Angioimmunoblastic T-Cell Lymphoma: an Epstein-Barr Virus-Associated, Bilineage Lymphoid Neoplasm

Surapong Sorcharoen, MD, Supaporn Suwiwat, PhD.* Winyou Mitarnun, MD.
Department of Pathology, Faculty of Medicine Prince of Songkla University, Hat-Yai, THAILAND 90110
Correspondence Winyou Mitarnun, MD
E-mail: miwinyou@medicine.psu.ac.th Tel: +66 074 45 1551 Fax: +66 074 212908
Received 13 Nevember 2013; Accepted 2 February 2014

Abstract

In this study, 50 cases of angioimmunoblastic T-cell lymphoma (AITL) were investigated for histological patterns, their association with Epstein-Barr virus (EBV) infection, B cell and T cell clonalities, and the clinical outcomes. Three histological patterns were identified: the classical pattern (34 cases), large-cell rich pattern (14 cases), and interfollicular pattern (2 cases). The median survival time of patients from this group was 36 months. There was no significant difference in survival between patients from the classical pattern and the large-cell rich pattern. EBV-encoded early RNAs (EBERs) in tumor cells were identified in 46% of the cases. The expression of EBERs had no correlation with the histological patterns. We detected T-cell receptor-gamma chain (TCR-γ) T-cell clones in 70% of the cases, and immunoglobulin heavy chain (IgH) B-cell clones in 32% of the cases. Bilineage monoclonal proliferation of both T and B cells were detected in 28% of the cases. Clonal B cell expansion had no correlation with number of B cells in the lymph node and the status of EBERs in the tumor cells. Based on the findings from molecular study, we believe that AITL is a bilineage lymphoid neoplasm and may contribute to the pathogenesis of composite lymphomas.

Keywords: Angioimmunoblastic T-cell lymphoma, Epstein-Barr virus, EBV-encoded early RNA, Clonality, T-cell receptor (TCR), immunoglobulin heavy chain (IgH), Composite lymphoma

Running Title: Angioimmunoblastic T-cell lymphoma

---

INTRODUCTION

Angioimmunoblastic T-cell lymphoma (AITL) is a systemic disease of peripheral (mature) T cell neoplasm found mainly in the lymph nodes but may also occur in the spleen, bone marrow, liver, and skin^1-3. Histologically, the lymph nodes show partial or totally obliteration of the normal architecture by a polymorphic infiltrate lymphocytes, plasma cells, eosinophils, histiocytes and a proliferation of venules and follicular dendritic cells (FDCs)^1,4. In many instances the histopathological features of malignancy are not readily identifiable. This disease was first described as an atypical reactive process, however, molecular studies have consistently shown clonal abnormalities, with monoclonal T cell proliferation in most cases and monoclonal B cell proliferation in some cases[3]. In Western countries, AITL is one of the more common specific subtypes of the peripheral T-cell lymphoma, accounting for approximately 15-20% of peripheral T-cell lymphoma cases. The primary site of this disease is the lymph nodes and it occurs mainly in the middle-aged and elderly (median age = 64 years) with an equal incidence in both sexes. In Thailand, the incidence of AITL is about 67% of nodal peripheral T-cell lymphoma cases, and occurs in every age group (median age = 51 years) with a male to female ratio of about 2:1⁵.

AITL is highly associated with latent Epstein-Barr virus (EBV) infection. The presence of EBV genomes, EBV-encoded small nuclear RNAs (EBERs) in tumor cells is detected in most of AITL cases^5-7. In one study, in about 26% of cases, EBERs were detected in both B and T cells, the majority of which were B cells of immunoblastic morphology⁷. In a study from Thailand⁵, EBERs-positive tumor cells were detected in 50% of AITL cases, and in most cases were detected in both background lymphocytes (T and B cells) and im-munoblasts. The presence of EBV genomes in tumor cells had no correlation with the clinical outcomes⁶.

Studies have found that T cell receptor (TCR) genes showed clonal rearrangement in 60% to 100% of cases^8,9, and clonal immunoglobulin heavy chain (IgH) gene rearrangements were found in 8% to 40% of cases^10,11. Composite lymphomas which composed of EBV-associated AITL and EBV-associated diffuse large B-cell lymphoma and EBV-associated diffuse large B-cell lymphoma arising in the preexisting EBV-associated AITL have also been reported^12,13.

The aim of this study was to investigate the clonality of the T and B cells in EBV-positive and EBV-negative AITL cases in Thailand and to correlate these findings with histological patterns and clinical outcomes.

