the Department of Pathology, Ghent University Hospital, Ghent
    Department of Morphology and Molecular Pathology, Department of Haematology, and the Center of Human Genetics, Catholic University of Leuven, Leuven, Belgium

    ABSTRACT

    PURPOSE: The reliability of immunohistochemistry for subdividing diffuse large B-cell lymphomas (DLBCL) into germinal center B-cell-like (GCB) and non-GCB prognostic subgroups is debated. In this study we evaluated the prognostic significance of such subgrouping on a series of 153 DLBCL patients. Furthermore, we investigated whether both subgroups could comprise clinicopathologic entities recognized by their morphology and characterized by a distinct phenotype, specific genetic abnormalities, and clinical characteristics.

    PATIENTS AND METHODS: All samples from patients were reviewed and morphologically subdivided into large cleaved, immunoblastic, and not otherwise specified DLBCL. GCB and non-GCB immunohistochemical profiles were established. The presence of chromosomal translocations involving BCL2, BCL6, and MYC and/or rearrangements of these genes was investigated.

    RESULTS: Subdividing DLBCL with either a GCB or non-GCB immunophenotypic profile was not of prognostic significance. Nevertheless, CD10 expression was a predictor of favorable outcome, whereas high bcl-2 expression and BCL6 rearrangement were adverse predictors of disease-free survival. Interestingly, large cleaved DLBCL was clearly associated with a GCB immunophenotypic profile, CD10 expression, BCL2 rearrangement, age younger than 60 years, and low to low/intermediate International Prognostic Index risk, but was not of prognostic significance. In contrast, immunoblastic morphology was associated with a non-GCB profile and was a significant predictor of unfavorable DFS.

    CONCLUSION: Subdividing DLBCL into subgroups based on their immunohistochemical profile was not of prognostic significance. Nevertheless, it allowed the additional characterization of two lymphoma subgroups previously recognized in the Working Formulation. Both correspond to two distinct clinicopathologic entities within the DLBCL.

    INTRODUCTION

    Diffuse large B-cell lymphomas (DLBCLs) account for approximately 40% of adult non-Hodgkin's lymphomas. They are clinically, morphologically, and genetically heterogeneous. Gene expression studies have shown that DLBCL can be subdivided into two main prognostically different groups based on the differential expression of genes characteristic of either normal germinal center B-cells or genes typical of activated blood B-cells. The germinal center B-cell-like (GCB) group is correlated with a significantly better prognosis than the activated B-cell-like (ABC) group. A third group, type 3, does not express genes characteristic of the GCB or ABC group and has a poor outcome similar to that of the ABC group. The gene expression profile of DLBCL is considered to be a prognostic factor independent of the International Prognostic Index (IPI), which is based on clinical parameters.1–5 Cytogenetic studies have supported the gene array conclusion that GCB and ABC groups are pathogenetically and biologically different tumors because BCL2 gene rearrangements were demonstrated almost exclusively in GCB tumors.4,6,7 Furthermore, patients with rearranged BCL2 genes share a similar gene expression profile, suggesting that they represent a unique subset within the group of GCB-like DLBCL.

    Recent studies have investigated the usefulness of immunohistochemistry in the identification of GCB and ABC (or non-GCB) subgroups.8–10 Most of these studies use CD10 and bcl-6 as GCB B-cell markers and MUM1/IRF4 and CD138 as activated or non-GCB B-cell markers. Several of these studies conclude that immunohistochemistry performed on formalin-fixed and paraffin-embedded tissue allows reliable identification of both subgroups using only three markers (CD10, bcl-6 and MUM-1) and prediction of survival similar to microarray analysis subgrouping.8,10 To date, morphology was not considered helpful for differentiation of the GCB group from the non-GCB group.

    In this study we have analyzed a series of 153 patients diagnosed as having DLBCL and collected within one center. Patient cases were subdivided into a GCB and a non-GCB subgroup based on the immunophenotypic profile of the neoplastic cells, and the prognostic significance of this subgrouping was evaluated. In addition, we investigated whether clinicopathologic entities with a characteristic morphology could be identified within both subgroups.

