Spatially and Clonally Multiple B‐/Plasma Cell Proliferations in T Follicular Helper Cell Lymphoma: A Case Report

ABSTRACT

T follicular helper cell lymphoma (TFHL) is a systemic T‐cell lymphoma with a phenotype reminiscent of TFH cells and frequent RHOA p.G17V mutation. Here, we report a case of TFHL with spatially multiple and clonally independent B‐/plasma cell proliferations (B/PC‐Ps), which posed diagnostic difficulties. An 87‐year‐old female presented with a skin rash, and a skin biopsy from the chest was initially diagnosed as low‐grade B‐cell lymphoma with plasmacytic differentiation. Three months after rituximab monotherapy, bilateral pleural effusions developed. Cytology of the pleural effusion revealed abnormal lymphocytes and plasma cells. The abnormal T‐cells were positive for CD3, CD4, CD10, and PD1, while the neoplastic plasma cells showed kappa‐predominant light‐chain restriction. Additional cutaneous biopsies from the left forearm and upper arm revealed Epstein‐Barr virus‐positive and ‐negative B/PC‐Ps, respectively. Polymerase chain reaction identified RHOA p.G17V and the same monoclonal rearrangement of T‐cell receptor gamma gene (TRG) in all the four specimens. In addition, three cutaneous lesions showed distinct monoclonal immunoglobulin heavy chain gene (IGH) rearrangements. This case demonstrates phenotypic and clonal diversity of monoclonal B/PC‐Ps in TFHL, which can obscure the neoplastic TFH proliferation. In such cases, careful clinicopathological correlation and the integration of immunophenotypic and genetic analyses may aid in the diagnosis.

A patient with TFHL developed spatiotemporally, pathologically, and clonally multiple B/PC‐Ps. The cellular composition of the neoplastic TFH cells and monoclonal B/PC‐Ps significantly differed by time and location; the pleural effusion was T‐cell‐predominant, whereas cutaneous lesions predominantly consisted of monoclonal B/PC‐Ps, which obscured TFHL. The clonal relationship between neoplastic TFH cells and monoclonal B/PCPs was eventually elucidated by molecular analyses.

Abbreviations

AITL

angioimmunoblastic T‐cell lymphoma

B/PC‐P

B‐/plasma cell proliferation

EBER

EBV‐encoded small RNA

EBV

Epstein‐Barr virus

FTCL

follicular T‐cell lymphoma

HTLV‐1

human T‐lymphotropic virus 1

IGH

immunoglobulin heavy chain

ISH

in situ hybridization

NOS

not otherwise specified

PCR

polymerase chain reaction

R‐mini‐CHOP

rituximab combined with low‐dose doxorubicin, cyclophosphamide, vincristine, and prednisolone

TFH

T follicular helper

TFHL

T follicular helper cell lymphoma

TRG

T‐cell receptor gamma

WHO

World Health Organization

1. Introduction

T follicular helper (TFH) cells represent a subset of CD4‐positive helper T cells that localize to the germinal center of secondary lymphoid follicles and support B‐cell maturation. T‐cell neoplasms exhibiting the TFH phenotype have been recognized, and the revised 4th edition of the World Health Organization (WHO) classification introduced an umbrella category that included angioimmunoblastic T‐cell lymphoma (AITL), follicular T‐cell lymphoma (FTCL), and nodal peripheral T‐cell lymphoma with TFH phenotype [1]. Now, both the 5th edition of the WHO classification and the International Consensus Classification consider T‐follicular helper cell lymphomas (TFHLs) a single family comprising three histological types: angioimmunoblastic, follicular, and not otherwise specified (NOS) [2, 3]. The 5th WHO classification defines the TFH phenotype as the expression of CD4 and two or more TFH markers, including PD1, ICOS, CXCL13, CD10, and BCL6 [2, 4, 5].

TFHLs occasionally develop monoclonal B‐/plasma cell proliferation (B/PC‐P), which is often associated with Epstein‐Barr virus (EBV) infection [6]. However, some cases of EBV‐negative monoclonal B/PC‐P have also been reported [6, 7]. The in‐depth pathogenesis of monoclonal B/PC‐P in TFHL remains to be elucidated; however, the clonal hematopoiesis and/or the immunosuppressive microenvironment in TFHL presumably induce monoclonal B/PC‐P. TET2 and DNMT3A mutations have been shown to be positive not only in neoplastic TFH cells but also in B‐/plasma cells [8, 9], supporting their origin as common clonal hematopoietic precursors. Immunosuppressive microenvironment generated by the dysregulated TFH function of the neoplastic cells may also foster EBV‐positive B‐cell proliferation and neoplasms [10].