MATERIALS AND METHODS

Patient Samples

All ofthe cases were from the Department of Pathology, Faculty of Medicine, Prince of Songkla University, Songkhla, Thailand. Consecutive cases from 1990 through 2007 were examined. There were 50 cases of AITL in the lymph nodes. The diagnosis of AITL was made according to the WHO classification!. Three patterns of AITL were recognized: classical, large-cell rich and interfollicular^14-16.

Morphological criteria of AITL subtypes

Classical pattern (Fig.1): This pattern showed partial or diffuse effacement of the nodal architecture, vascular proliferation with prominent abori-zation of high endothelial venules, extrafollicular meshwork of FDCs, an atypical population of CD3+ T cells, and large CD20+ B cells. The percentage of large cell was less than 10% of the total cell population. The cellular infiltration was admixed with variable numbers of eosinophils, plasma cells, histiocytes and occasionally Reed-Sternberg-like B cells and granulomatous formations were seen.

Figure 1 Angioimmunoblastic T-cell lymphoma, classical pattern.

 

Large-cell rich pattern (Fig.2): This subtype of AITL had histopathological changes similar to the classical pattern except for an increase in number of large cells. The large cell population (B and/or T cells) was more than 10% of the total cell population.

 

Figure 2 Angioimmunoblastic T-cell lymphoma, large-cell rich pattern.

Figure 3 Angioimmunoblastic T-cell lymphoma, mterfollicular pattern.

 

Interfollicular pattern (Fig.3): This pattern had paracortical expansion with hyperplastic germinal centers. The histopathology in this subtype consisted of the combination of hyperplastic germinal centers along with angioimmunoblastic reaction.

 

Immunohistochemistry

Immunohistochemical staining for the leukocyte common antigen (LCA), CD3, CD5, CD10, CD20, CD23 was performed on formalin-fixed, paraffin-embedded tissues using the following antibodies: monoclonal mouse antihuman LCA (CD45) (Dako Cytomation, Demmark), monoclonal mouse antihuman CD3 (Novocastra, UK), monoclonal mouse antihuman CD5 (Novocastra, UK), monoclonal mouse antihuman CD10 (Novocastra, UK), monoclonal mouse antihuman CD20 (Zymed Labaratories, USA) and monoclonal mouse antihuman CD23 (Zymed Laboratories, USA).

In situ hybridization study for EBV

Detection of EBV RNA expression was carried out using fluorescein-conjugated EBV oligonucleotides complementary to portions of the EBV-encoded early RNA transcripts (EBERs) (EBER1/2, Novocastra, UK) that were actively transcribed in latently infected cells, as previously described¹⁷. Appropriate positive and negative controls were run in every batch tested. Few positive cells or positive less than 5% of the total tumor cell population were considered to be negative⁵.

DNA extraction from the lymph nodes for PCR analysis

DNA extraction was performed using a QIAamp DNA Mini Kit (QIAGEN, Germany). Three to five pieces of tissue section (3-5 pm) were taken from paraffin blocks, deparaffinized, resuspended in 180 pL of buffer ATL, digested with 50 pL of proteinase K, mixed by vortexing and incubated at 55°C until the tissue was completely lysed. Additional steps followed the tissue protocol outlined in the manufacturer’s manual. The presence of DNA was confirmed by agarose-gel electrophoresis.

Clonality studies of the tumor cells

The TCR-γ chain gene was amplified using the method as previously described[18]. We used five separate reactions with the following oligonucleotide primers combination: VyI+VynI/IV+Jy1/2 (product sizes 70-95 base pairs), VyI+VyIII/IV+JPy1/2 (product size 80-100 base pairs), VyII+Jy1/2 (product sizes 150-180 base pairs) VyII+JPy1/2 (product sizes 160-190 base pairs), and VyI+JPy (product sizes 70-95 base pairs). The sequences of oligonucleotide primers were listed in Table 1. For the monoclonal rearrangement, one or two clearly visible band/s were identified.

Determination of IgH gene rearrangement was performed by seminested PCR using FR1c-JH,

FR2a-JH and FR3a-JR primer amplification (Table1) [19]. Briefly, each 25 pl reactions mixture contained 5x PCR buffer, 2 mM MgCl2, 200 pM of each dNTP, 10 pmol of each primer, 1.25 U of Fast Start Taq DNA polymerase (Roche Diagnostics GmbH, Mannheim, Germany) and about 200 ng of DNA. Each PCR cycle consisted of 1 min at 93°C, 1 min at 54°C for FR1c and FR3a, and 50°C for FR2a, and 1 min at 72°C. For the first round of PCR amplification, the primers FR1c, FR2a, FR3a and LJH were used. For the second round of PCR, the primers FR2a and FR3a were used with inner primer VLJH. The first round of PCR consisted of 30 cycles and 1% of the PCR amplification product was used as a template for the second round of PCR consisting of 20 cycles. The PCR products were analyzed using 6% and 10% nondenaturing polyacrylamide gel electrophoresis, stained with ethidium bromide.

