgical apparatus. Animals were staged for treatment using a T2-weighted MRI sequence and treatment groups were populated with a tumor volume of 10-20mm3. Treatment was delivered for 5 days as follows 1) Control and 2) TMZ. DCE MRI images were acquired with a gadolinium (Gd) contrast agent and dynamic 18F-FDG PET imaging was acquired at baseline and days 7 and 14 post-treatment. Initial AUC (at 30, 60 and 90s) and Ktrans were determined by fitting the Gd time course data to a generalized 2 compartment kinetic model (Tofts-Kermode approach) and calculated on a voxel-by-voxel basis. PET data was analyzed with traditional kinetic modeling (Ki) using Patlak analysis from 12-70 min and for the SUVmean and SUVmax. Unified metrics that combined Ktrans, Ki and/or SUV were evaluated for predictive power, as compared to each parameter alone. RESULTS: Animals treated with TMZ had a significant increase in median lifespan of 87 days, as compared to controls (59 days; p=0.002). Comparisons of the influx constant, Ki, showed no significant differences between control and treated groups, regardless of whether mean or maximum values were used. Similarly, there were no significant differences between groups in the SUV mean and SUVmax. DCE MRI data is currently undergoing analysis. CONCLUSIONS: Control and TMZ-treated groups produced expected median lifespan results as compared to historical data. Static and dynamic analysis of 18F-FDG PET data did not demonstrate differences between treated and untreated groups. It is possible that the time interval to differentiate the two groups was too proximate to extrapolate response from the derived parameters or that 18F-FDG is not the optimal tracer for this type analysis. It is therefore proposed that future studies utilize later time points in evaluating 18F-FDG-derived biomarkers and/or evaluate the utility of alternate radiotracers (e.g. 18F-FLT). Additionally, it is expected that combining radiation therapy with TMZ will produce more measurable differences in these biomarkers at early time points. A hybrid metric that combines PET and DCE MRI outputs is under development as an enhanced predictor of response. Citation Format: Erin Trachet, John Chunta, Chris Bull, Jeni Baranski, Tracey Woolliscroft, Deanne Lister, Patrick McConville, Dick Leopold. Combined Dynamic 18F-FDG PET Imaging and DCE MRI prediction of treatment response in an orthotopic model of glioblastoma multiforme. [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 2681. doi:10.1158/1538-7445.AM2013-2681