Egyptian Journal of Pathology:
The role of immunohistochemistry in subclassification of diffuse large B-cell lymphoma with a vague nodular pattern
El-Tanaihy, Eman Y.a; El-Eisawy, Basem H.a; Ali, Mohamed Y.b
aPathology Department, Faculty of Medicine, Mansoura University, Mansoura
bPathology Department, Faculty of Medicine, Al Azhar University, Cairo, Egypt
Correspondence to Eman Y. El-Tanaihy, MD, BCH, Pathology Department, Faculty of Medicine, Mansoura University, 35116 Mansoura, Egypt Tel: +20 122 466 0879; fax: +2050 2248203; e-mail: firstname.lastname@example.org
Received February 5, 2013
Accepted February 20, 2013
Background: The aim of the current study was to elucidate the histology, immunophenotype, and differential diagnosis of diffuse large B-cell lymphoma (DLBCL) with a vague nodular pattern. The heterogeneity of DLBCL is widely acknowledged. T-cell/histiocyte-rich B-cell lymphoma (T/HRBCL) is an uncommon aggressive variant of DLBCL. It has both nodal and extranodal presentation. The diagnosis of T/HRBCL is difficult as it may share several morphological and immunophenotypic similarities to other lymphoid diseases, such as nodular lymphocytic predominance Hodgkin’s lymphoma (NLPHL) and classic Hodgkin’s lymphoma. Thus, histologic subclassification of DLBCL is useful to avoid the risk of treatment failure. Accurate diagnosis therefore requires a careful immunohistochemical analysis.
Patients and methods: A total of 332 cases diagnosed with DLBCL with vague nodularity were reviewed between 2009 and 2012. They were reclassified as DLBCL, T/HRBCL, or NLPHL according to the 2008 WHO classification of lymphoid neoplasms. Immunohistochemical studies were carried out on paraffin-embedded tissue to study the immunophenotype of the large neoplatic cells (CD20, CD15, CD30, EMA, Bcl-2) and the background of non-neoplastic cells (CD3, CD20).
Results: In total, 332 cases diagnosed with DLBCL with vague nodularity were reviewed. Of which 318 (95.7%) cases were diagnosed as DLBCL, whereas 12 (3.6%) were reclassified as T/HRBCL and two (0.6%) were reclassified as NLPHL. There was male predominance (75.0%) among cases of T/HRBCL. Most patients with T/HRBCL were diagnosed at a younger age in comparison with patients with DLBCL, with a mean age of 46.50±14.2 years. Cervical lymph nodes were the most frequently involved lymph node type in both DLBCL (56.6%) or in T/HRBCL (50.0%).
In terms of disease stage, there was a statistically significant difference (P=0.031) between T/HRBCL and DLBCL patients. Most patients [eight (66.7%)] with T/HRBCL presented at an advanced stage. All patients with DLBCL were CD3− and showed a diffuse positive reaction for CD20. However, patients with T/HRBCL showed a positive reaction for CD3 in 80–90% of the reactive cells and showed a positive reaction for CD20 in 10–20% of the large neoplastic cells. In NLPHL cases, lymphocytic and/or histiocytic cells showed positivity for EMA and CD20 and were surrounded by reactive T cells positive for CD3. There was a statistically significant difference between patients with T/HRBCL and DLBCL in the mean Bcl-2 expression (16.3±7.1 and 43.2±18.2, respectively).
Conclusion: T/HRBCL is an uncommon clinicopathologic subtype of DLBCL. Careful attention must be paid at the time of diagnosis because it may be confused with other lymphoid neoplasms characterized by a marked host immune reaction such as Hodgkin’s disease (NLPHL). Immunophenotyping analysis is essential in establishing the diagnosis of T/HRBCL in all cases.
