INFLUENCE OF LOCATION AND EXTENT OF SURGICAL RESECTION ON SURVIVAL OFPATIENTS WITH GLIOBLASTOMA-MULTIFORME - RESULTS OF 3 CONSECUTIVE RADIATION-THERAPY ONCOLOGY GROUP (RTOG) CLINICAL-TRIALS

Purpose: The influence of tumor site, size, and extent of surgery on t he survival of patients with glioblastoma multiforme treated on three consecutive prospectively randomized Radiation Therapy Oncology Group trials employing surgery and irradiation plus or minus chemotherapy wa s studied. Methods and Materials: Six hundred forty-five patients with a diagnosis of glioblastoma multiforme on central pathological review were analyzed for survival with respect to known prognostic factors, that is, age and Karnofsky Performance Status, as well as extent of su rgery, site, and size. Surgical treatment consisted of biopsy only in 17%, partial resection in 64%, and total resection in 19%. Tumors were located in frontal lobe in 43%, temporal lobe in 28%, and parietal lo be in 25%. Maximum tumor diameter as determined on computed tomography or magnetic resonance imaging scans was less than 5 cm for 38%, betwe en 5-10 cm for 56% and greater than 10 cm for 6% of patients. The exte nt of surgical therapy was the same for tumors greater than 5 or great er than 10 cm, whereas total resection was more often performed for tu mors less than 5 cm. The extent of surgery did not appear to vary with age or site. Results: Patients undergoing total resection had a media n survival of 11.3 months compared to 6.6 months for patients with a b iopsy only. A significant difference in median survival was also found for partial resection versus biopsy only treatment (10.4 vs. 6.6 mont hs). There was no difference in survival for the different tumor sizes . Patients with frontal lobe tumors survived longer than those with te mporal or parietal lobe lesions (11.4 months, 9.1 months, and 9.6 mont hs, respectively) (p = 0.01). A Cox multivariate model confirmed a sig nificant correlation of age, Karnofsky Performance Status, extent of s urgery, and primary site with survival. The best survival rates occurr ed in patients who had at least three of the following features: < 40 years of age, high Karnofsky Performance Status, frontal tumors, and t otal resection (17 months median). Conclusion: We conclude that biopsy only yields inferior survival to more extensive surgery for patients with glioblastoma multiforme treated with surgery and radiation therap y.