    This study was approved by the Institutional Ethics Commission of the Catholic University of Leuven (Leuven, Belgium). Informed consent was provided according to the declaration of Helsinki.

    PATIENTS AND METHODS

    Patients and Samples

    We retrospectively studied 153 patients with de novo DLBCL diagnosed at the University Hospital of Leuven between 1990 and 2003. Patients were selected based on availability of adequately fixed paraffin-embedded surgical biopsy material and fresh tissue for cytogenetic and/or molecular genetic studies. Burkitt or Burkitt-like lymphomas, T-cell/histiocyte-rich DLBCL, primary mediastinal lymphoma, intravascular lymphoma, primary effusion lymphoma, primary cutaneous, and primary CNS lymphomas were excluded from the study. The series included 28 primary extranodal (stage IE or IIE) and 125 primary nodal DLBCL. A complete clinical data set including the different clinical parameters of the IPI was available in 138 patients. The male-to-female ratio was 1.2:1 (76 men, 62 women). The mean age was 63 years (range, 20 to 89 years). The mean overall survival was 55 months (range, 0 to 174 months). Detailed information on treatment and follow-up was available for 134 patients. One hundred twenty-five patients were treated with curative intent. Eight patients presenting with localized early-stage disease were treated with radiotherapy only. Sixty-eight patients received combination chemotherapy and 49 patients were treated with combined chemotherapy and radiotherapy. Of these 117 patients, 113 received anthracycline-containing combination chemotherapy, predominantly cyclophosphamide, doxorubicin, vincristine, and prednisone or a similar regimen. The remaining patients (n = 9) received supportive therapy.

    Immunohistochemical Analysis

    The immunohistochemical panel included the monoclonal antibodies CD10, CD20, CD21, bcl-2, bcl-6, MUM1/IRF4, and Ki-67. Table 1 shows the characteristics of the antibodies used. CD10, bcl-6, and MUM1 stainings were scored positive or negative according to Hans et al,8 with cutoff levels of 30%. Ki-67 and bcl-2 were semiquantitatively scored negative (score 0), less than 25% (score 1), between 25% and 75% (score 2), and more than 75% of tumor cells (score 3).

    Subdivision into GCB and non-GCB subtypes was based on the algorithm of Hans et al.8 The GCB profile was defined by CD10 expression (regardless of expression of other markers) or by a CD10–/bcl-6+/MUM1– phenotype. Other patterns of expression (CD10–/bcl-6+/–/MUM1+ or lack of expression of the three markers) were considered to define the non-GCB profile.

    Genetic Studies

    Cytogenetic, fluorescent in situ hybridization (FISH), and Southern blot analyses followed standard protocols.11–13 Involvement of the BCL2, BCL6, and MYC genes was considered when respective translocations affecting 18q21, 3q27, and 8q24 were found or rearrangement of these genes was documented by FISH and/or Southern blot.

    Statistical Analysis

    The nine patients receiving supportive therapy were excluded from the survival analysis. Fisher's exact test was used to identify significant correlations between variables.

    The relationship between the immunohistochemical markers individually, the different immunohistochemical profiles, the morphologic subgroups, and the clinical variables of interest on the one hand, and overall survival (OS) and disease-free survival (DFS) on the other hand, were assessed using the product limit method of Kaplan-Meier. The P values for these analyses are based on the log-rank test. OS and DFS were measured from the time of diagnosis to death and to time of first relapse, respectively. The Cox proportional hazards model was used to assess which of those covariates retained significant prognostic value after adjusting for IPI risk group. All statistical analyses were performed using the SAS System, version 8.1 (SAS Institute Inc, Cary, NC).