Notably, monoclonal B/PC‐Ps in TFHL are not always subsequent findings. Monoclonal B/PC‐Ps can be predominant in tumor tissues and can arise before the detection of TFHL, posing a significant diagnostic challenge. Here, we report a challenging case of TFHL, which developed spatially and clonally multiple B/PC‐Ps in the skin and pleural effusion.

2. Clinical Summary

An 87‐year‐old female with non‐itching erythema on her trunk and extremities visited a community hospital. She had a medical history of hepatitis B viral infection. The human T‐lymphotropic virus 1 (HTLV‐1) was serologically negative. A dermatologist prescribed topical medications; however, no improvement was observed. Eight months after the initial visit, she visited another community hospital and underwent skin biopsy (namely, initial skin biopsy). Positron emission tomography revealed an equivocal cervical lymph node adenopathy with a standardized uptake value maximum (SUV_max) 3.45. Following the diagnosis of low‐grade B‐cell lymphoma with plasmacytic differentiation, she visited our hospital and was treated with five courses of rituximab monotherapy.

Three months after rituximab monotherapy, or twelve months after the initial visit, cough and dyspnea appeared. Computed tomography revealed bilateral pleural effusions and consolidation in the bilateral lungs. There were several slightly enlarged lymph nodes above the diaphragm and at the left pulmonary hilum; however, it was difficult to distinguish reactive or neoplastic by imaging analysis alone. No other lesion was observed. Laboratory testing revealed thrombocytopenia (77 000/µL) and an elevated soluble interleukin‐2 receptor level (1,264 U/mL). No monoclonal proteins were detected in the serum or urine. The pleural effusion was sampled for cytopathological and histopathological diagnoses, and a cutaneous nodule in the left forearm and a telangiectatic erythema in the upper arm were biopsied (secondary skin biopsies).

Following the diagnosis of TFHL with monoclonal B/PC‐Ps, five courses of rituximab combined with low‐dose doxorubicin, cyclophosphamide, vincristine, and prednisolone (R‐mini‐CHOP) were administered. After the initiation of treatment, pleural effusion decreased, and oxygenation improved. However, following the chemotherapy, the patient developed paralytic bowel obstruction, and the treatment was discontinued.

3. Pathological Findings

3.1. Initial Skin Biopsy of the Chest

Plasma cells with eccentric nuclei and no significant atypia formed dense clusters in the dermis (Figure 1a,b). These plasma cells were positive for CD138 (Figure 1c) but negative for CD20 and CD56. In situ hybridization (ISH) showed kappa‐predominant light chain restriction in plasma cells (Figure 1d,e). CD20‐positive B‐cells and CD3‐positive T‐cells were small and mainly distributed around clusters of plasma cells (Figure 1f,g). There were only a few EBV‐encoded small RNA (EBER)‐positive cells in the background. Initially, the differential diagnoses included low‐grade B‐cell lymphoma with prominent plasmacytic differentiation and plasma cell neoplasm. However, based on the following molecular findings, we eventually concluded that the cutaneous lesion was a monoclonal B/PC‐P associated with TFHL.

Figure 1.

Histological and immunohistochemical findings of the initial cutaneous lesion of the chest. (a) Hematoxylin and eosin‐stained sections show a dense cellular infiltration in the dermis. (b) Infiltrating cells exhibit plasma cell morphology. (c) Infiltrating plasma cells are weakly positive for CD138. (d and e) In situ hybridization for kappa (d) and lambda (e) shows kappa‐predominant light chain restriction. (f) CD20‐positive B‐cells and (g) CD3‐positive T‐cells are distributed around the clusters of plasma cells.

3.2. Pleural Effusion

Pleural effusion smears were stained with Papanicolaou and May‐Giemsa stains. Small‐to‐medium‐sized, monotonous, and abnormal lymphocytes were predominantly observed, and scattered plasma cells, eosinophils, and neutrophils were intermixed (Figure 2a–c). The abnormal lymphocytes had cleaved nuclei with irregular nuclear contours and reticulated chromatin. The cytoplasm was pale to slightly basophilic upon Giemsa staining (Figure 2d). Immunohistochemistry on the cell block revealed that the abnormal lymphocytes (Figure 3a) were predominantly T‐cells positive for (cytoplasmic) CD3 (Figure 3b), weakly positive for CD5 (Figure 3c) and CD7 (Figure 3d). These T‐cells also expressed CD10 (Figure 3e), PD1, and BCL6, corresponding to the TFH phenotype defined by the 5th WHO classification. While there were few B‐cells expressing both CD20 (Figure 3f) and CD79a, plasma cells positive for CD79a and CD138 and negative for CD20 were present. ISH revealed that these plasma cells exhibited kappa‐predominant light chain restriction, suggesting a monoclonal/neoplastic process. Flow cytometry also detected an abnormal CD4‐ and CD10‐positive T‐cell population with an aberrant loss of surface CD3. Karyotyping by Giemsa banding detected two distinct clones: 46,XX,t(3;16)(q21;q24)[9]/47,XX, + 5,del(6)(q?)[5]/46,XX[6]. Collectively, we diagnosed the malignant pleural effusion as a TFHL with neoplastic plasma cell proliferation.

