AJCP / Original Article
Characterization of Tissue Findings in Bone Marrow Biopsy Specimens With Small Monoclonal B-Cell Populations Beverly P. Nelson, MD,1 Anmaar Abdul-Nabi, MD,1 Charles Goolsby, PhD,1 Jane Winter, MD,2 and LoAnn Peterson, MD1 From the 1Department of Pathology and 2Department of Medicine, Division of Hematology Oncology, Northwestern University, Feinberg Medical School, Chicago, IL. Key Words: Hematopathology; Flow cytometry; Immunopathology Am J Clin Pathol May 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
ABSTRACT Objectives: Bone marrows (BMs) with incidentally identified, small monotypic B-cell populations (MBPs) were evaluated. Methods: BM aspirates with MBPs representing 5% or less of total events by flow cytometry, less than 5.0 × 109/L B cells in blood, and no history of lymphoma or MBP with a different phenotype from prior lymphoma were selected. Clinical, immunophenotypic, and histologic findings were evaluated. Results: Forty-one of 3,052 BMs had MBPs at 5% or less of total events (median, 1%); 17 were females and 24 were males aged 30 to 87 years (median, 73 years). The MBPs were CD5– in 24, CD5+ resembling chronic lymphocytic leukemia (CLL) in 13, and CD5+ unlike CLL in four. Eighteen of 40 had lymphoid aggregates (LAs) with mostly T cells or a mixture of B and T cells, but three cases had B-cell–rich LAs. Conclusions: Unlike monoclonal B lymphocytosis in blood, MBPs in BMs were more commonly CD5–. Fortyfive percent of BMs had LAs; none were interpreted as lymphoma, although three were suspicious for B-cell lymphoma.
Monoclonal B lymphocytosis (MBL) is defined as the presence of less than 5.0 × 109/L monoclonal B cells in the blood of individuals who do not have extramedullary tissue involvement, lymphadenopathy/organomegaly, autoimmune disorders, or a B-cell lymphoma/leukemia.1-3 These small populations of monoclonal B cells are more common with increasing age and in males.4,5 MBL exhibits different phenotypes but is typically CD5+, resembling chronic lymphocytic leukemia (CLL), and less commonly CD5–.5,6 Small monotypic B-cell populations (MBPs) also have been reported in bone marrow aspirates; these MBPs have been interpreted either as of unknown significance or as evidence of lymphoma even when no morphologic evidence of lymphoma was found.7,8 Therefore, we performed this retrospective study to evaluate the clinical, immunophenotypic, and histologic findings of small MBPs identified in bone marrow biopsy specimens obtained for common indications such as unexplained cytopenias, myeloid malignancies, plasma cell neoplasms, and lymphoma staging. Cases with previously diagnosed lymphoma were included in this study only when the immunophenotype of the MBPs differed from the lymphoma.
Materials and Methods Case Selection This study was approved by the Institutional Review Board of Northwestern Memorial Hospital (Chicago, IL). In a retrospective review from January 2008 through December 2011 at Northwestern Memorial Hospital, we © American Society for Clinical Pathology
687
Am J Clin Pathol 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
687 687
Nelson et al / Bone Marrow Biopsy Specimens With Monoclonal B-Cell Populations
identified MBPs in 441 of 3,052 bone marrow aspirates, on which four- to eight-color flow cytometry immunophenotyping was done. Cases in which the MBPs comprised 5% or less of total events, the B-cell lymphocyte count in peripheral blood was less than 5.0 × 109/L, and there was either no history of lymphoma or the immunophenotype of the small MBPs differed from the established lymphoma were selected. These selection criteria identified 41 cases. The clinical findings, characteristics of the MBPs, and morphologic findings in the bone marrow biopsy specimens of these cases were further evaluated. Flow Cytometry Analysis Data collection was performed on a Becton Dickinson Biosciences (San Jose, CA) LSR II flow cytometer employing 488-nm and 635-nm laser excitation and using a 505-nm long pass (LP), a 550-nm dichroic LP (DCLP), a 530-nm band pass (BP), a 575-nm BP, a 685-nm DCLP, a 695-nm BP, a 735-nm DCLP, and a 780-nm BP for the 488-nm excitation source and a 735-nm DCLP, a 660-nm BP, and a 780-nm BP for the 635-nm excitation source. Negative and positive staining was determined by comparison to internal nonreactive (negative) and reactive (positive) small lymphocytes in the sample. Data analysis was performed using
FACSDiva software (Becton Dickinson Biosciences). The primary antibodies used and their combinations are listed in ❚Table 1❚. The samples were prepared and stained using a modified QPrep (Beckman Coulter, Miami, FL) procedure as previously described.9 For each specimen, greater than 100,000 events with a goal of a minimum of 2,000 B cells were collected. Any B-cell population or subset of B cells identified by heterogeneous antigen expression with a k/l ratio below 0.5 or above 5.0 was regarded as monotypic. The MBP cases were divided into three groups according to their phenotypes: CLL-like, CD5+ with phenotype not typical for CLL, and CD5–. Cases with the following phenotype were defined as CLL-like: CD5+, surface immunoglobulin light chain (SIg) dim/–, CD10–, CD20 dim+, CD23+, FMC7 dim+/–, and CD79b dim+/–. The CD5+ group with a phenotype not typical for CLL was defined as follows: CD5+, SIg dim/bright, CD20 bright+, CD10–, CD23+/–, FMC7 dim+/–, and CD79b+/–, while the CD5– group was defined as CD5– with any combination of staining patterns for the other markers. Cases were defined as dim+ if staining intensity overlapped with the nonreactive lymphocytes in the sample and bright if there was no overlap with nonreactive cells. Not all antigens were tested for each case.
