Prognostic significance of Bcl-6 protein expression in DLBCL treated with CHOP or R-CHOP: a prospective correlative study

Abstract

Bcl-6 protein expression, a marker of germinal center origin, has been associated with a favorable prognosis in diffuse large B-cell lymphoma (DLBCL). To determine the prognostic significance of this marker when rituximab (R) was added to cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy, we prospectively studied Bcl-6 protein expression by immunohistochemical staining of 199 paraffin-embedded specimens from patients enrolled in the US Intergroup phase 3 trial comparing R-CHOP to CHOP with or without maintenance R. In Bcl-6(-) patients, failure-free survival (FFS) and overall survival (OS) were prolonged for those treated with R-CHOP alone compared to CHOP alone (2-year FFS 76% versus 9%, P < .001; 2-year OS 79% versus 17%, P < .001). In contrast, no differences in FFS and OS were detected between treatment arms for Bcl-6(+) cases. In the multivariate analysis, treatment arm (CHOP versus R-CHOP) was the major determinant of both FFS (P < .001) and OS (P < .001) for the Bcl-6(-) subset, whereas the International Prognostic Index risk group was the only significant predictor of outcome among Bcl-6(+) cases. Bcl-2 protein expression was not predictive of outcome in either group. In this study, we observed a reduction in treatment failures and death with the addition of R to CHOP in Bcl-6(-) DLBCL cases only. Our finding that Bcl-6(+) cases did not benefit from the addition of R to CHOP requires independent confirmation.

Figures

Figure 1.

FFS and OS based on Bcl-6 expression. Failure-free survival (A-B) and overall survival (C-D) according to Bcl-6 expression for cases treated on R-CHOP and CHOP induction arms. Analysis excludes patients who were randomized to receive MR.

Figure 2.

FFS and OS according to induction treatment. Failure-free survival (A-B) and overall survival (C-D) according to induction treatment for Bcl-6+ and Bcl-6– cases. Analysis excludes patients who were randomized to receive MR.