Impact of concordant and discordant bone marrow involvement on outcome in diffuse large B-cell lymphoma treated with R-CHOP.

PURPOSE: In diffuse large B-cell lymphoma (DLBCL), prior studies suggest that concordant bone marrow involvement with DLBCL portends a poorer prognosis, whereas discordant bone marrow involvement with small B-cell lymphoma does not. We examined the significance of bone marrow involvement in patients treated in the current era of therapy including rituximab. PATIENTS AND METHODS: We performed a retrospective analysis of the prognostic impact of bone marrow involvement in an unselected population of patients with newly diagnosed DLBCL treated with rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone in British Columbia and Auckland, New Zealand, with complete clinical information and evaluable staging bone marrow biopsies. RESULTS: In total, 795 patients were identified. Six hundred seventy (84.3%) of 795 had a negative bone marrow, 67 patients (8.4%) had concordant and 58 (7.3%) had discordant involvement. Median follow-up was 41 months (range, 1 to 115). Progression-free survival (PFS) was inferior in those with concordant (P < .001) and discordant (P = .019) involvement while overall survival (OS) was inferior in those with concordant involvement (P < .001) only. In a multivariate analysis controlling for the International Prognostic Index (IPI) score, concordant involvement remained an independent predictor of PFS (P < .001) and OS (P = .007). Discordant involvement was associated with older age, elevated lactate dehydrogenase, advanced stage, and increased number of extranodal sites and was not a negative prognostic factor independent of the IPI score. CONCLUSION: The negative prognostic impact of discordant involvement is adequately represented by the IPI score, while the risk with concordant involvement is greater than that encompassed by this predictor. The results emphasize the need for accurate staging assessment of bone marrow involvement in DLBCL.