    RESULTS

    Morphologic Analysis

    All patient cases were reviewed and a diagnosis of DLBCL according to the WHO classification was confirmed. The samples were subdivided in three groups based on the features of the neoplastic cells and associated stromal component.

    The first group (40 samples) was classified as large cleaved lymphoma because the diffuse proliferation was predominantly composed of large lymphoid cells, mostly spindle shaped. Their nuclei were indented or twisted and were characterized by fine nuclear chromatin, several small to medium sized nucleoli, and a well-delineated but delicate nuclear membrane; these features were best appreciated in B5-fixed sections. The presence of tiny fibrous septa results in a compartmentalization of the lymphoid proliferation, in most cases with an almost pericellular fibrosis. Within the septa, a variable amount of stromal T-cells was present (Fig 1A). Nodular or follicular areas suggestive of follicular lymphoma were absent. A CD21 staining to visualize follicular dendritic cells was performed on these 40 samples considered to be large cleaved lymphoma. Any evidence for the presence of a follicular dendritic network was absent in 35 of 40 samples. In the remaining five samples, a membranous CD21 expression was found throughout the tumor. As it was unclear whether CD21 was expressed by the neoplastic B-cell population or whether the diffuse staining pattern was indeed due to proliferating follicular dendritic cells, these samples were considered as DLBCL, not otherwise specified (NOS) category.

    The second group of samples (n = 8) was classified as immunoblastic lymphoma according to the WHO classification.14 These were characterized by a rather monotonous proliferation of almost exclusively (> 90%) immunoblasts, featuring a large round nucleus with centrally located prominent nucleolus (Fig 2A). In some samples, the tumor cells showed features of plasmacytic differentiation as illustrated by the abundant eccentrically located basophilic cytoplasm. In contrast to the large cleaved samples, immunoblastic lymphoma did not show a prominent fibrohistiocytic stromal reaction; a variable, occasionally high number of tingible body macrophages and stromal T cells could be found.

    All other samples (n = 110) showing a variety of morphologic features, including large cleaved lymphomas with CD21 expression, were grouped together as DLBCL, NOS category.

    Bone Marrow Involvement

    Bone marrow biopsy involvement at the time of diagnosis was present in 10 of 120 patients (8.3%). In eight patients, histology showed infiltration of the marrow by large atypical B cells. In only two patients the trephine showed paratrabecular infiltrates predominantly composed of small lymphoid cells. In both latter patients, the diagnostic sample showed a mixture of centroblasts and immunoblasts; both samples were classified in the DLBCL, NOS category.

    Immunohistochemical Results

    Reliable evaluation of the different immunohistochemical markers was possible in all patients except for MUM1 in one patient and CD10 in two patients due to background staining. Therefore, the algorithm of Hans et al8 could be applied on 150 patients. Expression of CD10 was seen in 40%, bcl-6 in was seen 70.5%, MUM1 was seen in 58.5%, strong bcl-2 expression (score 2 or 3) was seen in 77%, and a high proliferation index (score 2 or 3) was observed in 75.5% of patients.

    Using the algorithm of Hans et al,8 49% (74 of 150) of the patients showed a GCB profile (CD10+ or CD10–/BCL-6+/MUM1–) versus 51% (76 of 150) showing a non-GCB profile. Within the GCB group, 25% of patients expressed MUM1. There was no significant difference in the bcl-2 score between the GCB and the non-GCB group.

    Molecular Genetic Aberrations

    Table 2 lists an overview of BCL2, BCL6, and MYC rearrangements in the total group of DLBCLs and the different subgroups.

    BCL2 rearrangements were detected in 17% of patients (22 of 128 patients with successful genetic analysis), BCL6 rearrangements were detected in 24% (25 of 103 successful genetic analyses), and MYC rearrangements were detected in 6.8% of patients (seven of 102 successful analyses).