 

Table 1 Sequence of primers Primer
 

 

RESULTS

Clinical data

The patients consisted of 35 males and 15 females, (male: female=2.3:1), ranging in age from 17 years to 84 years; 5 patients (10%) were under 20 years of age. The median age was 56 years, and the mean (+SD) age was 50.7+19.3 years. Using the morphological criteria of the AITL subtypes, 34 patients were classified into the classical pattern (27 males, 7 females), 14 into the large-cell rich pattern (8 males, 6 females), and two female patients showed the interfollicular pattern. Kaplan-Meier survival estimates for patients from the classical pattern, large-cell rich pattern and all patterns combined are shown in Fig.4.

The median survival time of all patterns was 36 months. The median survival times of patients from the classical pattern and large-cell rich pattern were 24 months and 40 months, respectively. There was no significant difference in the survival profile between patients with the classical pattern and large-cell rich pattern (P=0.5464). The expression of EBERs in the tumor cells had no association with the disease outcome (P=0.5987).

Immunohistochemistry

Results of the immunohistochemistry are summarized in Table 2. The majority of lymphoid infiltrates were CD3+ cells in 72% of cases, and CD20+ cells in 14% of cases. Fourteen percent of the cases had an equal amount of CD3+ and CD20+ cellular infiltration. One patient showed dual staining of both CD3 and CD20 in nearly all tumor cells. Loss of CD5 expression on the surface of the T lymphocytes was detected in 6 of 48 cases (12.5%). CD10 expression was seen in 8 of 34 cases (23.5%) of the classical pattern and was not identified in the other two patterns. The fraction of CD10+ cells varied from around 5% to a maximum of 40% of the total lymphoid population. FDCs (CD23+ cells) were identified in 22 of the 47 cases (46.8%): 11 of the 31 cases in the classical pattern, 9 of the 14 cases in large-cell rich pattern and the 2 cases of the inter-follicular pattern. FDCs proliferation varied from minimal and limited to a few follicles, extended beyond the follicles into the paracortex, to the expansion into the interfollicular area and some FDCs were seen surrounding proliferating vessels. For the interfollicular pattern, the FDCs proliferation was limited to inside the hyperplastic follicles. In the large-cell rich pattern, the large cell populations were identified as CD3+ cells in 4 cases and CD20+ cells in 10 cases.

Figure 4 Probability of patient survival according to the patterns

 

Table 2 Immunohistochemistry of the AITL

 

Table 3 EBV genomes and the clonality studies of AITL

 

EBV genomes and the clonality studies

The results of the EBV genomes and the clonality studies are summarized in Table 3. EBERs were identified by ISH in 23 of the 50 cases (46%). The frequencies of EBERs-positive in each of the histological patterns were as follows: classical pattern = 50%, large-cell rich pattern = 35.7%, and interfollicular pattern = 50%.

Using five sets of oligonucleotide primers for the TCR-γ gene, monoclonal band/s were detected in 35 of the 50 cases (70%): 70.6% in the classical pattern, 64.3% in the large-cell rich pattern, and two of the two cases (100%) in the interfollicular pattern. For the IgH gene, monoclonal band/s were detected in 16 of the 50 cases (32%): 38.2% in the classical pattern, 14.3% in the large-cell rich pattern, and one of the two cases (50%) in the interfollicular pattern.

Monoclonalities of both the TCR-γ gene and the IgH gene were identified in 14 of the 50 cases (28%), indicating bileneages monoclonal proliferations of T and B lymphocytes in a single lymph node. Of sixteen patients who showed monoclonality of the B lymphocytes, 14 also showed monoclonality of the T lymphocytes. Of seven patients with B cells predominate (more than 50% of the lymphoid infiltrates), four showed IgH clonal products. There was no correlation between B cell clonality and the presence of B cell proliferation (Pearson chi-squared, P=0.30). The presence of EBERs in tumor cells also had no correlation with the clonalities of either T or B cells (Pearson chi-squared, P=0.58 and 0.70, respectively).

Composite lymphomas

Three additional patients with nodal composite lymphomas of synchronous occurrence of AITL combined with peripheral T-cell lymphoma, unspecified (PTCL-NOS), anaplastic large-cell lymphoma, ALK-positive (ALCL-ALK+), and non-Hodgkin lymphoma, diffuse large B-cell (DLBCL) are shown in Fig.5-Fig.7, respectively. Figure 5 shows a lymph node from an 81-year-old female patient. Some portions of the lymph node were replaced by AITL and another portions were replaced by PTCL-NOS. EBERs were detected in both AITL and PTCL-NOS. Figure 6 shows a lymph node from a 23-year-old male patient. The lymph node showed simultaneous occurrence of admixed AITL and ALCL-ALK+, and the EBERs were negative in the entire lymph node. Figure 7 shows a lymph node from a 36-year-old male patient. Half of the lymph node showed AITL features and the other half was DLBCL. EBERs were detected only in the AITL.