Diffuse large B-cell lymphomas (DLBCLs) is a type of aggressive non-Hodgkin’s lymphoma (NHL), which accounts for ∼40% of lymphoma (Jennifer et al., 2011). DLBCL is the most common histologic subtype of NHL, accounting for ∼25% of adult NHL cases (Morton et al., 2006). The molecular pathogenesis of DLBCL is a complex multistep process leading to the replication of malignant clone of germinal or postgerminal B-cell origin. The diagnostic category of DLBCL is heterogenous in terms of morphology, genetics, and biologic behavior. There is a B lineage lymphoma that shows overlapping clinical, morphological, and/or immunophenotypic characteristics, and is now recognized as a separate diagnostic category known as gray-zone lymphoma (Eberle and Jaffe, 2011). Some authors acknowledge the existence of some gray-zone cases showing features common to both T-cell/histiocyte-rich B-cell lymphoma (T/HRBCL) and NLPHL because lymphoma cells in T/HRBCL morphologically and phenotypically resemble Hodgkin’s cells, and some nodular growth patterns have been described in T/HRBCL (Anagnostopoulos et al., 2000; Achten et al., 2002). This biologic overlap between both diseases as well as the observation of their sequential or simultaneous occurrence in the same patient has led to the hypothesis that H/T-cell rich B cell lymphoma (TCRBCL) can result from the transformation of nodular lymphocytic predominance Hodjken disease (Müller-Hermelink and Rüdiger, 2003). TCRBCL and nodular lymphocytic predominance Hodgkin’s lymphoma (NLPHL) represent a major challenge to both the pathologist and the clinician (Dogan, 2005).
In 2008, the WHO defined TCRBCL as a morphological variant of DLBCL characterized by fewer than 10% large neoplastic cells amid a prominent inflammatory infiltrate, the majority of which are small polyclonal T cells. The percentage of T cells required to classify lymphoma as TCRBCL is somewhat controversial but most authors would consider tumors as TCRBCL if at least 65–90% of the lymphoid infiltrate constitutes T cells, with the remainder being large B cells. The large cells showed the CD20+, CD15−, and CD30− phenotype. However, the background small lymphocytes were CD3+ (Boudova et al., 2003). Factor(s) responsible for the reduced number of B-cell and increased T-cell infiltrate in TCRBCL have not been well characterized. However, tumor cell apoptosis (perhaps cytotoxic T-cell mediated) may partly account for the decreased number of large (neoplastic) B cells, but other factors (i.e. decreased Bcl-2 expression) may also be needed (Felgar, 1998) Clinically, TCRBCL is an aggressive B-cell NHL and should be treated as high-grade large-cell lymphoma. It occurs in younger patients, predominantly affecting men, and the patient presents at an advanced stage involving the liver, spleen, and bone marrow, with greater frequency than traditional DLBCL, accompanied by B symptom (Sehn et al., 2005). NLPHL is a distinct type of Hodgkin’s lymphoma (HL), representing ∼5% of all HL cases. Clinical, morphologic, immunophenotypic, and molecular features set NLPHL apart from other types of HL. NLPHL is characterized by expansile nodules composed of sparse, relatively large neoplastic cells, referred to as lymphocytic and/or histiocytic (L&H) cells intermixed with numerous reactive lymphocytes and a variable number of histiocytes (Drakos et al., 2009). L&H cells are CD20+ cells that have an immunoglobulin gene rearrangement with somatic hypermutations and, therefore, are considered to be derived from follicular center B cells. NLPHL is the most favorable prognostic histologic subtype of HL (Pijuan et al., 2005).
The aim of this study was to elucidate the role of immunophenotypical studies in the evaluation of accurate diagnosis of lymphoma with a vague nodular pattern and excess reactive lymphocytes, as these require different treatment protocols and carry different prognosis, and to determine whether the increased apoptosis or reduced cellular proliferation is the possible mechanism that might account for the lower number of neoplastic large B cells in TCRBCL in comparison with DLBCL.
Patients and methods
Tissue material that consisted of 332 specimens of patients diagnosed with DLBCL with a vague nodular pattern and rich in small lymphocytes was selected and collected from the archives of the Pathology Department, Mansoura University, Egypt, and the Pathology Department, Al Azhar University, Cairo, Egypt, in the period from 2009 to 2012.
Information on age, sex, type of lymph nodes involved, histological diagnosis, laboratory studies including complete blood counts and lactate dehydrogenase, pathological staging including bone marrow aspirates or biopsies, and appropriate radiologic studies (including at least a chest radiograph, abdominal ultrasound, computed tomography scan, or MRI) was retrieved from the files of the patients.
