LAUSR

AbstractThe updated WHO 2016 classification of CNS tumors combines histological and molecular features for an integrated classification. We aimed to redefine grade II/III diffuse astrocytomas earlier diagnosed as per WHO 2007 using molecular markers: IDH1/2 mutation, 1p/19q co-deletion, EGFR amplification as per new 2016 WHO classification guidelines. A total of 180 grade II (n=94) and grade III (n=86) astrocytic tumors were included diagnosed from 2014 to 2016. H&E slides were reviewed by three neuropathologists. An algorithm-based stepwise analysis was undertaken. Immunohistochemistry for IDH1-R132H and ATRX was done first. IDH1 negative cases were subjected to Sanger sequencing for IDH1/2 mutation. Cases with loss of ATRX and with IDH mutation by IHC/sequencing were categorized as astrocytomas IDH mutant. ATRX retained cases were analysed for 1p/19q deletion by FISH. All IDH wild-type tumors, and IDH mutant tumors with no ATRX loss or 1p/19q co-deletion were subjected to FISH to look for EGFR amplification. p53 IHC and MIB-1 were done in all cases. A cut-off of 50% or more of nuclei showing strong immunostaining was taken as p53 positive. Of the cases diagnosed as astrocytoma grade II/III by morphological criteria, 57% had the molecular fingerprint of astrocytoma (IDH+/ATRX loss). Molecular profile of oligodendroglioma (IDH mutant/ATRX retained/1p/19q co-deleted) was noted in 1% of cases while 1% were primary glioblastomas (IDH wild-type/ EGFR amplified). p53 expression varied in these groups, being 78% in the first and nil in the latter two. The remaining 41% included 23% IDH wild-type, EGFR non-amplified and 18% IDH mutant but with retained ATRX and no 1p/19q co-deletion which did not match the expected molecular profile of astrocytomas, oligodendrogliomas or glioblastomas. These unclassified cases encompass a heterogeneous group which needs further molecular evaluation for accurate classification. Algorithmic approach to molecular characterization of gliomas is important for accurate diagnosis and prognostication. Use of such an approach alters the frequency of astrocytoma, oligodendroglioma and glioblastoma diagnosis as shown by change in diagnosis from astrocytoma to oligodendroglioma and glioblastoma in some of our cases. Reclassification of tumors with histological-genetic discordance is warranted especially for improved patient care and inclusion in clinical trials in order to ensure uniformity of patients subgroups. Current WHO category of both IDH mutant and wild-type astrocytomas encompass a heterogeneous group of tumors, especially the subgroups of astrocytomas that do not have classical molecular fingerprint of astrocytoma/oligodendroglioma/glioblastoma, which need further evaluation to identify novel pathogenetic pathways.