Figure 2.

Cytological findings of pleural effusion. (a) Papanicolaou staining shows abnormal small‐to‐medium‐sized lymphocytes. (b) Abnormal lymphocytes with irregular nuclear contours. (c) May‐Giemsa staining reveals atypical lymphocytes intermingled with scattered plasma cells, eosinophils, and neutrophils. (d) Abnormal lymphocytes display irregular nuclear contours, nuclear slits, reticulate chromatin, and pale to slightly basophilic cytoplasm.

Figure 3.

Immunophenotype of abnormal lymphocytes and plasma cells in the pleural effusion by immunochemistry and flow cytometry. (a) Hematoxylin and eosin‐stained sections. (b) Lymphocytes are predominantly CD3‐positive T‐cells with (c) diminished CD5 and (d) CD7 expression. (e) T‐cells are partially positive for CD10. (f) Few CD20‐positive B‐cells are observed.

3.3. Secondary Skin Biopsy of the Left Forearm

The lesion predominantly consisted of abnormal B‐cells that were intensely positive for CD79a and weakly positive for CD20 (Figure 4a–c). Unlike the initial chest cutaneous lesion and pleural effusion, these abnormal B‐cells were EBV‐positive by ISH for EBER (Figure 4d). A small number of CD138‐positive plasma cells were intermixed; however, these plasma cells expressed polytypic light chains. A small number of abnormal T‐cells expressing CD10 were also observed in the periphery of the lesion. Collectively, this lesion was diagnosed as an EBV‐positive monoclonal B‐cell proliferation associated with TFHL.

Figure 4.

Histopathology of additional cutaneous lesions in the left forearm and upper arm. (a–d) Hematoxylin and eosin‐stained sections from the left forearm show a proliferation of mononuclear cells (a) positive for CD79a (b), weakly positive for CD20 (c) and positive for EBER (d). (e–h) Hematoxylin and eosin‐stained sections from the left upper arm shows another proliferation of mononuclear cells (e) positive for CD79a (f) and CD20 (g), but negative for EBER (h).

3.4. Secondary Skin Biopsy of the Left Upper Arm

The lesion mainly contained neoplastic proliferations of EBV‐negative B‐cells expressing CD79a and CD20 (Figure 4e–h), kappa‐restricted monotypic plasmacytes, and a small number of abnormal CD10‐positive T‐cells. Based on these findings, this lesion was diagnosed as an EBV‐negative monoclonal B‐cell proliferation associated with TFHL.

3.5. Genetic Analysis

The results of the genetic analysis are summarized in Table 1. RHOA p.G17V was examined by locked nucleic acid‐peptide nucleic acid‐clamp polymerase chain reaction (PCR) [11]. RHOA p.G17V was positive in all the four samples from pleural effusion and three cutaneous lesions. We also examined T‐cell and B‐cell clonality by PCR for T‐cell receptor gamma (TRG) and immunoglobulin heavy chain (IGH) gene rearrangements [12]. Identical monoclonal TRG rearrangements were observed in all the four specimens (215 bp in tube A). With regard to B‐cell clonality, the initial chest skin biopsy and left pleural effusion showed the same monoclonal IGH rearrangements (115 bp in FR3), whereas skin biopsies from the left forearm and upper arm showed monoclonal but independent rearrangements (152 bp and 128 bp in FR3, respectively). Overall, monoclonal T‐cells with RHOA p.G17V were consistently present in all lesions throughout the clinical course; however, the cellular composition differed significantly depending on time and location. Monoclonal B/PC‐P was detected in all specimens; however, the latter two cutaneous lesions were clonally independent and distinct from the initial two samples obtained from the chest skin and pleural effusion.

Table 1.

Summary of genetic analysis.

	Skin, chest	Pleural effusion	Skin, forearm	Skin, upper arm
RHOA p.G17V	+	+	+	+
T‐cell clonality
TRG‐A	215 bp, equivocal	215 bp	215 bp	215 bp
TRG‐B	No peak	No peak	No peak	No peak
B‐cell/plasma cell clonality
IGH‐FR1	Equivocal	No peak	279 bp	Equivocal
IGH‐FR2	Equivocal	No peak	218 bp	262 bp
IGH‐FR3	115 bp	115 bp	152 bp	128 bp
EBV	+, a few in background	n/a	+	—
Karyotype	n/a	46,XX,t(3;16)(q21;q24)[9]	n/a	n/a
		47,XX,+5,del(6)(q?)[5]
		46,XX[6]

Abbreviations: bp, base pair; EBV, Epstein‐Barr virus; n/a, not available.