Tissue material in all cases was an excisional lymph node biopsy. All samples were fixed in 10% formalin, routinely processed, and paraffin embedded. Serial 5-µm sections were cut and stained with hematoxylin and eosin and examined independently by three observers (Eman, Basem, and Mohamed) to verify the histological diagnoses as DLBCL with vague nodularity. The cases collected were classified according to the WHO (2008) criteria.
Unstained sections from the paraffin blocks were subjected to immunohistochemistry. Immunolabeling was performed on tissue sections using an avidin–biotin–peroxidase system (Invitrogen, California, USA) as described by Wood and Warnke (1981) with diaminobenzedene was used as a chromogen and Mayer’s hematoxylin as a light counterstain. The standard immunophenotyping panel included antibodies to the following CD antigens: CD20, CD3, CD15, and CD30. Also, EMA and Bcl-2 was performed. Microwave heating in a solution of sodium citrate (pH 6) was performed before incubation with the antibodies. For negative control stains, the primary antibodies were omitted and replaced by PBS. Control tissues and internal controls were stained appropriately.
Interpretation of the staining
CD3, CD20, CD15, CD30, and Bcl-2 reactivity was assessed in all cases. For the purpose of data analysis, immunoexpression was evaluated according to the proportion of the positive cells. Antibody staining was evaluated independently and jointly by the pathologists (Eman, Basem, and Mohamed). Staining of the small lymphocytes was scored as follows: 0, no staining; 1+, 1–25% small lymphocytes positive; 2+, 26–50% small lymphocytes positive; 3+, 51–75% small lymphocytes positive; and 4+, 76–100% small lymphocytes positive. Staining of the large (neoplastic) lymphocytes was also evaluated using the following scale: 0, no staining; 1+, 1–10% large cells positive; 2+, 11–25% large cells positive; 3+, 26–50% large cells positive; and 4+, more than 50% large cells positive (Felgar et al., 1998).
The Statistical Package for Social Sciences (SPSS, version 16, IBM, Chicago, Illinois, USA) was used for statistical analysis. Comparison between numbers and percentages was performed using the χ2-test. A P value of less than 0.05 was considered statistically significant. The t-test was used to compare the mean and SD between groups.
There was a statistically significant difference (P=0.03) between TCRBCL and DLBCL patients in terms of sex. Among 318 patients still diagnosed with DLBCL, there were 136 men (42.8%) and 182 women (57.2%) (male-to-female ratio=0.7). However, among patients diagnosed with TCRBCL, there were nine men (75.0%) and three women (25.0%) (male-to-female ratio=3). Most patients with TCRBCL were diagnosed at a younger age in comparison with patients with DLBCL (with a statistically nonsignificant difference, P=0.13), as seven (58.4%) patients with TCRBCL were diagnosed in the age range of 20–40 years, whereas 181 (56.9%) of patients with DLBCL were diagnosed in the age range of 40–60 years, with a mean of 47.8±12.9 years. In terms of disease stage at the time of diagnosis, there was a statistically significant difference (P=0.031) between patients with TCRBCL and DLBCL. Eight patients with TCRBCL (66.7%) presented at stage III, whereas 116 patients with DLBCL (36.5%) presented at stage I. In our study, organomegaly showed a statistically significant difference (P=0.00) between TCRBCL and DLBCL patients. Five patients (41.7%) with TCRBCL presented at the time of diagnosis with spleen involvement, whereas only 67 patients (21.1%) with DLBCL presented with spleen involvement. Also, there was a statistically significant difference (P=0.006) between the two groups in bone marrow infiltration at the time of diagnosis. Nine patients with TCRBCL (75%) presented with bone marrow infiltration, whereas 115 patients with DLBCL (36.2%) presented with bone marrow infiltration. No central nervous system involvement was detected in any patient in both groups. There was no statistically significant difference between TCRBCL and DLBCL patients in the type of lymph nodes involved. Involvement of the cervical lymph nodes was most frequently found in both groups. Evaluation of lactic dehydrogenase levels showed a nonstatistically significant difference in the mean level between both TCRBCL and DLBCL patients (374.9±122.9 and 392.7±138.9, respectively) (Table 1).
