Proceedings of the 40th Annua l A S T R O Meeting

2075 RESULTS OF DEFINITIVE SURGERY AND RADIATION THERAPY IN ELDERLY PATIENTS WITH SUPRATENTORIAL

GLIOBLASTOMA MULTI~ORME

John H. Suh*, Dasarahally S. Mohan*, Jennifer L. Phan'l', Patrick A. Kupelian*, Bruce H. Cohen$ and Gene H. Barnett

Departments of *Radiation Oncology, *Neurology and ~Neurosurgery, The Cleveland Clinic Foundation, Cleveland, OH, tUniversity of South Carolina, School of Medicine, SC..

Purpose: To determine the efficacy of definitive surgery and radiation in patients aged 70 years and older with supratentorial gliobiastoma multiforme.

Materials & Methods: All elderly patients (_> 70 years) who had primary treatment for glioblastoma multiforme at our tertiary care institution from 1977 through 1997 was analyzed. The study group (n=l I 1) included 61 patients treated with definitive radiation, 23 treated with palliative radiation and 27 who received no radiation. To compare our results with published findings, we grouped our patients according to the applicable prognostic categories developed by the Radiation Therapy Oncology Group (RTOG): RTOG group IV (n = 7), V (n = 71), and VI (n = 33). Patients were retrospectively assigned to prognostic group IV, V, or VI based on age, performance status, extent of surgery, mental status, neurologic function and radiation dose. Treatment included surgical resection and radiation (n = 54), biopsy alone (n = 29), and biopsy followed by radiation (n = 28). Patients were also stratified according to whether they were optimally treated (gross total or subtotal resection with postoperative definitive radiation) or suboptimally treated (biopsy, biopsy + radiation, surgery alone, or surgery + palliative radiation). Patients were considered to have a favorable prognosis (n = 41) if they were optimally treated and had a Karnofsky performance status (KPS) score of at least 70.

Results: The median survival for patients according to RTOG groups IV, V, and VI was 9.6, 6.7, and 3.4 months, respectively (logrank, P <0.001 ). The median overall survival was 5.7 months. The definitive radiation group (n=61) had a median survival of 7.4 months compared to 2.3 months in the palliative radiation group (n=23) and 1.3 months in the biopsy-alone group (P <0.0001). Optimally treated patients had a median survival of 7.4 months compared to 3.1 months in those suboptimally treated (P <0.0001). The favorable prognosis group had 9.1 months median survival compared to 2.4 months in the unfavorable group (P<0.0001). On multivariate analysis, the KPS, RTOG group, favorable/unfavorable prognosis group, and optimal treatment/suboptimal treatment were significant predictors of survival (P =0.003).

Conclusions: Elderly patients with good performance status (> 70 KPS) when treated aggressively with maximal resection and definitive radiation had better results than those treated with palliative radiation and biopsy. Aggressive treatment in such patients should be considered.

265

2076 A PHASE II TRIAL OF WEEKLY PACLITAXEL AND CONVENTIONAL RADIOTHERAPY FOR SUPRATENTORIAL GLIOBLASTOMA MULTIFORME: RTOG 96-02

James Ruffer ~, Charles Scott 2, Corey Langer 3, Benjamin Movsas 3, Kevin Murray 4

l.University of Pennsylvania, Philadelphia, PA, 2. RTOG, Philadelphia, PA, 3. Fox Chase Cancer Center, Philadelphia, PA, 4. Medical College of Wisconsin, Milwaukee, WI.

Purpose/Objective: Fractionated external beam radiotherapy (EBRT) with or without BCNU has remained the standard of care for patients with glioblastoma multiforme (GBM), but survival results remain poor. In phase I studies, the maximal tolerated dose (MTD) of paclitaxel combined with EBRT for patients with primary CNS malignancy is much higher than in other disease sites with oz" without concurrent radiation. Paclitaxel 3-hoar infusion is more convenient and less myelosuppressive than 24-hour infusion. We mounted a phase II study to determine the feasibility and efficacy of conventional EBRT combined with weekly paclitaxel at its MTD

in patients with GBM.

Materials & Methods: Between 8/16/96 and 3/21/97, 62 patients with the histologic diagnosis of GBM were enrolled in a multi- institutional phase II trial evaluating concurrent EBRT and weekly paclitaxel 225 mg/m-/3hr infusion (1-3 hours prior to radiation) given the first treatment day of each radiation week. The total radiation dose was 60Gy in 200cGy fractions, 5 days per week with a conedown to gross disease with margin at 46Gy.

Results: 59 patients (95%) were evaluable (2 canceled, I had no on study form). Median age is 54 years (range 28-78); 75% were older than 50 years of age. Recursive Partitioning Analysis (RPA) classes III, IV, V, and VI included 10 (17%), 20 (34%), 24 (41%), and 5 (8%) patients respectively. Seven acute grade IV toxicities were seen (2 cutaneous, 2 pulmonary, 1 neurologic, and 1 allergic). 19 (33%) sustained a maximum grade IlI toxicity. No grade Ill or IV neutropenia or thrombocytopenia was observed. 91% of patients received acceptable (per protocol) dose delivery of the paclitaxel with 75% completing chemotherapy. Of 35 patients with measurable disease, complete or partial responses were observed in 23% of patients, with 17°7o showing progression at first follow-up. With median potential follow-up of 9,0 months, The estimated median survival is 7.5 months with median survivals for RPA IIL IV, V, and VI being not reached, 11.1, 6.0, and 2.5 months respectively, Twenty-six patients remain alive and evaluable at last follow-up.

Coneluslon: Delivery of concurrent full dose EBRT and weekly high dose paclitaxel was achievable in the majority of patients with acceptable acute grade IV toxicity and modest myelosuppression. Despite a considerably higher overall dose intensity of paclitaxel compared to that achievable in other disease sites, the overall median survival (and median survival by RPA class) remains comparable to historical controls treated with standard radiation and BCNU. Because the results of EBRT and short course chemotherapy appear comparable to a full year of BCNU, concurrent paclitaxel and radiation, with a variation of infusion technique, may still lead to clinical benefit and an evaluation of quality adjusted survival remains to be performed.