4. Discussion

This is an instructive case showing spatially, pathologically, and clonally multiple B/PC‐Ps in a single patient with TFHL. The cellular composition of the neoplastic TFH cells and monoclonal B/PC‐Ps significantly differed by time and location; pleural effusion contained abundant neoplastic TFH cells as the dominant component, whereas cutaneous lesions predominantly consisted of monoclonal B/PC‐Ps masking TFHL. The clonal relationship between neoplastic TFH cells and monoclonal B/PC‐Ps was eventually elucidated by molecular analyses.

B/PC‐P is an occasional finding seen in TFHL. A study by Segura‐Rivera et al. reviewed the clinicopathologic characteristics of nTFHLs with B/PC‐P and reported that 22% of the B/PC‐Ps occurred before the diagnosis of nTFHL [6], as seen in the present case. The pathogenesis of B/PC‐P has not been fully elucidated; however, the complication of the unique TFH cell‐mediated microenvironment and clonal hematopoiesis may induce B/PC‐P. TFH cells play a crucial role in the adaptive immune response by interacting with B‐cells to promote antibody production, and the neoplastic TFH cells induce dysregulated immune responses, leading to prominent polymorphous infiltrates, follicular dendritic cell expansion, and EBV reactivation in TFHL [10]. This unique immunosuppressive microenvironment in TFHL likely fosters clonal B/PC‐Ps. Another potential etiology is clonal hematopoiesis. TFHL is characterized by frequent clonal hematopoiesis‐associated mutations, including TET2 and DNMT3A mutations, and these mutations are shown to be present in both neoplastic TFH cells and B/PC‐Ps [8, 9]. TET2‐ and/or DNMT3A‐mutated hematopoietic progenitor cells likely lead to the parallel evolution of neoplastic TFH cells and B/PC‐Ps [9]. In the present case, the role of hematopoiesis is obscure, as we did not investigate clonal hematopoiesis‐associated mutations. Nonetheless, anatomically multiple and clonally independent B/PC‐Ps accompanied by neoplastic TFH cells suggest the importance of the local microenvironment mediated by neoplastic TFH cells.

It is challenging to diagnose lymphoma in body fluids by cytomorphology alone, especially when diverse inflammatory cells are intermixed or when lymphoma cells exhibit minimal cellular atypia. In such situations, immunophenotyping by flow cytometry and molecular testing can aid in the diagnosis. Tan et al. reported a case in which a primary diagnosis of AITL was made using body fluid specimens [13]. Cytological findings revealed a predominance of small‐to‐intermediate‐sized lymphocytes with irregular nuclear contours; scattered Hodgkin/Reed‐Sternberg‐like cells were also present. They showed that a definitive diagnosis of AITL can be made using body fluid specimens by combining cytomorphological features with immunostaining, flow cytometry, and next‐generation sequencing. In our case, TFHL‐associated monoclonal B/PC‐Ps made the differential diagnosis more complicated, and the flow cytometry of the pleural effusion suggested the presence of T‐cell neoplasms. Further clonal relationships were revealed using RHOA p.G17V testing and PCR for IGH and TRG rearrangements.

In conclusion, monoclonal B/PC‐Ps in TFHL can be spatially multiple and clonally independent, and their cellular composition is significantly different by lesions. Moreover, monoclonal B/PC‐Ps can precede the identification of TFHL and occasionally dominate the lesion, masking neoplastic TFH cells. In such challenging cases, it is important to integrate clinicopathological correlation, detailed morphological observations, and flow cytometry and molecular analyses.

Author Contributions

Yuka Yokota drafted the manuscript based on clinical and pathological information. Kazunari Kasai, Masataka Kawai, Ippei Tahara, Tomohiro Inoue, conducted pathological analysis and diagnosis and reviewed the manuscript. Yuma Sakamoto, Megumi Koshiish‐Yamada, and Kazunari Kasai managed the patient and revised the article critically for important clinical content. Naoki Oishi supervised the pathological diagnosis and molecular analysis and reviewed the manuscript. Tetsuo Kondo reviewed the manuscript. All authors have contributed to the intellectual content of this manuscript and have approved the final manuscript as submitted.

Ethics Statement

This study was conducted in accordance with the principles of the Declaration of Helsinki (1975). Ethical approval was not required for this single‐case presentation in accordance with the local/national guidelines.

Conflicts of Interest

Dr. Naoki Oishi and Dr. Tetsuo Kondo are Editorial Board members of Pathology International and coauthors of this article. To minimize bias, they were excluded from all editorial decision‐making related to the acceptance of this article for publication.