ITVT-04. Management of intraoperative technological advances [intraoperative MRI, neuronavigation system using PET, and 5-aminolevulinic acid (5-ALA)–induced fluorescence image-guided surgery] for glioblastoma

Abstract

Objective: Glioblastoma (GBM) is the most aggressive neoplasm among the gliomas and is characterized by histopathologic intratumoral heterogeneity, with respect to tumor morphology. The maximum resection of GBM is the standard therapy and is expected to improve prognosis. Image-guided surgery using a neuronavigation system is the standard technique for glioma surgeries. Because of the brain shift occurring during surgery, neuronavigation systems are limited by the proper presence and metabolism of GBM cells. Therefore, intraoperative technologies, such as 5-ALA fluorescence and intraoperative MRI (IoMRI), are employed. Radiotracers are used during positron emission tomography (PET) for metabolic and molecular imaging and assist the evaluation of glioma metabolism. We compared the effectiveness of these intraoperative technologies. Methods: Between January 2016 and December 2018, 45 patients with gliomas underwent IoMRI. After excluding patients who underwent biopsy only and those who could not undergo multiple PET studies (MET, FLT, and FMISO), 15 patients were selected for 5-ALA fluorescence-guided resection of GBM. Each patient received 5-ALA approximately 3 h before surgery, and a modified neurosurgical microscope was used for intraoperative visualization of 5-ALA-induced fluorescence. We graded fluorescence level as strong, vague, or none. Following tumor resection, we identified the fluorescence level and evaluated the residual volume of gadolinium-enhanced T1WI (T1-Gd) on IoMRI and at each PET study. After calculating the extent of resection (EOR) for T1-Gd on a MET PET study, we selected an end of radiation (EOR) of 96%. Subsequently, we compared the residual volume on T1-Gd for IoMRI and each PET study, between EOR>96% and EOR < 96%. Results: We detected strong 5-ALA fluorescence during induction and before tumor resection in all 15 (100%) patients with a newly-diagnosed and histopathologically-confirmed GBM. Following tumor resection, we noted an EOR>96% for T1-Gd in 8 cases (vague, 2; none, 6) and an EOR<96% for T1-Gd in 7 cases (vague, 4; none, 3). The compared median residual volume (mL) with no fluorescence between EOR > 96% and EOR < 96% for T1-Gd were T1-Gd (0.08, 0.19), MET (0.79, 0.2), FLT (0.5, 0.17), and FMISO (0.29, 0.16). For MET, 7 and 8 cases had an EOR > 96% (vague, 2; none, 5) and EOR < 96% (vague, 4; none, 4), respectively. The compared median residual volume (mL) with no fluorescence between EOR > 96% and EOR< 96% for MET were T1-Gd (0.03, 0.21), MET (0.13, 1.17), FLT (0.11, 0.73), and FMISO (0.02, 0.51). Conclusions: It is challenging to fuse IoMRI images with multiple preoperative PET images during navigation. Moreover, GBM cells are difficult to distinguish in cases without 5-ALA fluorescence. For cases without 5-ALA fluorescence, the residual volume of T1-Gd, MET, FLT, and FMISO focusing on the MET was lesser than T1-Gd. Extracting the accumulation of MET facilitated the maximum resection of GBM.