P10.12 Prognostic impact of 2016 WHO classification in anaplastic gliomas

Abstract

OBJECTIVE: 2016 World Health Organization (WHO) classification of Tumors of the Central Nervous System (CNS) added molecular information to histology to redefine CNS tumors. We investigated whether the updated 2016 WHO classification in anaplastic gliomas has more prognostic impact for disease progression and mortality than 2007 WHO classification. METHODS: A total of 113 consecutive patients with newly diagnosed anaplastic gliomas by 2007 WHO classification at our hospital from Jan. 2001 to Dec. 2013 were enrolled in this study. We integrated the molecular profiles in each patient and reclassified the diagnosis according to 2016 WHO classification. The Kaplan-Meier methods and a stepwise multivariate Cox proportional regression analysis were performed to evaluate the survival and prognostic factors. To further evaluate the predictability of 2016 WHO classification across the entire follow-up period, we applied a time-dependent receiver operating characteristic (ROC) analysis for censored survival data. We then compared the global concordance probability (integrated area under the curve, iAUC) of the 2016 model with 2007 model. RESULTS: A total of 57 anaplastic astrocytoma (AA) patients in the 2007 classification were classified by the new 2016 classification as 51 patients of AA, IDH-wildtype (AAw); 3 patients of AA, IDH-mutant (AAm); 2 patients of Anaplastic oligodendroglioma, NOS (AOnos); 1 patient of AO, IDH-mutant and 1p/19q-codeleted (AOmc). A total of 27 anaplastic oligoastrocytoma (AOA) patients were newly classified as 7 patients of AAw; 10 patients of AAm; 2 patients of AOnos; 8 patients of AOmc. And 29 AO patients were newly classified as 4 patients of AAw; 2 patients of AAm; 6 patients of AOnos; 17 patients of AOmc. The overall survival of total group of anaplastic gliomas was 42.7 months, of which the AAw group was 21.0 months. The progression free survival of total group was 31.7 months, of which the AAw group was 16.8 months. After a multivariate Cox regression analysis, age, preoperative Karnofsky Performance Scale (KPS) score, postoperative enhanced tumor volume in contrast enhanced T1-weighted magnetic resonance imaging (MRI), postoperative tumor volume in T2-weighted MRI, extent of resection measured in T2-weighted MRI, deep location of tumor were associated with the overall survival. Time-dependent ROC curve demonstrated that concordance probability increased from 2007 WHO classification model to 2016 WHO classification model. On visual inspection, the 2016 WHO classification model had a higher iAUC throughout the entire follow-up period than the 2007 WHO classification model. These findings suggest that the 2016 WHO classification to the survival increases the predictive accuracy for mortality. CONCLUSION: We showed that 2016 WHO classification of Tumors of the CNS may be more predictable of survival in patients with anaplastic gliomas.