Diffuse large B-cell lymphoma (DLBCL) with bone marrow (BM) involvement. clinical presentation, central nervous system (CNS) relapses, and outcomes of 121 patients treated at M. D. Anderson Cancer Center.

Abstract

8072 Background: DLBCL with BM involvement is an uncommon condition, and it is suspected to be a more aggressive condition with worse outcome and increased risk for CNS. There is controversy about the best modality of treatment, and it is unclear if CNS prophylaxis should be performed in every patient. Methods: We examined a database of new patients seen at M. D. Anderson Cancer Center from 1998 to 2008. We found patients who had DLBCL and LBCL in the bone marrow (140 patients with discordant histology or HIV positive serology was excluded). Patient characteristics included 51% male, 73% elevated LDH, and 53% older than 60 years. IPI distribution included 4% low-risk, 16.5% IL, 37% IH, 40% high-risk (2% N/A). All patients received rituximab CHOP-like therapy, and 52% received intrathecal prophylaxis. Results: With a median follow-up of 3.5 years for living patients, 45 of patients have developed relapse, with a median time to progression of 3.1 years. The 5-year overall survival rate was 51%; survival was associated with IPI (p=0.02), with 46% of the patients with high-risk dying of their disease. There were 10 (9%) CNS relapses, 5 in the prophylaxis and 5 in the non-prophylaxis group. All CNS relapses occurred in patients with elevated serum LDH at diagnosis, and in those with high- intermediate (50%) and high (50%) scores. Risk IPI scores nine of 10 patients with CNS relapse died of their disease. There were no differences between the CNS relapse rates for high-risk patients whether they received CNS prophylaxis (10% CNS relapse) or not (16% CNS relapse, p=0.4). Conclusions: In our patients, DLBCL with BM involvement commonly presented with high-risk features, CNS relapses occurred mostly in patients with HI and H-IPI scores and elevated LDH. The role of CNS prophylaxis in prevention of relapse remains unclear and new options for prevention of CNS relapse should be considered. No significant financial relationships to disclose.