On low-power examination, the nodal architecture is usually completely effaced with a diffuse pattern of tumor infiltration, although vague nodularity may be apparent (Fig. 1), with few scattered large neoplastic cells in a background rich in histiocytes and small lymphocytes (Fig. 2).
Immunohistochemical staining of CD15 and CD30 was negative in all patients with DLBCL, TCRBCL, and NLPHL. Immunohistochemical staining of CD20 was positive in 100% of neoplastic cells in all patients with DLBCL, whereas positivity for CD20 antibody was observed in all patients with TCRBCL but was restricted to large neoplastic cells, which comprise 10–20% of cells (Fig. 3). Immunohistochemical staining of the CD3 antibody was negative in all patients with DLBCL and positive in 80–90% of the small reactive lymphocytes in patients with TCRBCL (Fig. 4). In NLPHL, popcorn cells showed a positive reaction for EMA (Fig. 5) and the CD20 antibody.
There was a statistically significant difference between patients with TCRBCL and patients with DLBCL in the mean Bcl-2 expression (16.3±7.1 and 43.2±18.2, respectively). Patients with TCRBCL had a lower percentage of tumor cells expressing Bcl-2 (12 of 12 immunoreactive cases had <25% Bcl-2-positive tumor cells) (Fig. 6).
TCRBCL is a recent and uncommon described morphological variant of B-cell NHL, representing 1–3% of all DLBCL (Sehn et al., 2005). The term gray-zone lymphoma is mostly used for cases of malignant lymphoma in which there are morphological, biological, and clinical features suggesting an overlap between HL and NHL (Achten et al., 2002). As this differential diagnosis has direct implications for management strategies, and despite its inclusion in the recent WHO classification, it represents a major problem area for both pathologists and hemato-oncologists (Poppema et al., 2005). This difficulty is faced in NLPHL, T/HRBCL, and DLBCL with a vague nodular pattern. They are distinct tumors and are treated differently. They are linked by a morphologic and probably a biologic continuum, which renders the differential diagnosis difficult (Müller-Hermelink and Rüdiger, 2003). Thus, immunohistochemistry and in some cases molecular studies are essential to establish an accurate diagnosis of neoplastic cells in gray-zone lymphoma (Eberle and Jaffe, 2011).
TCRBCL was originally differentiated as a unique subset of DLBCLs because of distinct histologic features. Since then, however, clinical distinctions have likewise become apparent (Olaniyi et al., 2011). In our study, TCRBCL has been described in 12 patients of younger age than DLBCL, with a mean age of 12.87±47.50 years. This is similar to the results of studies carried out by Bouabdallah et al. (2003)and Greer et al. (1995), who reported no diagnosis in patients younger than 18 years of age. The 12 cases reported in our study included both men (nine) and women (three). This result is similar to that of a study carried out by Aki et al. (2004), who reported male predominance. Conversely, Lones et al. (2000) reported all cases in men. However, this in contrast to DLBCL, which showed female predominance. Most patients with TCRBCL in our study presented with advanced-stage disease as eight (66.7%) patients with TCRBCL presented at stage III, and extranodal nodal involvement often involved spleenomegaly. Bone marrow involvement in TCRBCL was found in nine patients (75%). This is in agreement with the results of Aki et al. (2004) and Fraga et al. (2002), who reported the propensity of TCRBCL to present with advanced stage, with multiple extranodal sites. TCRBCL is characterized by the presence of less than 10% (sometime as low as 1%) of usually large, atypical B cells in a background rich in T cells (Abramson, 2006). In this study, the ratio of reactive T cells and neoplastic B cells was somewhat different. The T-cell ratio in our cases was 80–90%, confirmed by positivity for the CD3 antibody. This result almost similar to that obtained by Farhi (1995), who reported that the percentage of T cells is at least 65–90% of the lymphoid infiltrate. The extensive reactive T-cell infiltrate in TCRBCL has been postulated to be an immune reaction to the neoplastic B cells (Abramson, 2006). Another possible source of the extensive T-cell population in TCRBCL may be an immune response to viral-associated antigens in the neoplastic B cells such as