Background: The prognosis of patients with anaplastic oligodendrogliomas (AOD) and oligoastrocytomas (AOA) is variable. Biomarkers might be helpful to identify more homogeneous disease subtypes and improve therapeutic index. The aim of this study is to develop new clinical, pathological and molecular prognostic models for locally diagnosed anaplastic gliomas with oligodendroglial features (AOD or AOA).
Methods: Data from 368 patients with AOD or AOA recruited in The European Organisation for Research and Treatment of Cancer (EORTC) trial 26951 on adjuvant PCV (Procarbazine, CCNU, Vincristine) chemotherapy in anaplastic oligodendroglial tumours were used to develop multifactor models to predict progression free survival (PFS) and overall survival (OS). Different models were compared by their percentage of explained variation (PEV). Prognostic calculators were derived from these new models.
Results: Treatment (for PFS only), younger age, confirmed absence of residual tumour on imaging, frontal location, good World Health Organisation (WHO) performance status, absence of endothelial abnormalities and/or necrosis, 1p/19q codeletion and Isocitrate dehydrogenase 1 (IDH1) mutation were independent factors that predicted better PFS and OS.
Conclusions: We identified important prognostic factors for AOD and AOA and showed that molecular markers added a major contribution to clinical and pathological factors in explaining PFS and OS. With a positive predictive value of 92% for PFS and 94% for OS, our models allow physicians to precisely identify high risk patients and aid in making therapeutic decisions.
Keywords: Cox models; Oligoastrocytomas; Oligodendrogliomas; Prognostic factors.
Copyright © 2013 Elsevier Ltd. All rights reserved.