Prognostic impact of concordant and discordant bone marrow involvement and cell-of-origin in Korean patients with diffuse large B-cell lymphoma treated with R-CHOP

Aims Previous studies have suggested many prognostic factors in diffuse large B-cell lymphoma (DLBCL), but the prognostic importance of cell-of-origin and discordant bone marrow involvement remains unclear. The aim of this study was to evaluate the prognostic impact of bone marrow involvement histological subtype, cell-of-origin subtype and international prognostic index (IPI) scores in patients with DLBCL. Methods Patients who were newly diagnosed with DLBCL and treated with rituximab plus cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP) were analysed. Clinical information was reviewed retrospectively. Patients were classified into negative, concordant and discordant bone marrow involvement by histological review. The cell-of-origin types were defined using immunohistochemical analysis. Results Both concordant and discordant bone marrow involvement had a negative prognostic impact on progression-free survival, independent of the standard and National Comprehensive Cancer Network (NCCN) IPI scores and cell-of-origin. Patients with non-germinal centre B-cell type showed significantly shorter progression-free survival than those with germinal centre B-cell type. However, non-germinal centre B-cell type did not have a prognostic impact on progression-free survival or overall survival after controlling for the standard and NCCN-IPI and bone marrow involvement. Conclusions Both concordant and discordant bone marrow involvement had an adverse prognostic impact on progression-free survival and overall survival; this was independent of the standard and NCCN-IPI and cell-of-origin (non-germinal centre B-cell type). The NCCN-IPI had more powerful prognostic value than the standard IPI (sIPI). The non-germinal centre B-cell type lost significant prognostic impact on progression-free survival after adjustment for standard and NCCN-IPI and bone marrow involvement.