epstein bar virus (Eberle and Jaffe, 2011). Also, the CD3 antibody showed a positive reaction in the small reactive lymphocytes forming the expansile nodules in NLPHL, with rosettes of T lymphocytes commonly surrounding the neoplastic L&H cells, although such rosettes are not observed within the reactive T-cell population of T/HRBCL. Conversely, CD3 was negative in all cases of DLBCL and the result is similar to that reported by Inaba et al. (2001). The neoplastic B-cell ratio was 10–20% in our cases, confirmed by positivity for CD20, and this ratio is different from that reported in the study carried out by Farhi (1995), who reported that the percentage of large B cells was 10–35%, and the study carried out by Abramson, 2006, who reported that the atypical large lymphocytes composed less than 10% of the lymphocyte population. Neutrophils and eosinophils were almost absent. Also, the CD20 antibody showed a positive reaction in the large neoplastic cells (popcorn cells) in NLPHL and diffusely positive in all (neoplastic) cells of DLBCL. On low-power examination, the nodal architecture in TCRBCL is usually completely effaced with a diffuse pattern of tumor infiltration, although vague nodularity may be apparent, creating the potential for confusion with DLBCL and NLPHL. Thus, immunohistochemistry for CD15 and CD30 was performed and showed a negative reaction in all patients with DLBCL, TCRBCL, and NLPHL. These results are similar to those reported by Abramson, 2006. In addition, EMA immunohistochemistry was performed and showed a positive reaction in popcorn cells that was negative for the traditional Hodgkin’s markers CD15 and CD30.
In terms of the expression of the antiapoptotic gene Bcl-2, the results of our study showed a low percentage in patients with TCRBCL, and this may be an alternative explanation for the decreased number of tumor cells in TCRBCL in comparison with DLBCL as there is increased Bcl-2 expression in DLBCLs, providing resistance to apoptosis, thereby allowing more tumor cells to accumulate in DLBCLs. This result is similar to that reported by Felgar et al. (1998).
Finally, a conclusion can be drawn from the results of this study. Although the diagnosis of TCRBCL is sometimes difficult in cases of gray-zone lesions, especially in developing countries, careful attention must be paid at the time of diagnosis to distinguishing this entity from other lymphoid neoplasms characterized by a marked host immune reaction and nodular pattern because these diseases follow distinct natural histories and require different therapies such as DLBCL and NLPHL on the basis of accurate histologic and phenotypic criteria. This will ensure prompt target therapy and improved chance of survival in this category of patients.
Conflicts of interest
There are no conflicts of interest.
Abramson JS.T-cell/histiocyte-rich B-cell lymphoma: biology, diagnosis, and management.Oncologist2006;11:384–392.
Achten R, Verhoef G, Vanuytsel L, De Wolf-Peeters C.Histiocyte-rich, T-cell-rich B-cell lymphoma: a distinct diffuse large B-cell lymphoma subtype showing characteristic morphologic and immunophenotypic features.Histopathology2002;40:31–45.
Aki H, Tuzuner N, Ongoren S, Baslar Z, Soysal T, Ferhanoglu B, et al..T-cell-rich B-cell lymphoma: a clinicopathologic study of 21 cases and comparison with 43 cases of diffuse large B-cell lymphoma.Leuk Res2004;28:229–236.
Anagnostopoulos I, Hansmann M-L, Franssila K, Harris M, Harris NL, Jaffe ES, et al..European Task Force on Lymphoma project on lymphocyte predominance Hodgkin disease: histologic and immunohistologic analysis of submitted cases reveals 2 types of Hodgkin disease with a nodular growth pattern and abundant lymphocytes.Blood2000;96:1889–1899.
Bouabdallah R, Mounier N, Guettier C, Molina T, Ribrag V, Thieblemont C, et al..T-cell/histiocyte-rich large B-cell lymphomas and classical diffuse large B-cell lymphomas have similar outcome after chemotherapy: a matched-control analysis.J Clin Oncol2003;21:1271–1277.
Boudová L, Torlakovic E, Delabie J, Reimer P, Pfistner B, Wiedenmann S, et al..Nodular lymphocytic predominant Hodgkin lymphoma with nodules resembling T-cell/histiocytic rich B-cell lymphoma.Blood2003;15:3753–3758.
Brown JR, Freedman AS, Aster JCLister A, Connor RF.Pathobiology of diffuse large B cell lymphoma and primary mediastinal B cell lymphoma2011UpToDate.
