Standard Chemotherapy With or Without High-Dose Chemotherapy for Aggressive Non-Hodgkin's Lymphoma: Randomized Phase III EORTC Study

BACKGROUND: The long-term outcome for patients with aggressive non-Hodgkin's lymphoma (NHL) is poor. Consequently, the European Organization for Research and Treatment of Cancer Lymphoma Group designed a prospective randomized trial to investigate whether high-dose chemotherapy plus autologous bone marrow transplantation (ABMT) after standard combination chemotherapy improves long-term survival. METHODS: Patients aged 15-65 years with aggressive NHL received three cycles of CHVmP/BV polychemotherapy (i.e., a combination of cyclophosphamide, doxorubicin, teniposide, and prednisone, with bleomycin and vincristine added at mid-cycle). After these three cycles, patients with a complete or partial remission and at that time no lymphoma involvement in the bone marrow were randomly assigned to the ABMT arm (a further three cycles of CHVmP/BV followed by BEAC i.e., a combination of carmustine, etoposide, cytarabine, and cyclophosphamide chemotherapy and ABMT) or to the control arm (five more cycles of CHVmP/BV). All statistical tests are two-sided. RESULTS: From December 1990 through October 1998, 311 patients (median age = 44 years) were registered and received the first three cycles of CHVmP/BV, and 194 patients were randomly assigned to the treatment arms. Approximately 70% (140 patients) of these patients were of low or low-intermediate International Prognostic Index (IPI) risk. After a median follow-up of 53 months, an intention-to-treat analysis showed a time to disease progression and overall survival at 5 years of 61% (95% confidence interval CI = 51% to 72%) and 68% (95% CI = 57% to 79%), respectively, for the ABMT arm and 56% (95% CI = 45% to 67%) and 77% (95% CI = 67% to 86%), respectively, for the control arm. Differences between arms were not statistically significant. A subset analysis on IPI risk groups, although too small for reliable statistical analysis, yielded similar results. CONCLUSIONS: Standard combination therapies remain the best choice for most patients with aggressive NHL. We recommend that patients with IPI low or low-intermediate risk not be subjected to high-dose chemotherapy and ABMT as a first-line therapy.