    BCL2 gene rearrangement was correlated with CD10 expression (P < .0001), a GCB profile (P < .0001), and cleaved morphology (P = .02). In fact, nearly all (21 of 22 patients) of the BCL2-rearranged patients displayed a GCB immunophenotypic profile. In one third of DLBCL with a GCB immunophenotypic profile, a BCL2 gene rearrangement was detected. Interestingly, 50% (11 of 22 patients) of BCL2-rearranged patients were characterized by cleaved morphology.

    In 85% (17 of 20) of BCL2-rearranged cases, high bcl-2 protein expression was found. However, this observation did not reach statistical significance (P = .34; Table 3).

    Of the 25 patients with BCL6 rearrangement, 18 were detected by conventional cytogenetics (karyotyping) and further proved by FISH in nine patients. Translocation t(3;14)(q27;q32) resulting in the BCL6/IGH fusion was found in seven patients, and three-way translocations [t(3;14;18)(q27;q32;q21) and t(3;8;14)(q27;q24;q32)] were identified in one patient each. In the nine remaining patients, variant 3q27 translocations, possibly involving a non-IGH partner gene, were found.

    Main Clinical, Phenotypic, and Molecular Genetic Characteristics of the Full Series of DLBCL in Comparison With Those of the Morphologic Subgroups

    The 153 DLBCLs were classified as large cleaved (n = 35, 23%), immunoblastic (n = 8, 5.2%), and NOS (n = 110, 72%) lymphomas (Table 4). The group of large cleaved lymphomas included 13 men and 22 women (male-to-female ratio of 0.6:1). The mean age of these patients was 60 years (range, 20 to 79 years); almost half of the patients were younger than 60 years. Statistically significant correlations were observed between cleaved morphology and GCB profile (P < .0001), CD10 expression (P = .003), lack of MUM1 expression (P < .0001), BCL2 rearrangement (P < .0001), and age-adjusted low/intermediate-risk IPI (P = .003). The immunoblastic subgroup comprised eight patients, five male and three female (male-to-female ratio of 1.6:1). The mean age of these patients was 70.1 years (range, 42 to 84 years). In seven of the eight patients, the neoplastic cells were characterized by a non-GCB phenotype. Most of the patients presented with intermediate/high or high IPI at diagnosis.

    Survival Analysis and Response to Treatment

    The 5-year OS was 61.78%; the 5-year DFS was 48.29%. By univariate analysis, the following variables had a significant adverse impact on DFS and OS: high bcl-2 expression (P = .002 and P = .06, respectively), Ann Arbor stage III/IV (P = .003 and P = .001, respectively), bone marrow invasion (P < .0001 for both DFS and OS), Eastern Cooperative Oncology Group performance status more than 2 (P = .0002 for DFS and P < .0001 for OS), high lactate dehydrogenase level (P = .008 and P = .003, respectively), and intermediate/high and high IPI (P = .0016 and P < .0001, respectively). CD10 expression (P = .07, borderline) was a predictor of favorable DFS, whereas immunoblastic morphology (P = .001) and presence of BCL6 rearrangement (P = .05) were predictors of an unfavorable DFS.

    The prognostic significance of high bcl-2 protein expression was also investigated in the GCB and non-GCB subgroups (Table 5). High bcl-2 expression seemed to be a prognostic indicator for worse DFS and OS, especially in the GCB subgroup (P = .01 and P = .09, respectively). Finally, using a Cox proportional hazards analysis, high bcl-2 expression was a predictor for DFS and OS (P = .003 and P = .07), whereas immunoblastic morphology and CD10 expression were predictors for DFS only.

    The overall complete remission (CR) rate was 61%. The following variables predicted CR achievement: female sex (P = .04), early Ann Arbor stage (P = .001), absence of bone marrow invasion (P = .04), performance status less than 2 (P = .0002), normal lactate dehydrogenase level (P = .001), low- to low/intermediate-risk IPI (P = .0002), CD10 expression (P = .02), and absence of BCL6 rearrangement (P = .03). Statistical data are summarized in Table 6.

