P15.17.b dynamic fdg pet as a predictor of survival in patients with gliomas

Abstract BACKGROUND The neuroimaging characteristics of treatment-related necrosis (TRN) induced by chemotherapy and radiotherapy mimic those of tumor progression (TPR). This can make it challenging for clinicians to deliver prognoses for patients with gliomas and subsequently optimize their treatment regimens. In this study, we evaluate the utility of dynamic FDG (dFDG) PET, which provides information about dynamic metabolic changes, for predicting survival in patients who have undergone treatment for gliomas. MATERIAL AND METHODS Clinical and radiographic information were obtained for 21 adult patients with known or suspected recurrent gliomas and had radiographic worsening reported on MRI within 6 months of completion of radiation and chemotherapy. All patients underwent dFDG PET imaging, and the prognostic significance of rate constants from the suspected tumor region were analyzed using Cox regressions. RESULTS Based on dFDG PET imaging, higher values of the k3 rate constant predicted a hazard ratio of 6.7 of death within 12 months of diagnosis. Hazard ratios of the k1 and k2 parameters were not significant in prediction of 12-month survival. CONCLUSION The k3 kinetic parameter of dFDG PET imaging, a marker for glucose uptake and metabolism, is predictive of 12-month survival in glioma patients who have questionable tumor progression. Clinicians can use this information to estimate the likelihood of TPR versus TRN in patients with gliomas and subsequently tailor their treatment plans based upon expected survival.