PION RADIATION FOR HIGH-GRADE ASTROCYTOMA - RESULTS OF A RANDOMIZED STUDY

Purpose: This study attempted to compare within a randomized study the outcome of pion radiation therapy vs. conventional photon irradiation for the treatment of high-grade astrocytomas. Methods and Materials: Eighty-four patients were randomized to pion therapy (33-34.5 Gy pi), or conventional photon irradiation (60 Gy). Entry criteria included as trocytoma (modified Kernohan high Grade 3 or Grade 4), age 18-70, Karn ofsky performance status (KPS) greater than or equal to 50, ability to start irradiation within 30 days of surgery, unifocal tumor, and trea tment volume < 850 cc. The high-dose volume in both arms was computed tomography enhancement plus a 2-cm margin. The study was designed with the po,ver to detect a twofold difference between arms. Results: Eigh ty-one eligible patients were equally balanced for all known prognosti c variables. Pion patients started radiation 7 days earlier on average than photon patients, but other treatment-related variables did not d iffer. There were no significant differences for either early or late radiation toxicity between treatment arms. Actuarial survival analysis shows no differences in terms of time to local recurrence or overall survival where median survival was 10 months in both arms (p = 0.22). The physician-assessed KPS and patient-assessed quality of life (QOL) measurements were generally maintained within 10 percentage points unt il shortly before tumor recurrence. There was no apparent difference i n the serial KPS or QOL scores between treatment arms. Conclusion: In contrast to high linear energy transfer (LET) therapy for central nerv ous system tumors, such as neutron or neon therapy, the safety of pion therapy, which is of intermediate LET, has been reaffirmed. However, this study has demonstrated no therapeutic gain for pion therapy of gl ioblastoma. (C) 1997 Elsevier Science Inc.