Secondary Analysis of Radiation Therapy Oncology Group study (RTOG) 9310: An Intergroup Phase II Combined Modality Treatment of Primary Central Nervous System Lymphoma

Abstract

To determine whether a lower dose of hyperfractionated whole brain radiation reduces central nervous system morbidity without compromising survival for primary CNS lymphoma (PCNSL) patients receiving combined modality treatment. One hundred and two patients received a course of pre-radiation chemotherapy, followed by whole brain radiation, followed by cytosine-arabinoside. Initial radiation dose was 45 Gy/25 fractions (RT) then the study was amended to reduce this dose for complete responders to induction chemotherapy to 36 Gy/30 fractions/3 weeks (HFX). Eighty-two patients received radiotherapy and were evaluable for toxicity analysis (66 RT patients and 16 HFX patients). MMSE scores and survival for the 40 patients who received radiotherapy after complete response to chemotherapy (27 RT and 13 HFX) were compared. There were no notable differences in pre-treatment patient characteristics between the RT and HFX groups. Neurotoxicity: By 4 years, there were 8/82 (10%) grade 5 neurotoxicities which included 2/16 (13%) grade 5 encephalopathies and 0/27 in the RT group of complete responders to chemotherapy. Survival: There was no statistically significant difference in overall or progression-free survival (PFS) between the chemotherapy-complete responders who received RT and HFX. Cognitive function testing: MMSE scores improved at 8 months across both treatment groups. Analysis of the area under the MMSE curve at 8 months showed no statistically significant difference between RT and HFX groups (P=0.81). Leukoencephalopathy occurred later in the HFX group than in the RT patients. Although the HFX schedule represented a 25% reduction in biologically effective tumor dose in comparison, PFS and overall survival were not significantly affected. The HFX regimen delayed but did not eliminate severe neurotoxicity from chemoradiation in PCNSL patients.