2029 Background: The prognostic value and clinical utility of Ki67 tumor cell proliferation index in anaplastic oligodendroglial tumors is unclear. Methods: We performed anti-Ki67 immunostaining (Mib-1 antibody) of formalin-fixed and paraffin embedded tumor specimens of 70 patients that were treated in the Phase III EORTC trial 26951 investigating adjuvant chemotherapy with procarbazine, lomustine and vincristine (PVC) in newly diagnosed anaplastic oligodendroglial tumors (J Clin Oncol 2006). Ki67 index was independently assessed by 3 observers, 1 of whom performed re-assessment several weeks after initial evaluation. Lin's concordance correlation coefficient (CCC), Bland-Altman plots and Cohen′s kappa (k) were used to evaluate pair-wise observer agreement on Ki67 index values. The average Ki67 index was correlated to clinical (patient age, performance status, type of surgery and adjuvant therapy, tumor location), centrally assessed histopathological (diagnosis, presence of high cellularity, nuclear and endothelial abnormalities, mitoses, necrosis) and molecular features (1p19q loss, EGFR amplification, trisomy 7, loss of chromosome 10 or 10q). The association of Ki67 index, clinical, histopathological and molecular features with progression-free survival (PFS) and overall survival (OS) were assessed using log-rank test, cox model, and bootstrap validation. Results: Ki67 indices ranged from 0.3 to 73% (median 19%) and there was high inter- and intraobserver agreement (CCC > 0.9, k > 0.7). High Ki67 index (cut-off 19%) was significantly associated with presence of high cellularity, endothelial abnormalities, mitoses and necrosis. At multivariate analysis, low Ki67 index (cut-off 19%; p = 0.001), 1p19q loss (p = 0.01) and absence of EGFR amplification (p = 0.03) correlated significantly with favorable PFS. Low Ki67 index (p < 0.01) and 1p19q loss (p = 0.02) were independently and significantly associated with favorable OS. Conclusions: Ki67 index is a strong and reproducibly assessable prognostic factor in anaplastic oligodendroglial tumors. A cut off value of 0.19 is suggested. Validation in a prospective trial is needed. No significant financial relationships to disclose.