    Although the group with immunoblastic lymphoma showed a worse OS (median OS, 41 months) in comparison with the total group of DLBCLs and the large cleaved group (median OS, 115 months in both groups), this observation did not reach statistical significance. Cleaved morphology, conversely, showed a higher CR rate and a longer DFS in comparison with the total group of DLBCLs. Again, these findings were not statistically significant.

    DFS curves of the large cleaved (Fig 1B) and immunoblastic lymphomas (Fig 2B) as well as DFS and OS curves of the GCB versus non-GCB immunophenotypic groups are shown in Figs 3A and 3B.

    DISCUSSION

    Classification of DLBCL was, until the 1990s, largely based on morphology (the Kiel classification that subdivided large B-cell lymphomas into centroblastic and immunoblastic lymphomas) or on their biologic behavior (Working Formulation that classified large B-cell lymphomas into an intermediate-grade and a high-grade category). Large B-cell lymphomas were lumped under the category DLBCLs in the Revised European-American Lymphoma classification. Several clinicopathologic entities with characteristic morphologic, immunophenotypic, cytogenetic, and clinical features have since been recognized within the DLBCLs and were classified separately in the WHO classification. These lymphomas (mediastinal large B-cell lymphoma, intravascular large B-cell lymphoma, and primary effusion lymphoma) as well as T-cell/histiocyte rich DLBCL were not included in the present study. In addition to these entities, several variants of DLBCL (immunoblastic, centroblastic, anaplastic, and T-cell/histiocyte rich) are listed in the WHO classification, but none of them are accepted currently as representing distinct lymphoma entities.

    Microarray studies provided new insights into the biology of DLBCL by identifying at least two major groups of DLBCLs based on different gene expression profiles, suggesting different underlying pathogenetic mechanisms.3–5 These GCB-like and ABD-like (or non-GCB-like) molecularly defined groups carry a different prognostic significance independent of the IPI. To date, no clear correlation has been found between the gene profiles and the different morphologic variants listed in the WHO classification.4 Cytogenetic studies showed differences in occurrence of BCL2 gene rearrangement within the different molecular groups. The t(14;18)(q32;q21) translocation, which can be detected in 12% to 30% of all DLBCLs,15–19 occurs almost exclusively in the GCB subgroup.4,7,20 Furthermore, these BCL2 gene-rearranged patients were shown to share a similar gene expression profile, suggesting that they represent a unique subset within the group of GCB-DLBCLs.20 However, no significant differences in OS and DFS were found between the translocation-positive and -negative GCB patients.7

    Several studies have applied immunohistochemistry to identify the molecularly defined subgroups by their immunophenotypic profile.6,8,10,21 Although all investigators use a similar panel of antibodies consisting of CD10 and bcl-6 as GCB markers and MUM1 and CD138 as activation markers, they differ in defining the GCB and non-GCB subgroups, which complicates comparison of the different studies. Nevertheless, the general conclusion is that the immunophenotypic profile of the neoplastic B cells based on three markers only (CD10, bcl6, and MUM1) allows identification of the molecularly defined subgroups and may predict the final outcome of the patient.8,10

    In our study, the algorithm of Hans et al8 was used to subdivide DLBCLs into GCB and non-GCB lymphomas. This subdivision did not give significant differences in terms of OS or DFS. Our results are in line with data published by Colomo et al,21 who studied 128 patients with de novo DLBCL. Comparable to our findings, the different immunophenotypic profiles were not predictive for outcome or response to treatment.21

