Staging and Prognosis | Treatment > Thromboembolism

Meeting Abstract

Venous thromboembolism is common but does not alter survival in glioblastoma multiforme

A. Shet, A. E. Schorer, M. A. Kuskowski

University of Minnesota, Minneapolis, MN

To determine venous thromboembolism (VTE) rate and impact on survival in patients with primary central nervous system (CNS) malignancy, we performed a retrospective analysis of 185 consecutive adult patients.
Diagnoses required histological confirmed glioblastoma multiforme (GBM), other high grade gliomas, or another CNS malignant tumor.
All patients were diagnosed and treated at the Minneapolis VA Medical Center and identified by search of pathology and tumor registry records.
The diagnosis of VTE was confirmed by duplex ultrasonography, venography, ventilation-perfusion scanning or high resolution CT scanning.
We assigned patients to two groups: GBM (n=98) and non GBM (n=87) and performed statistical analysis using the following variables:
incidence of VTE, overall survival (OS), thrombosis free survival (TFS), and post VTE survival.
There were 181 (97.8%) males and 4 (2.2%) females with a mean age of 56.5 years (range 19-86).
VTE developed in 27 patients (14.4 %), with 21 having lower extremity deep vein thrombosis (DVT) and 6 having pulmonary emboli (PE).
The rate of VTE was higher (p=0.05) in GBM patients, but owing to the longer survival of non GBM patients, the cumulative risk for GBM (n=17 events; 17% VTE from diagnosis to death) and non GBM (n= 10 events; 11% VTE from diagnosis to death) patients was similar (p =0.20).
Between GBM and non GBM patients, there was significantly shorter OS (301±87 vs. 2613±223 days; p< 0.0001) and TFS (287±87 vs. 2545±325 days; p<0.0001).
VTE risk was not clustered around the time of biopsy or excision.
Rather, GBM patients remained at a constant high risk for VTE until death.
The development of VTE was not predictive for poorer survival in patients with CNS malignancy.
In particular, when GBM patients with VTE were compared to those without VTE, survival was the same (p= 0.26).
These data confirm previous speculations that GBM is associated with a high rate of VTE, but do not support the conclusion that VTE contributes to mortality in this disease.
Clinical factors that identify high risk subgroups should be determined, but if trials of anticoagulation are contemplated, survival is unlikely to be a suitable study endpoint.

© Copyright 2003 American Society of Clinical Oncology All rights reserved worldwide