Abstract
Interleukin 2 (IL-2) is an important cytokine that controls the proliferation and differentiation of not only T- but also B-lymphocytes. Recently, we reported that CD25 (IL-2 receptor alpha chain, IL-2R) is expressed in diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL), and high expression of CD25 in the two types of lymphoma is correlated with a poor prognosis following chemotherapy regimens containing rituximab (ASH annual meeting, 2011 118:2666, 2012 120:1543). We evaluated the clinical significance of CD25 expression in a larger series of different mature B-cell lymphomas (BCL).
Four hundred and thirty-seven newly diagnosed patients who were admitted to our hospital between 2002 and 2013 were retrospectively evaluated. Lymph node or related tissue biopsy samples of BCL were analyzed using flow cytometry, as follows: 182 patients, DLBCL; 92, FL; 48, chronic lymphocytic leukemia (CLL); 21, mantle cell lymphoma (MCL); 23, marginal zone lymphoma (MZL); 8, Burkitt lymphoma (BL); 18, B-cell lymphoma unclassifiable with features intermediate between BL and DLBCL (BL/DLBCL); 5, lymphoplasmacytic lymphoma (LPL); and 39, reactive lymphadenopathy with sufficient B-cells. CD25-positivity was defined as >20% of clonal B-cells in a gated region.
CD25 expression in patients with MCL, CLL, MZL, and DLBCL was significantly higher than that in patients with reactive lymphadenopathy (P<0.001,<0.001, =0.019, and <0.001, respectively). BL and FL, which were derived from germinal center B-cells, did not express CD25. These results indicate that pre- or post- germinal center-derived B-cells, activated by IL-2/IL-2R signaling, may give rise to CD25+ BCL such as CD25+ MCL, CLL, MZL, and DLBCL.
The highest median CD25 expression (41.5%) was observed in MCL. CD25 expression was higher in MCL than CD5+ BCL (CLL and CD5+ DLBCL) (median, 41.5 vs. 16.9%, respectively; P<0.001). With a cut-off value of 60% CD25-positivity, patients with CD25-high (>60%) MCL (n=9) were not treated with aggressive chemotherapy regimens such as Hyper-CVAD due to their age and characteristics, compared with those with CD25-low (<60%) MCL (n=12) (11.1 vs. 72.7%, respectively, P=0.021).
In patients with CLL, the range of CD25 expression was wide (0.4-90.7%), and 29 patients (60%) showed CD25-positivity (CD25+ CLL). CD25+ CLL showed higher soluble IL-2R (sIL-2R) levels and an inferior overall survival (OS) than CD25- CLL (median sIL-2R, 2,195 vs. 706 U/ml P=0.047; 5-year OS, 62.7 vs. 100%; P=0.037). There was a significant correlation between levels of CD25 and sIL-2R (r=0.53, P=0.0053).
It is clinically important to distinguish between DLBCL and BCL involving MYC oncogene rearrangement (BL and BL/DLBCL, MYC+ BCL). The former showed higher CD25 expression than the latter (median, 10.2 vs. 2.1%, respectively, P=0.04). The progression-free survival rate (PFS) after rituximab containing chemotherapy was inferior in patients with CD25+ DLBCL (n=72) than those with CD25- DLBCL (n=110) and MYC+ BCL (5-year PFS, 49 vs. 70.4, 66.3%, respectively). In patients with DLBCL, central nerve system (CNS) involvement was observed in 15 patients (7 at diagnosis and 8 at relapse). CD25+ DLBCL showed a higher frequency of CNS involvement than CD25– DLBCL (13.8 vs. 4.5%, respectively, P=0.049).
Regarding MZL, CD25 was highly expressed in nodal MZL, but it showed a low expression in splenic MZL. Regarding the sites of extranodal MZL, CD25 expression was lower in the thyroid than at other sites (median, 5.1 vs. 21.2%, respectively, P=0.37). There were some differences between CD25+ (n=9) and CD25- (n=14) MZL concerning the presence of B symptoms (33.3 vs. 0%, respectively) and advanced stage (66.6 vs. 35.7%, respectively).
CD25 expression using flow cytometry can potentially provide diagnostic and prognostic implications on BCL patient. The high expression of CD25 in MCL and CLL suggests the possibility of targeted anti-CD25 immunotherapy. These findings may shed light on the role of CD25 expression in B-cell lymphomagenesis.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.