    Significant predictors of DFS and achievement of CR in our study were BCL6 gene rearrangement and CD10 expression. The prognostic role of BCL6 gene rearrangement is controversial. BCL6 rearrangement has been associated with a more favorable outcome,22,23 a worse outcome,24 or not correlated with outcome.25 Most BCL6 rearrangements involve a translocation between BCL6 and the immunoglobulin heavy chain (IGH) gene, but other partner genes have been described as well.13 Fusion between BCL6 and a nonimmunoglobulin gene resulted in a worse prognosis when compared with BCL6/IGH fusions.26 This observation offers a possible explanation for the adverse prognostic effect of BCL6 rearrangement on DFS in our study, given that half of the BCL6 rearrangements found most likely involved a non-IGH partner. CD10 is generally considered a marker for follicle center cell origin and is thus used as an indicator of GCB-like DLBCL. Confirming several reports,9,20,27 we found CD10 expression a favorable prognostic factor for DFS and achievement of CR. However, in contrast to the findings of Hans et al,8 bcl-6 and MUM1 expression had no impact on survival or response to treatment in our study, and this may explain why we could not confirm the predictive power of the immunohistochemical profiles as defined by Hans et al. Differences in results between our study and that of Hans et al can be explained further by the study design (a single-center study v a multicenter study), the application of antibodies to whole tissue sections instead of tissue microarray slides, and the use of a monoclonal bcl-6 antibody (clone PG-B6p), whereas Hans et al used a polyclonal bcl-6 antibody. More studies with possible adjustment of the algorithms or redefining of the cutoff levels of the different antibodies are needed to validate the predictive value of the immunohistochemical approach in the DLBCL.

    In agreement with several other studies,15,16,18,28 BCL2 gene rearrangements that were found in 17% of patients in our study had no impact on prognosis. Furthermore, there was no statistically significant correlation between high bcl-2 expression and BCL2 gene rearrangement; high bcl-2 protein expression was found to occur independently of the t(14;18) translocation and/or BCL2 rearrangement. High bcl-2 expression might result from amplification of the BCL2 gene, a phenomenon that was reported in up to 30% of DLBCL.29 The prognostic value of bcl-2 protein expression in DLBCL has been studied extensively and our data are in line with previously reported findings demonstrating that a high bcl-2 expression represents a strong unfavorable predictor of OS and DFS.15,16 However, when the full cohort of patients into four small subgroups was subdivided for the evaluation of the prognostic significance of high bcl-2 expression within the GCB and non-GCB subgroups, the prognostic significance of high bcl-2 expression in the non-GCB group was lost (Table 5).

    Interestingly, by adding morphological characteristics to the immunophenotypic profiles we were able to identify two lymphoma entities within the large group of DLBCLs. Large cleaved DLBCL showed a slight female predominance and presented at a younger age in comparison with the other DLBCLs. A significant correlation was found among large cleaved morphology, germinal center-like profile, CD10 expression, and BCL2 gene rearrangement, indicating that large cleaved DLBCL is a distinct entity within the GCB group associated with a recurrent chromosomal aberration. We assume that large cleaved DLBCL corresponds to the molecular subset recognized within the GCB group (detected by gene array studies) associated with BCL2 rearrangement.20 Similar to this molecularly recognized subgroup, large cleaved DLBCL in our analysis did not differ in terms of DFS and OS from the other DLBCL.

    Conversely, immunoblastic morphology was associated with a non-GCB profile and was characterized by consistent MUM1 expression and infrequent bcl-6 expression. Patients with immunoblastic DLBCL were mostly men presenting with primary nodal disease at an advanced stage. Immunoblastic morphology was a significant adverse predictor of DFS in univariate and multivariate analyses. These findings confirm the results of a previous study in which immunoblastic morphology, defined as a monotonous proliferation of more than 90% immunoblasts, was found to represent an independent prognostic factor for OS and relapse-free survival.30 Others found that when comparing patients with DLBCL with and without immunoblasts, those with immunoblastic differentiation had a significantly worse OS and DFS.31

    These findings suggest the presence of distinct clinicopathologic entities within GCB and non-GCB DLBCL groups. The association between a CD10+ GCB profile and centroblastic morphology (including large cleaved cell morphology) on the one hand and a non-GCB profile and immunoblastic morphology on the other hand was also reported by Colomo et al.21 In their study, these morphologic subtypes had no significant prognostic value.

