Bone Marrow Histology in Monoclonal B-Cell Lymphocytosis Shows Various B-Cell Infiltration Patterns

Abstract

Table 1

Infiltration Pattern and Immunophenotype of Bone Marrow B-Cell Infiltration in Patients With MBL

Open in new tab

Materials and Methods

Monoclonal B-cell lymphocytosis (MBL) was first recognized as an indolent variant of chronic lymphocytic leukemia (CLL) in patients with Rai stage 0 disease. The patients showed no disease progression, even after a period of more than 24 years.¹ A similar indolent form of CLL was later described in otherwise healthy family members of patients with CLL and in cohort studies of outpatients.^2-5 The condition has since been called MBL, and its natural course and genetics have been studied extensively.⁶ MBL is now defined as less than 5 × 10⁹/L monoclonal B cells in the blood of otherwise healthy patients. Large population-based screening studies have revealed that the prevalence of MBL increases with age, with a frequency of up to 14% at an age of more than 62 years.^7-9 Virtually all patients with clinically evident CLL had a preceding MBL, whereas the risk for patients with CLL-like MBL to develop leukemia is about 1% per year.^7,8 MBL likely arises through chronic stimulation of a limited B-cell repertoire in older individuals. Thus, it is a phenomenon associated with senescence of the lymphoid system.¹⁰ MBL shows genetic changes such as 13q14 deletion, trisomy 12, and deletion of 17p and 11q, as also seen in CLL.⁷ CD38 and ZAP-70 expression, VH gene usage, and cytogenetic abnormalities have not allowed prediction of the transition of MBL to CLL, but B-cell counts, especially low-count MBL, have been associated with a very low risk of progression to CLL.^11,12 In addition to the more frequent CLL-like immunophenotype of MBL, characterized by CD5 and CD23 expression and weak expression of CD20 and immunoglobulins, an atypical CLL-like phenotype with moderate or strong CD20 and immunoglobulin expression is also recognized. Furthermore, MBL with a non–CLL-like phenotype, characterized by the absence of CD5 expression, has also been described.^5,6 In view of the frequency of MBL in the general population, MBL is often seen in combination with other diseases, including those for which a trephine biopsy regularly is performed as part of the diagnostic workup. Trephine biopsy is part of the panel of tests performed for a large number of diseases, including primary nonmalignant as well as malignant hematologic diseases, nonhematologic malignancies, and certain infectious diseases. Hitherto, there have been no systematic studies of bone marrow infiltration, including bone marrow histologic findings, in patients with MBL despite its high frequency in the population. We therefore investigated whether MBL involved the bone marrow and, if so, what its histology was like and whether it could be differentiated from bone marrow involvement by lymphoma. We studied the bone marrow histology in 26 patients with well-documented MBL.

Patients were selected from the database of the Department of Pathology, Oslo University Hospital, Oslo, Norway. Only patients with an established MBL diagnosed between 2008 and 2011 and from whom a trephine biopsy was procured at the time of MBL diagnosis were selected for this study. All patients with non-Hodgkin lymphoma were excluded from the study, with the exception of 2 cases in which lymphoma involvement was easy to distinguish from MBL by morphology and immunohistochemistry. In total, 26 patients were retrieved from the database, including 18 men and 8 women with a mean age 70 years (range, 50-94 years). The patients underwent bone marrow examination as part of the clinical workup of their initial symptoms or primary disease. In 17 patients who presented with nonspecific symptoms such as fatigue, weight loss, or fever, no disease was eventually diagnosed. Chronic myeloproliferative disease was diagnosed in 4 patients and myelodysplastic syndrome in 1 patient. Three patients had a plasma cell neoplasia, 1 had a primary cutaneous follicular lymphoma, and 1 had a hepatosplenic T-cell lymphoma. One patient had renal cell carcinoma. A bone marrow aspirate had also been procured in all 26 patients.

Flow cytometry analyses were performed on blood and bone marrow samples from all 26 patients. Samples were anticoagulated with heparin. Between 0.5 and 1 × 10⁶ cells were stained for surface antigens. For samples analyzed before 2011, a 4-color analysis¹³ was performed with the following antibody combinations labeled with fluorescein isothiocyanate (FITC)/phycoerythrin (Pe)/peridinin-chlorophyll cyanine 5.5 (PercPCY5.5)/allophycocyanin (APC): (1) CD20/CD5/CD19/CD43, (2) FMC7/CD23/CD19/CD5, (3) Igκ/Igλ/CD20/CD19, (4) CD22/CD24/CD19/CD34, and (5) cyBcl2/CD10/CD19/CD38. From 2011, an 8-color flow cytometry analysis¹⁴ was used with the following antibody combinations labeled with Pacific Blue/e450 (PB/e450), Krome Orange (KO), FITC/Pe/PercPCy5.5/phycoerythrin cyanine 7(PeCy7)/APC/APC Hilite7, or APC/cyanine 7 (APCH7/cy7): (1) CD20+CD4/CD45/CD8+Igλ/CD56+Igκ/CD5/CD19+TCRγδ/CD38 and (2) CD20/CD45/CD23/CD10/CD79b/CD19/CD200/CD43. All antibodies for the 4-color panels were purchased from Becton-Dickinson (San Jose, CA) except anti-FMC7, CD22, CD23, and anti-Bcl2, which were purchased from DAKO (Glostrup, Denmark). For the 8-color panels, CD56, CD5, CD3, and 79b were purchased from Becton-Dickinson; CD23 from DAKO; CD200 from eBioscience (San Diego, CA); CD8, κ, and λ from Cytognos (Salamanca, Spain); and the rest from Beckman Coulter (Brea, CA). Flow cytometry analysis was performed on a FACSCalibur or LSRII instrument (Becton-Dickinson) using CellQuest Pro and FACSDiva software (Becton-Dickinson), respectively.