Dogan A.Gray zone lymphomas.Hematology2005;10Suppl 1190–192.
Drakos E, Rassidakis GZ, Leventaki V, Cotta CV, Vega F, Medeiros LJ.Nodular lymphocyte predominant Hodgkin lymphoma with clusters of LP cells, acute inflammation, and fibrosis: a syncytial variant.Am J Surg Pathol2009;33:1725–1731.
Eberle FC, Jaffe ES.Gray zone lymphomas: abiologic experiment and a challenge for diagnosis and management.Ann Oncol2011;22Suppl 4iv64–iv66.
Farhi DC.T-cell-rich B-cell lymphoma: reflections on changes in hematopathology.Am J Clin Pathol1995;103:4–5.
Felgar RE, Steward KR, Cousar JB, Macon WR.T-cell-rich large-B-cell lymphomas contain non-activated CD8+ cytolytic T cells, show increased tumor cell apoptosis, and have lower Bcl-2 expression than diffuse large-B-cell lymphomas.Am J Pathol1998;153:1707–1715.
Fraga M, García-Rivero A, Sánchez-Verde L, Forteza J, Piris MA.T-cell/histiocyte-rich large B-cell lymphoma is a disseminated aggressive neoplasm: differential diagnosis from Hodgkin’s lymphoma.Histopathology2002;41:216–229.
Greer JP, Macon WR, Lamar RE, Wolff SN, Stein RS, Flexner JM, et al..T-cell-rich B-cell lymphomas: diagnosis and response to therapy of 44 patients.J Clin Oncol1995;13:1742–1750.
Inaba T, Shimazaki C, Sumikuma T, Nakagawa M.T-cell associated antigen-positive B-cell lymphoma.Leuk Lymphoma2001;42:1161–1171.
Lones MA, Cairo MS, Perkins SL.T-cell-rich large B-cell lymphoma in children and adolescents: a clinicopathologic report of six cases from the Children’s Cancer Group Study CCG-5961.Cancer2000;88:2378–2386.
Morton LM, Wang SS, Devesa SS, Hartge P, Weisenburger DD, Linet MS.Lymphoma incidence patterns by WHO subtype in the United States, 1992–2001.Blood2006;107:265–276.
Müller-Hermelink H-K, Rüdiger T.Nodular lymphocyte-predominant Hodgkin lymphoma with nodules resembling T-cell/histiocyte-rich B-cell lymphoma: differential diagnosis between nodular lymphocyte-predominant Hodgkin lymphoma and T-cell/histiocyte-rich B-cell lymphoma.Blood2003;102:3753–3758.
Olaniyi JA, Oluwasola AO, Ibijola A.Therapy outcome of a T-cell-rich-B-cell lymphoma (TCRBCL) patient with R-CHOP in Ibadan, Nigeria: a case report.Mediterr J Hematol Infect Dis2011;3:e2011008.
Pijuan L, Vicioso L, Bellosillo B, Ferrer MD, Baró T, Pedro C, et al..CD20-negative T-cell-rich B-cell lymphoma as a progression of a nodular lymphocyte-predominant Hodgkin’s lymphoma treated with rituximab: a molecular analysis using laser capture microdissection.Am J Surg Pathol2005;29:1399–1403.
Poppema S, Kluiver JL, Atayar C, Van Berg AD, Rosenwald A, Hummel M, et al..Report: workshop on mediastinal grey zone lymphoma.Eur J Haematol Suppl2005;75:45–52.
Sehn LH, Donaldson J, Chhanabhai M, Fitzgerald C, Gill K, Klasa R, et al..Introduction of combined CHOP plus rituximab therapy dramatically improved outcome of diffuse large B-cell lymphoma in British Columbia.J Clin Oncol2005;23:5027–5033.
Swerdlow SH, Campo E, Harris NL, et al..World Health Organization classification of tumors of haematopoietic and lymphoid tissue2008.Lyon:IARC Press.
Wood GS, Warnke R.Suppression of endogenous avidin-binding activity in tissues and its relevance to biotin–avidin detection systems.J Histochem Cytochem1981;29:1196–1204.
©2013Egyptian Journal of Pathology