    In summary, we could not confirm the prognostic significance of subdividing DLBCLs based on the GCB and non-GCB immunophenotypic profile of the neoplastic cells. Nevertheless and more importantly, our study allowed us to identify and define, adapting the principles of the Revised European-American Lymphoma and WHO classification, two new, distinct clinicopathologic entities within the group of DLBCLs. Large cleaved DLBCL, represented by almost one fourth of patients, and immunoblastic DLBCL, represented by more than 5% of patients included in this study, are both clinicopathologic entities characterized by a distinct morphology, immunophenotype, clinical features, and cytogenetic abnormalities. Interestingly, both were recognized previously in the Working Formulation and Kiel-classification.

    Authors' Disclosures of Potential Conflicts of Interest

    The authors indicated no potential conflicts of interest.

    Acknowledgment

    We thank J. Meuleman, G. Bullock, J. Van de Broeck, and especially M. Van Herck for their excellent technical assistance.

    NOTES

    Supported by Grant No. G.0362.01 from the Fund for Scientific Research (FWO) Flanders; G.V. is a research fellow of the Belgian Federation against Cancer; and G.V. is holder of the Roche Chair in Hematology.

    Authors' disclosures of potential conflicts of interest are found at the end of this article.

    REFERENCES

    A predictive model for aggressive non-Hodgkin's lymphoma: The International Non-Hodgkin's Lymphoma Prognostic Factors Project. N Engl J Med 329:987-994, 1993

    Atkins CD: A predictive model for non-Hodgkin's lymphoma. N Engl J Med 330:574-575, 1994

    Alizadeh AA, Eisen MB, Davis RE, et al: Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403:503-511, 2000

    Rosenwald A, Wright G, Chan WC, et al: The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 346:1937-1947, 2002

    Shipp MA, Ross KN, Tamayo P, et al: Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nat Med 8:68-74, 2002

    Barrans SL, Evans PAS, O'Connor SJM, et al: The t(14;18) is associated with germinal center-derived diffuse large B-cell lymphoma and is a strong predictor of outcome. Clin Cancer Res 9:2133-2139, 2003

    Iqbal J, Sanger WG, Horsman DE, et al: BCL2 translocation defines a unique tumor subset within the germinal center B-cell-like diffuse large B-cell lymphoma. Am J Pathol 165:159-166, 2004

    Hans CP, Weisenburger DD, Greiner TC, et al: Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood 103:275-282, 2004

    Barrans SL, Carter I, Owen RG, et al: Germinal center phenotype and bcl-2 expression combined with the International Prognostic Index improves patient risk stratification in diffuse large B-cell lymphoma. Blood 99:1136-1143, 2002

    Chang CC, McClintock S, Cleveland RP, et al: Immunohistochemical expression patterns of germinal center and activation B-cell markers correlate with prognosis in diffuse large B-cell lymphoma. Am J Surg Pathol 28:464-470, 2004

    Martin-Subero JI, Harder L, Gesk S, et al: Interphase fish assays for the detection of translocations with breakpoints in immunoglobulin light chain loci. Int J Cancer 98:470-474, 2002

    Michaux L, Wlodarska I, Theate I, et al: Coexistence of BCL1/CCND1 and CMYC aberrations in blastoid mantle cell lymphoma: A rare finding associated with very poor outcome. Ann Hematol 83:578-583, 2004

    Wlodarska I, Mecucci C, Stul M, et al: Fluorescence in situ hybridization identifies new chromosomal changes involving 3q27 in non-Hodgkin's lymphomas with BCL6/LAZ3 rearrangement. Genes Chromosomes Cancer 14:1-7, 1995

    Gatter KC, Warnke RA: Diffuse large B-cell lymphoma, in Jaffe ES, Harris NL, Stein H, Vardiman JW (eds): World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues, Lyon, France, 2001, pp 171-174

