BACKGROUND: Incidence of central nervous system (CNS) recurrence in patients with aggressive non-Hodgkin's lymphoma who did not receive meningeal prophylaxis is about 5%. Controversy remains regarding risk factors associated with such an event preventing a rational approach of prophylactic strategies. PATIENTS AND METHODS: We analyzed a cohort of 974 patients with aggressive lymphoma in complete remission (CR). All the patients received a CNS prophylaxis consisting of intrathecal injections and intravenous high-dose methotrexate. The risk repartition on the basis of the international prognostic index (IPI) of these 974 CR-patients was low (L): 41%, low-intermediate (LI): 27%, high-intermediate (HI): 19%, high (H): 13%. RESULTS: The incidence of isolated CNS relapse was 1.6%. In a first multivariate logistic regression analysis an increased LDH (P = 0.05, RR = 5) and the presence of more than one extranodal site (P = 0.05, RR = 3) were identified as independent risk factors for isolated CNS relapse. Another multivariate analysis incorporating IPI as a unique parameter showed that only IPI remained significantly associated with a higher risk of CNS relapse (L-LI: 0.6% vs. HI H: 4.1%, P = 0.002; RR = 7). CONCLUSION: Prophylaxis notably reduces the risk of CNS recurrence in the higher risk patients. By contrast, we propose the deletion of prophylactic intrathecal injections in the lower risk patients.
BACKGROUND: Incidence of central nervous system (CNS) recurrence in patients with aggressive non-Hodgkin's lymphoma who did not receive meningeal prophylaxis is about 5%. Controversy remains regarding risk factors associated with such an event preventing a rational approach of prophylactic strategies. PATIENTS AND METHODS: We analyzed a cohort of 974 patients with aggressive lymphoma in complete remission (CR). All the patients received a CNS prophylaxis consisting of intrathecal injections and intravenous high-dose methotrexate. The risk repartition on the basis of the international prognostic index (IPI) of these 974 CR-patients was low (L): 41%, low-intermediate (LI): 27%, high-intermediate (HI): 19%, high (H): 13%. RESULTS: The incidence of isolated CNS relapse was 1.6%. In a first multivariate logistic regression analysis an increased LDH (P = 0.05, RR = 5) and the presence of more than one extranodal site (P = 0.05, RR = 3) were identified as independent risk factors for isolated CNS relapse. Another multivariate analysis incorporating IPI as a unique parameter showed that only IPI remained significantly associated with a higher risk of CNS relapse (L-LI: 0.6% vs. HI H: 4.1%, P = 0.002; RR = 7). CONCLUSION: Prophylaxis notably reduces the risk of CNS recurrence in the higher risk patients. By contrast, we propose the deletion of prophylactic intrathecal injections in the lower risk patients.
Authors: Steven H Bernstein; Joseph M Unger; Michael Leblanc; Jonathan Friedberg; Thomas P Miller; Richard I Fisher Journal: J Clin Oncol Date: 2008-12-01 Impact factor: 44.544