Treatment Effect of Rituximab Plus Chemotherapy Is Obtained by Improving Survival from Non-Germinal Center-Type Untreated Diffuse Large B-Cell Lymphoma.

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous disorder that can be classified into at least three subgroups according to DNA microarray-based expression profiles: germinal center B-cell-like (GC) type, activated B-cell-like type, and type 3. Immunohistochemistry used to determine the GC and non-GC subtypes of DLBCL and to predict survival show results similar to those of cDNA microarray analysis. There are a few reviews of the treatment effect of rituximab in relation to immunohistological protein expression, but there are no reports comparing clinical outcomes of GC-type DLBCL with non-GC-type DLBCL for patients treated with rituximab plus chemotherapy. We analyzed relations between the treatment effect of rituximab plus conventional chemotherapy and immunohistochemically determined subtypes in cases of untreated DLBCL. Immunohistochemical stain results for CD10, bcl-6 and MUM1 were used to subclassify the case as same as previous reported studies. Sixty-nine patients were divided into 2 treatment groups, rituximab plus TCOP (cyclophosphamide, therarubicin, vincristine, and prednisolone) (R-TCOP) and TCOP alone. The R-TCOP group included 38 patients (12 women, 26 men; age range, 53–89 years; mean age, 68.1 years); median follow-up of surviving patients was 13 months. The TCOP group included 31 patients (15 women, 16 men; age range, 45–88 years; mean age, 66.4 years); median follow-up of surviving patients was 25 months. Of the 69 DLBCLs, 26 (38%) were considered GC type and 43 (65%) were considered non-GC type by immunohistochemical analysis. In the R-TCOP group, 18 (48%) DLBCLs were classified as GC type and 20 (52%) were classified as non-GC type. In the TCOP group, 8 (26%) DBLCLs were classified as GC type and 23 (74%) cases were classified as non-GC type. Overall survival (OS) was significantly longer for patients treated with R-TCOP than for those treated with TCOP alone (P=0.002). In both the R-TCOP and TCOP groups, no significant difference in OS was observed between GC-type and non-GC-type DLBCLs (P=0.36 and P=0.16, respectively). Among GC-type cases, OS did not differ significantly between the two treatments, but among non-GC DLBCLs, OS was significantly longer for patients treated with R-TCOP than for those treated with TCOP alone (P=0.005). Furthermore, among the low IPI, non-GC-type cases, OS did not differ significantly between the two treatments (P=0.8), but among the high IPI, non-GC-type cases, a significant difference was observed between the two treatments (P=0.008). In prior studies, the addition of rituximab to conventional chemotherapy was shown to improve outcome in patients with DLBCL. Our findings suggest that these improved outcomes may be due primarily to the beneficial effect of rituximab added to chemotherapy on non-GC DLBCLs, especially when the IPI is high. The non-GC type includes activated B-cell-like DLBCL, which is characterized by activation of NF-kappa B target genes. Rituximab inhibits the constitutive nuclear factor-kappa B signaling pathway in non-Hodgkin lymphoma cell lines; therefore, we believe our results point to the clinical importance of NF-kappa B as a therapeutic target for DLBCL.