    Gascoyne RD, Adomat SA, Krajewski S, et al: Prognostic significance of Bcl-2 protein expression and Bcl-2 gene rearrangement in diffuse aggressive non-Hodgkin's lymphoma. Blood 90:244-251, 1997

    Hill ME, MacLennan KA, Cunningham DC, et al: Prognostic significance of BCL-2 expression and bcl-2 major breakpoint region rearrangement in diffuse large cell non-Hodgkin's lymphoma: A British National Lymphoma Investigation Study. Blood 88:1046-1051, 1996

    Jacobson JO, Wilkes BM, Kwaiatkowski DJ, et al: Bcl-2 rearrangements in de novo diffuse large cell lymphoma: Association with distinctive clinical features. Cancer 72:231-236, 1993

    Kramer MHH, Hermans J, Wijburg E, et al: Clinical relevance of BCL2, BCL6, and MYC rearrangements in diffuse large B-cell lymphoma. Blood 92:3152-3162, 1998

    Yunis JJ, Mayer MG, Arnesen MA, et al: Bcl-2 and other genomic alterations in the prognosis of large-cell lymphoma. N Engl J Med 320:1047-1054, 1989

    Huang JZ, Sanger WG, Greiner TC, et al: The t(14;18) defines a unique subset of diffuse large B-cell lymphoma with a germinal center B-cell gene expression profile. Blood 99:2285-2290, 2002

    Colomo L, Lopez-Guillermo A, Perales M, et al: Clinical impact of the differentiation profile assessed by immunophenotyping in patients with diffuse large B-cell lymphoma. Blood 101:78-84, 2003

    Offit K, Lo CF, Louie DC, et al: Rearrangement of the bcl-6 gene as a prognostic marker in diffuse large-cell lymphoma. N Engl J Med 331:74-80, 1994

    Ohno H, Fukuhara S: Significance of rearrangement of the BCL6 gene in B-cell lymphoid neoplasms. Leuk Lymphoma 27:53-63, 1997

    Barrans SL, O'Connor SJ, Evans PA, et al: Rearrangement of the BCL6 locus at 3q27 is an independent poor prognostic factor in nodal diffuse large B-cell lymphoma. Br J Haematol 117:322-332, 2002

    Pescarmona E, DeSanctis V, Pistilli A, et al: Pathogenetic and clinical implications of Bcl-6 and Bcl-2 gene configuration in nodal diffuse large B-cell lymphomas. J Pathol 183:281-286, 1997

    Akasaka T, Ueda C, Kurata M, et al: Nonimmunoglobulin (non-Ig)/BCL6 gene fusion in diffuse large B-cell lymphoma results in worse prognosis than Ig/BCL6. Blood 96:2907-2909, 2000

    Ohshima K, Kawasaki C, Muta H, et al: CD10 and Bcl10 expression in diffuse large B-cell lymphoma: CD10 is a marker of improved prognosis. Histopathology 39:156-162, 2001

    Tang SC, Visser L, Hepperle B, et al: Clinical significance of bcl-2-MBR gene rearrangement and protein expression in diffuse large-cell non-Hodgkin's lymphoma: An analysis of 83 cases. J Clin Oncol 12:149-154, 1994

    Monni O, Joensuu H, Franssila K, et al: BCL2 overexpression associated with chromosomal amplification in diffuse large B-cell lymphoma. Blood 90:1168-1174, 1997

    Engelhard M, Brittinger G, Huhn D, et al: Subclassification of diffuse large B-cell lymphomas according to the Kiel classification: Distinction of centroblastic and immunoblastic lymphomas is a significant prognostic risk factor. Blood 89:2291-2297, 1997

    Diebold J, Anderson JR, Armitage JO, et al: Diffuse large B-cell lymphoma: A clinicopathologic analysis of 444 cases classified according to the updated Kiel classification. Leuk Lymphoma 43:97-104, 2002

查询更多Lymphoma相关信息在本站>>

《临床肿瘤学医学期刊》2005年10月第23卷第10期