Differences in BCL-2 Protein and t(14;18) in Patients with Primary Rituximab/Chemotherapy Refractory Diffuse Large B-Cell Lymphoma (DLBCL).

Background: Aberrant over-expression of BCL-2 is a poor prognostic factor in patients with DLBCL. t(14;18) (q32;q21) results in transcription of the anti-apoptotic BCL-2 protein. In vitro studies have shown that rituximab induces an IL-10 dependent down regulation of BCL-2 and sensitizes cancer cells to the effect of chemotherapy agents. It has been reported that the addition of rituximab to chemotherapy has eliminated the prognostic significance of BCL-2 over-expression in DLBCL. However limited information is available regarding the prognostic significance of BCL-2 over-expression driven by t(14;18) following rituximab/chemoimmunotherapy. To this end we correlate the levels of Bcl-2 expression and response to front-line treatment in patients with primary refractory DLBCL (PR-DLBCL).

Methods: A single-institution retrospective data analysis was done to identify all DLBCL patients treated between 1/1/02 and 12/31/06. Patients with PR-DLBCL, as defined by residual disease at the end of primary treatment or disease relapse within 6 months of completing it were identified. Patient demograpics, international prognostic index (IPI) score, treatments administered, outcome and Bcl-2 status were reviewed. Bcl-2 translocation was studied by a nested PCR-based assay of peripheral blood samples. The test is reproducibly sensitive to a detection level of 1:10⁵ and reported the presence or absence of t(14;18) as a positive or negative result. BCL-2 protein expression in tissue was studied with immunohistochemistry (IHC).

Results: A total of 246 patients with DLBCL were identified. Of these, 27 (11%) qualified as PR-DLBCL. Median age at diagnosis was 56 years (range 16–84). Thirteen (48%) were females and 14 (52%) were males. IPI score was 0 in 2 (7.5%), 1 in 9 (33.3%), 2 in 6 (22.2%), 3 in 4 (14.8%) and 4 in 6 (22.2%) patients. Disease was labeled ‘bulky’ (> 5 cm lymphadenopathy) in 11 (41%) patients at the time of diagnosis. Twenty four (89%) patients received an initial rituximab-containing regimen and radiation therapy was administered to 13 (48%) patients. Median overall survival for all the 27 patients was 570 days (range 151–1437). Ten (37%) patients received peripheral blood stem cell transplantation during the course of their disease. At the time of analysis, 14 (52%) patients were alive and of these, 7 (50%) had persistent disease, while the other 7 (50%) were disease-free. t(14;18) data was available on 18 of these 27 patients and all were negative for the translocation. All of these 18 patients had received rituximab-containing initial therapy. Of these 18 patients, IHC for BCL-2 expression was available for 8 (44%), and was positive in all 8 patients.

Conclusions: Bcl-2 over-expression is observed in a significant number of patients with primary rituximab/chemotherapy refractory DLBCL. In our selected group of patients, the t(14;18) does not appear to play a role in the expression of BCL-2, suggesting the existence of additional mechanisms responsible for BCL-2 upregulation. In addition, 70% of our PR-DLBCL patients did not express BCL-2, leading to the hypothesis that various regulatory proteins such as other members of the BH3 domain family play a role in determining rituximab/chemotherapy sensitivity.