|
|
Clinical prognostic
factors affecting survival in patients with newly diagnosed Glioblastoma
Multiforme (GBM)
T. Gorlia, R. Stupp, E. A. Eisenhauer, R. O. Mirimanoff, M. J. Van Den Bent,
K. Belanger, D. Lacombe, A. Allgeier, European Organisation for Research and
Treatment of Cancer (EORTC) Brain And Radiotherapy Groups, National Cancer
Institute of Canada (NCIC) Clinical Trials Group
EORTC Data Center, Brussels, Belgium; Centre Hospitalier
Universitaire Vaudois, Lausanne, Switzerland; NCIC Clinical Trial Group,
Kingston, ON, Canada; Erasmus University Medical Center, Rotterdam, Netherlands;
Hopital Notre-Dame du CHUM, Montréal, PQ, Canada
Background.
In the EORTC 26981-22981/NCIC CE3 phase III trial, 573
pts with newly diagnosed GBM were randomized to receive radiotherapy (RT) or RT
plus Temozolomide (results separately reported). [BrainLife]
In this analysis we evaluate clinical parameters as prognostic factors for
survival.
Methods.
The Cox Proportional Hazard model was used to assess factors
related to patients’ characteristics (age, gender, WHO Performance Status
(PS), Mini Mental State Examination (MMSE)), disease history (extent of
resection, time between surgery and start of radiotherapy, administration of
corticosteroids, baseline haematological and non-haematological toxicity
status), tumor location (hemisphere and lobe).
These factors were first screened by univariate technique and the hemisphere was
the only variable that did not pass the 10% significance level.
Results.
In the multivariate analysis, poorer prognosis was associated
with the extent of surgery (biopsy worse than partial or complete debulking)
(p<0.0001), greater age (p=0.016), male gender (p=0.013), lower MMSE Score
(p<0.0001), the administration of corticosteroids (p=0.012), tumor location
in the frontal lobe (p=0.002) or tumor on more than one lobe (p=0.02).
The time between surgery and start of radiotherapy, the baseline haematological
and non-haematological toxicity status and the WHO PS status were not
retained.
Gender seems related to overall survival but probably because the proportion of
male patients with a tumor on more than one lobe was significantly higher
(p<0.0001).
Conclusions.
This analysis confirms the extent of surgery and age as
major prognostic factors.
MMSE has a better correlation to survival than WHO PS.
The model will be validated with the bootstrap technique and on an independent
data set.
The results should be taken into consideration for future EORTC glioma trials.
Copyright 2004 American Society of Clinical Oncology All rights
reserved worldwide.
Source: http://www.asco.org/ac/1,1003,_12-002636-00_18-0026-00_19-001530,00.asp
|