The primary antibodies used for immunophenotypic analysis of the lymphoid infiltrates were anti-CD45, anti-CD20, anti-Bcl6, anti-Mum1, anti-IgA, anti-IgD, anti-IgG, anti-IgM, and anti-Ki67 (all from DAKO); anti-CD10, anti-Bcl2, anti-CD21, anti-CD23, anti-CD5, anti-CD43, and anti-cyclin D1 (all from Novocastra Laboratories, Newcastle upon Tyne, UK); anti-PAX5 (Becton-Dickinson); and anti-CD3 (Thermo Fisher Scientific, Fremont, CA). For all antibodies, heat-induced epitope retrieval was performed in a microwave oven by heating the slides for 5 minutes at 750 W and subsequently for 15 minutes at 500 W in retrieval buffer. The detection system EnVision (DAKO) was used for all antibodies. The stain was developed with 3,3′-diaminobenzidine and H₂O₂, and the slides were counterstained with hematoxylin. The immunostaining was performed in a DAKO Autostainer (DAKO) according to the instructions of the manufacturer.

Results

Patient Selection

Flow cytometry analysis revealed a CLL-, an atypical CLL-, or a non–CLL-like immunophenotype MBL in 11, 5, and 10 patients, respectively Table 1. Monoclonal B cells were also detected in the bone marrow samples of all 26 patients. The immunophenotypes of the cells in the bone marrow were identical to those of the respective blood samples.

Flow Cytometry

The trephine biopsy specimens showed no evidence of abnormal lymphoid infiltration in 6 of 26 patients. Infiltration of the bone marrow by small B cells was found in 20 patients. Three infiltration patterns were discerned: 7 patients showed small interstitial foci, varying from 1 to 5, with monotonous small lymphoid cells with a rounded nucleus. At the periphery of these foci, cells infiltrated between fat cells in small rows Image 1A-D. Eight patients showed more rounded lymphoid foci, varying from 1 to 9, without interstitial infiltration between fat cells at the periphery. These lymphoid foci consisted of small lymphoid cells with a slightly more irregular nuclear contour, as well as a few scattered immunoblasts and histiocytic cells Image 1E-H. Centrally located tiny germinal centers could occasionally be observed (Image 1H). The latter were highlighted by CD21 staining. Finally, 5 patients showed a discrete diffuse lymphocytosis that was only discerned upon immunohistochemical staining for CD20. Diffuse B-cell lymphocytosis in patients with MBL could be distinguished from the presence of hematogones by the homogeneous small size of B cells in the former and by the variable small to medium-large size in the latter. In addition, hematogones showed a variable CD20 expression in contrast to the homogeneous CD20 expression seen with MBL. Intrasinusoidal infiltration was not seen in any of the cases.

Of interest, there was a correlation between the infiltration pattern and the immunophenotype of MBL (Table 1). The interstitial pattern was exclusively seen in patients with MBL with a CLL-like or an atypical CLL-like immunophenotype. The 2 other patterns were most commonly seen in non–CLL-like MBL, although these patterns were also observed in 3 patients with a CLL-like or an atypical CLL-like MBL.

Immunohistochemistry

Discussion

Infiltration Pattern and Immunophenotype of Bone Marrow B-Cell Infiltration in Patients With MBL

The frequent involvement of bone marrow in patients with MBL displaying an infiltration pattern and immunophenotype that is indistinguishable from that of limited infiltration with lymphoma, especially CLL or marginal zone lymphoma, is of diagnostic importance. Our results indicate that a final diagnosis of lymphoma in the bone marrow should not be made in the absence of an established lymphoma diagnosis or in the absence of adequate clinical information. Whether the patient with limited lymphoid infiltration in the bone marrow, originally taken as part of the investigation for other diseases, should be screened for lymphoma and how extensive this screening should be is a matter of discussion. Noninvasive investigations with abdominal ultrasound and chest x-ray in addition to a complete blood cell count, whenever these have not been performed as part of the investigations for the patient’s primary disease, are most likely indicated, as recommended for MBL.^10,22

Whether the extent of bone marrow infiltration with MBL correlates with increased progression toward overt clinical disease is of interest but still needs to be studied. This question may be answered by studying a larger cohort of patients with a longer follow-up time.

This study was supported by a grant from the Norwegian Cancer Society.

References