Document Type


Publication Date



This article has been peer reviewed. It was published in: British Journal of Cancer.

Volume 104, Issue 9, April 2011, Pages 1365-71.

The published version is available at DOI: 10.1038/bjc.2011.123. Copyright © Nature


BACKGROUND: Patients with high-grade gliomas are treated with surgery followed by chemoradiation. The risk factors and implications of neurological side effects are not known.

METHODS: Acute and late ≥ grade 3 neurological toxicities (NTs) were analysed among 2761 patients from 14 RTOG trials accrued from 1983 to 2003. The association between acute and late toxicity was analysed using a stepwise logistic regression model. The association between the occurrence of acute NT and survival was analysed as an independent variable.

RESULTS: There were 2610 analysable patients (86% glioblastoma, 10% anaplastic astrocytoma). All received a systemic agent during radiation (83% chemotherapy, 17% biological agents). Median radiation dose was 60 Gy. There were 182 acute and 83 late NT events. On univariate analysis, older age, poor performance status, aggressive surgery, pre-existing neurological dysfunction, poor mental status and twice-daily radiation were associated with increased acute NT. In a stepwise logistic regression model the occurrence of acute NT was significantly associated with late NT (OR=2.40; 95% CI=1.2-4.8; P=0.014). The occurrence of acute NT predicted poorer overall survival, independent of recursive partitioning analysis class (median 7.8 vs 11.8 months).

INTERPRETATION: Acute NT is significantly associated with both late NT and overall survival.

PubMed ID


Included in

Oncology Commons