NI-18 INVASIONS OF WHITE MATTER AS A PROGNOSTIC FACTOR IN LOW GRADE GLIOMAS

INTRODUCTION: Gliomatosis cerebri (GC), which was characterized by widespread infiltration of the brain involving three lobes, was deleted in the 2016 WHO classification. However, it is known that gliomas with GC growth pattern have a poor prognosis even in low histological grading. In this study, we focused on tumor invasion into white matter fibers. We analyzed the MRI findings focusing on white matter fibers and compared in patients with histologically proven low grade gliomas (LGGs) with GC pattern and localized LGGs. METHOD: The patients can be classified into four groups according to the range of tumor invasion in T2-weighted image as follows: group 1, more than 3 lobes (n=6); group 2, 1 or 2 lobes infiltrate the basal ganglia (n=5); group 3, multicentric (n=2) and group 4 (n=12), localized. In reference to the human brain white matter atlas, the infiltration to the major white matter fibers (uncinate fasciculus, genu &splenium of corpus callosum, inferior fronto-occipital fasciculus, superior& inferior longitudinal fasciculus) was examined. RESULTS: Twenty-five patients (median 39.5 years) were included in the study. Of these, 20 patients were histologically diagnosed with diffuse astrocytomas, and 5 patients with oligodendrogliomas. The infiltrations into ifo, slf, and ilf of white matter fibers were a poor prognostic factor. The number of infiltrating white matter fibers correlated significantly with the Kaplan-Meier survival curve. CONCLUSIONS: The 2016 WHO classification defines diagnostic entities by combining molecular and histological information and remove GC as a distinct glioma entity. LGGs with GC pattern should be considered to be detected in different types of histologically and molecularly defined gliomas. As the patient numbers analyzed here were small, and larger series reproducing these results would be desirable. MRI findings particularly focusing on infiltration of LGGs into white matter fibers might be important to estimate the prognosis of patients.