Background: The therapeutic impact of tumor resection is poorly defined. Therefore the current study was conducted.
Methods: A retrospective, 2-institutional study was conducted (1991-1994) to compare the treatment results of stereotactic biopsy plus radiation therapy (99 patients; tumor dose: 60 gray [Gy]) with those of surgical resection plus radiation therapy (126 patients; tumor dose: 60 Gy). Only adult patients with supratentorial, lobar located, de novo glioblastoma were included. Survival time was analyzed with the Kaplan-Meier method. Prognostic factors were obtained from the multivariate Cox proportional hazards model.
Results: Patients were categorized in the Radiation Therapy Oncology Group (RTOG) Classes IV (46 patients), V (157 patients), and VI (22 patients). The resection group and the biopsy group did not differ in terms of age, pretreatment Karnofsky performance status KPS), gender, duration of symptoms, presenting symptoms, tumor location, tumor size, and the frequency of midline shift. Patients in the biopsy group more often were found to have left-sided tumors (P < 0.001). Transient perioperative morbidity and mortality rates were 1% and 1%, respectively, in the biopsy group and 5% and 1.6%, respectively, in the resection group (P > 0.05). The median survival time was 37 weeks for the resection group and 33 weeks for the biopsy group. The difference was not statistically significant (P = 0.09). The most favorable pretreatment prognostic factor was patient age < 60 years (P < 0.01). Tumor resection was highly effective in patients with midline shift (P < 0.01). In patients without midline shift radiation therapy alone was found to be as effective as tumor resection plus radiation therapy (P = 0.5). Patients with midline shift were more likely to have a worse KPS during the course of primary radiation therapy (P < 0.05).
Conclusions: For RTOG Classes IV-VI patients with moderate mass effect of the tumor, radiation therapy alone is a rational treatment strategy. Tumor resection should be performed in patients with pretreatment midline shift whenever possible.
Copyright 1999 American Cancer Society.