D cyclins positively regulate the cell cycle and mediate the pathogenesis of some lymphomas. Cyclin D1 overexpression is the hallmark of mantle cell lymphoma, whereas cyclins D2 and D3 are reportedly not as specific to certain lymphomas as cyclin D1. In this study, cyclin D2 was found to be overexpressed in 98% of de novo CD5-positive diffuse large B-cell lymphomas (DLBCLs) (50/51) and in 28% of CD5-negative DLBCLs (14/51). A statistically significant difference was observed between these two groups (p<0.0001). In contrast, no statistical difference was found in the cyclin D3 expression between CD5-positive (18/51) and CD5-negative (24/51) DLBCLs (p=0.23). Based on these findings, cyclin D2 is therefore considered to be closely associated with de novo CD5-positive DLBCLs. This insight may be useful for overcoming the inferior survival of this aggressive lymphoma.
The online version of this article (doi:10.1186/1746-1596-8-81) contains supplementary material, which is available to authorized users.
Competing interests
The authors declare that they have no competing interest.
Authors’ contributions
Conceived and designed the experiments: YS, TI. Performed the experiments: TI, YS. Analyzed the data: YS, TI, TY, KT, NI, TT, YM, NA, YO. Contributed reagents/materials/analysis tools: NN, SN, YM. Wrote the paper: TI, YS, TY. All authors read and approved the final manuscript.
Cyclophosphamide, doxorubicin, vincristine, and prednisone
R-CHOP
Rituximab with cyclophosphamide, doxorubicin, vincristine, and prednisone
RT-PCR
Reverse transcription polymerase chain reaction
NFkB
Nuclear factor kappa beta.
Background
D-type cyclins (D1, D2, and D3) positively regulate the cell cycle and mediate the pathogenesis of some lymphomas[1]. The overexpresison of cyclin D1 due to translocation t (11;14)(q13;q32) serves as a hallmark for the diagnosis of mantle cell lymphoma[1]. Recently, a rare type of cyclin D1-negative mantle cell lymphoma, which overexpresses cyclins D2 or D3, was identified[2]. Cyclins D2 and D3 are also detected in various lymphomas, but they have not been shown to be closely associated with any particular subtype of lymphoma[3].
We recently demonstrated cyclin D2 to be overexpressed in the proliferation centers of chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL)[4]. CLL/SLL is considered to be a disease caused by the accumulation of CD5-positive B lymphocytes, and proliferating tumor cells are located in the proliferation centers of lymph nodes[5]. The proliferating cells are composed of paraimmunoblastic cells and most of them are in the cell cycle, expressing Ki-67 antigen[5]. These findings suggest that de novo CD5-positive diffuse large B-cell lymphomas (CD5+ DLBCLs) share the characteristics of actively proliferating cells in CLL/SLL.
Anzeige
De novo CD5+ DLBCLs account for approximately 10% of all DLBCL cases, and they are characterized by a female predominance, a higher age at diagnosis, and a high degree of central nervous system relapse[6]. De novo CD5+ DLBCLs are also known to show a significantly poorer survival outcome than CD5-negative DLBCLs (CD5− DLBCLs) under both cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) and rituximab-CHOP (R-CHOP) therapy[6, 7].
According to these previous data, we sought to clarify the expression patterns of cyclins D2 and D3 in de novo CD5+ DLBCLs.
Methods
Case selection
We studied 51 Japanese patients with de novo CD5+ DLBCLs diagnosed between 1998 and 2011 at Okayama University, Tokai University, and Nagoya University. The patients included 26 males and 25 females between 32 and 90 years of age (median age 68 years). The examined tissue specimens were obtained from 32 lymph nodes and 19 extranodal sites. The CD5 antigen expression was examined by means of immunohistochemistry of paraffin sections and/or flow cytometry. All samples were immunohistochemically confirmed to be cyclin D1 and sox11 negative[2]. Any samples with a history of other lymphoproliferative disorders were excluded from the study.
As a control group, samples taken from 51 patients with CD5− DLBCLs diagnosed between 1997 and 2011 at Okayama University were also examined. The patients included 27 males and 24 females between 23 and 89 years of age (median age 68 years). The examined tissue specimens were obtained from 42 lymph nodes and 9 extranodal sites. In all cases, the CD5 antigen expression was examined by both immunohistochemistry and flow cytometry.