Figure 5 Composite lymphomas: angioimmunoblastic T-cell lymphoma (left) and peripheral T-cell lymphoma, NOS (right).

Figure 6 Composite lymphomas: angioimmunoblastic T-cell lymphoma (left) and anaplastic large cell lymphoma, ALK-positive (right).

Figure 7 Composite lymphomas: angioimmunoblastic T-cell lymphoma (left) and non-Hodgkin lymphoma, diffuse large B-cell (right).

 

DISCUSSION

Although AITL (previously called angio-immunoblastic lymphadenopathy, AILD) was first described as a non-neoplastic lymphoid proliferation associated with immunodeficiency, it is clear that the authors of the Revised European-American Classification of Lymphoid Neoplasms⁴ regard this entity as a peripheral T-cell lymphoma because most cases show clonal rearrangements of TCR genes. In the past decade, there have been many reports of IgH clonal rearrangements in AITL, up to 40% of reported cases^10,11. B cell and T cell clonal proliferations also have been reported in patients with PTCL-NOS, IgH clonal rearrangements were detected in 35% of PTCL-NOS patients²⁰.

In this study we investigated cell lineages in 50 cases of AITL. TCR-γ monoclonal band/s were observed in 35 patients (70%), and IgH monoclonal band/s were observed in 16 patients (32%). Bilineage monoclonal proliferation of T and B cells was observed in 14 patients (28%). The presence of EBV genomes has no association with B cell clonality. We believe that ATTL is a dual malignancy which occurs simultaneously in a lymph node instead of being a T cell neoplasm only. This hypothesis is supported by various reports of DLBCL arising in patients with preexisting AITL^13,21,22, and composite lymphomas of AITL and DLBCL¹². The above reported cases were EBV-associated with the possibility that the EBV-associated DLBCL was the secondary event in the AITL via EBV infection or reactivation followed by clonal expansion of an immortalized EBV-infected B cell clone¹². Other secondary lymphomas that occurred in AILT patients included small B-cell lymphomas, plasmacytoma, PTCL-NOS, and classical Hodgkin lymphoma²³²⁴. Our three additional patients with composite lymphomas confirmed the neoplastic nature of T and B cells in AITL which progressed to mature T-cell lymphoma or mature B-cell lymphoma, the transformation may or may not be associated with latent EBV infection. For the above reasons, AITL is considered to be a unique disease entity which should not be classified in either mature T-cell lymphoma or mature B-cell lymphoma.

The data presented here may indicate association of AITL with the latent EBV infection since EBERs positivity was detected in 46% of our cases which is somewhat lower than most reports from other countries (which vary from 37% to 97%)^6,7,25,26, which may be partially explain by noting that we used strict criteria for the interpretation of EBERs positivity. A few positive cells or positive less than 5% of the total tumor cell population were considered to be negative. We chose this strict standard, as a previous study we did showed that a small amount (less than 5%) of EBERs-positive cells was detected in 46% of nodal lymphoid hyperplasia cases⁵, a finding concurred with a previous report in which EBV DNA in peripheral blood CD3- lymphocytes was detected in 50% of the healthy individuals²⁷. These EBERs-positive cells were considered to be by-stander lymphocytes, and thus we decided to exclude them from our analysis, which may be resulted in the lower findings than in most other studies.

The clinical course in this group of patients was aggressive with a median survival of 36 months despite chemotherapy in most cases. Our median and overall survival times were better than in one reported European series¹⁴, which may have been due to younger age of our patients overall, with 10% being teenagers. The expression of EBERs in tumor cells had no association with the disease outcomes.

Overall, the two populations of classical pattern and large-cell rich pattern had no statistically significant difference in their survival profiles. A previous report also documented insignificant differences in a similar population in terms of clinical features, laboratory findings and treatment response¹⁴. The morphologic heterogeneity and lack of specific immunophenotypic marker for AITL could be a source of misdiagnosis to inexperienced pathologists, especially cases of AITL with an interfollicular pattern, AITL with very rich in epithelioid cells, and AITL with a granulomatous formations. The morphologic patterns have no clinical or prognostic values, but may be helpful in guiding pathologists to make the correct diagnosis of this disease.

ACKNOWLEDGMENTS AND REMARKS

This work was supported by a research grant from Faculty of Medicine, Prince of Songkla University. The study was approved by the Hospital Ethics Committee, Faculty of Medicine, Prince of Songkla University.

REFERENCES