Anzeige
Histological examination and immunohistochemistry
The tissue samples were fixed in 10% formalin and embedded in paraffin. The sections (4-μm thick) were stained with H&E. Immunohistochemistry was performed on the paraffin-fixed sections using an automated Bond-max stainer (Leica Biosystems, Melbourne, Australia) and anti-cyclin D2 (polyclonal; 1:150; Proteintech Group Inc., Chicago, IL, USA) and anti-cyclin D3 (DCS-22; 1:10; Progen Biotechnik GmbH, Heidelberg, Germany) antibodies. Based on previous studies, a sample was considered to be positive if ≥20% of the tumor cells were stained[3]. Faint cytoplasmic staining for cyclin D2 without corresponding nuclear staining was not considered positive.
Statistical analysis
The correlations between the 2 groups were examined by a chi-square analysis. All statistical analyses were carried out with the SPSS software program (version 14.0, SPSS Inc., Chicago, USA).
Results and discussion
In this study, our data showed that cyclin D2 was overexpressed in 98% of de novo CD5+ DLBCLs (50/51) and in 28% of CD5− DLBCLs (14/51) (Table 1, Figure 1). A statistically significant difference was observed between these two groups (p<0.0001). In contrast, no statistical difference was found in the cyclin D3 expression between CD5-positive (18/51) and CD5-negative (24/51) DLBCLs (p=0.23) (Table 1, Figure 1). Since de novo CD5+ DLBCLs are typically immunohistochemically negative for cyclin D1[6], these findings therefore indicate that cyclin D2 is closely associated with de novo CD5+ DLBCLs.
Table 1
Cyclins D2 and D3 expression inde novoCD5+and CD5−DLBCLs
De novo CD5+DLBCLs
CD5−DLBCLs
P
Cyclin D2
50/51 (98%)
14/51 (28%)
P < 0.0001
Cyclin D3
18/51 (35%)
24/51 (47%)
P = 0.23
×
Previous studies have examined the cyclins D2 and D3 expression in DLBCLs by immunohistochemistry (Table 2)[3, 8‐10]. Although the rates of cyclins D2 and D3 positive cases were highly variable in these studies, our findings suggest that cyclin D2-positive DLBCLs comprise the majority of de novo CD5+ DLBCLs.
Table 2
Previous reports on cyclins D2 and D3 expression by immunohistochemistry in DLBCLs
Several studies have shown cyclin D2 to be an independent indicator of an inferior survival in DLBCLs[8‐10]. In addition, Lossos et al. identified CCND2 to be the best predictor of an inferior survival in DLBCLs among the 36 genes associated with their prognosis based on quantitative RT-PCR[11]. These data suggest that cyclin D2 may be associated with the inferior survival observed in de novo CD5+ DLBCLs. The current gold standard therapy has evolved to include rituximab, and the outcome of DLBCLs has been reported to have dramatically improved[12]. However, both CCND2 and CD5 have been demonstrated to still be a good indicator of the inferior survival in DLBCLs, even in the era of R-CHOP[7, 12]. This fact also suggests an association between cyclin D2 and the poor survival in de novo CD5+ DLBCLs. Similar to de novo CD5+ DLBCLs, cyclin D2-positive CD5− DLBCLs may play a role in the negative prognostic impact of cyclin D2 in DLBCLs. In the current study, however, whether or not cyclin D2-positive CD5−DLBCLs show a poor survival was not examined. As a result, the prognostic impact of this population therefore still needs to be investigated.
Conclusions
Cyclin D2 is closely associated with de novo CD5+ DLBCLs. This insight might be useful for making treatment strategies and improving the survival of this aggressive lymphoma, since cyclin D2 expression is controlled by multiple signaling pathways, such as the NFkB-related pathways[13].
Consent
Informed consent was obtained from the patient for publication of this report and any accompanying images.
Open Access
This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License (
https://creativecommons.org/licenses/by/2.0
), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Competing interests
The authors declare that they have no competing interest.
Authors’ contributions
Conceived and designed the experiments: YS, TI. Performed the experiments: TI, YS. Analyzed the data: YS, TI, TY, KT, NI, TT, YM, NA, YO. Contributed reagents/materials/analysis tools: NN, SN, YM. Wrote the paper: TI, YS, TY. All authors read and approved the final manuscript.