❚Table 1❚ Antibodiesa Cluster Designation
Conjugate
CD2 PE-Cy7 CD3 APC CD4 PerCP-Cy5.5 CD5 FITC CD5 PE-Cy7 CD7 PE CD8 APC-H7 CD10 APC CD19 PerCP-Cy5.5 CD20 PE-Cy7 CD20 APC-H7 CD22 PE CD23 PE CD23 PE CD25 APC-H7 CD38 APC CD45 APC-H7 CD56 FITC CD57 FITC CD71 PE k APC k FITC l PE l FITC FMC7 FITC CD79b FITC CD52 APC a
Clone
Source
39C1.5 Beckman Coulter, Miami, FL UCHT1 Beckman Coulter SK3 BD Biosciences, San Jose, CA L17F12 BD Biosciences BL1a Beckman Coulter 8H8.1 Beckman Coulter SK1 BD Biosciences ALB1 Beckman Coulter SJ25C1 BD Biosciences B9E9 Beckman Coulter L27 BD Biosciences SJ10.1H11 Beckman Coulter HD50 Beckman Coulter EBVCS-5 BD Biosciences M-A251 BD Biosciences HB7 BD Biosciences 2D1 BD Biosciences NCAM16.2 BD Biosciences NC1 Beckman Coulter YDJ1.2.2 Beckman Coulter MHK-49 Biolegend, San Diego, CA Polyclonal Dako, Carpinteria, CA Polyclonal Dako MHL-38 Biolegend FMC7 Beckman Coulter 3A2-2E7 BD Bioscience HI186 Biolegend
Antibody combinations: CD5/FITC, CD7/PE, CD4/PerCPCy5.5, CD2/PECY7, CD3/APC, CD8/APCH7; CD56/57/FITC, CD22/PE, CD19/PerCPCy5.5, CD5/PECY7, CD52/ APC, CD45/APCH7; k/FITC, l/PE, CD19/PerCPCy5.5, CD20/PECY7, CD10/APC, CD45/APCH7; l/FITC, CD23/PE, CD19/PerCPCy5.5, CD5/PECY7, k/APC, CD25/ APCH7; FMC7/FITC, CD23/BD, CD19/PerCPCy5.5, CD5/PECY7, CD38/APC, CD20/APCH7; CD79b/FITC, CD71/PE, CD19/PerCPCy5.5, CD5/PECY7, CD38/APC, CD45/APCH7.
688 688
Am J Clin Pathol 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
© American Society for Clinical Pathology
AJCP / Original Article
Morphologic and Immunohistochemical Assessment All H&E- and immunohistochemical-stained sections performed on the bone marrow core biopsy specimens or clot sections as well as concurrent bone marrow aspirate and peripheral blood smears were reviewed by three observers (B.P.N., A.A.-N., and L.P.). All cases were evaluated for the presence of lymphoid aggregates, as well as their extent and patterns (paratrabecular, nonparatrabecular, or interstitial) of bone marrow involvement. The cellular composition of all lymphoid infiltrates was assessed by immunohistochemical stains for CD20 and/or PAX-5 and CD3. A cyclin D1 immunohistochemical stain was performed on all cases with a CD5+ phenotype that was not typical for CLL as well as both cases with a history of mantle cell lymphoma (MCL). The following are the primary antibodies that were used: monoclonal anti-CD20 at 1:800 dilution (clone LC6; Dako, Carpinteria, CA), ready-to-use monoclonal anti-CD3 (clone: PS1; Dako), monoclonal anti-PAX5 at 1:10 dilution (clone DAK. PAX5; Dako), and polyclonal anti–cyclin D1 at 1:40 dilution (clone SP4; Leica, Buffalo Grove, IL). Antigen retrieval and immunohistochemical stains were performed automatically on the BenchMark XT (Ventana Medical Systems, Tucson, AZ) using the Advanced HRP detection system. Tissues from reactive tonsil and normal lymph node were used as controls.
Results Clinical Characteristics Among the 41 patients, 17 were females and 24 were males ranging in age from 30 to 87 years (median, 73 years) ❚Table 2❚. One patient with T-cell prolymphocytic leukemia (T-PLL) had T lymphocytosis in the blood; the lymphocyte count was normal in all other cases. Indications for the bone marrow biopsies are summarized in Table 2. Lymphoma staging occurred in 12 cases (six diffuse large B-cell lymphomas [DLBCLs], three follicular lymphomas [FLs] , one marginal zone lymphoma [MZL], and two MCLs); other diagnoses included unexplained cytopenia (n = 9), acute myeloid leukemia (AML; n = 2), myelodysplastic syndrome (MDS; n = 6), blastic plasmacytoid dendritic cell neoplasm (n = 1), plasma cell neoplasms (n = 6 [plasma cell myeloma (PCM), n = 3; history of serum monoclonal protein, n = 3]), hepatitis C and cryoglobulinemia (n = 1), T-cell malignancy (n = 2 [T-cell large granular lymphocytic leukemia, n = 1; T-PLL, n = 1]), and primary myelofibrosis (n = 1). Organomegaly and/or a mass lesion were detected in 10 cases but were related to the previously diagnosed underlying diseases in all cases ❚Table 3❚. Flow Cytometric Immunophenotyping In 441 of the 3,052 bone marrow specimens in which flow cytometry was performed, an MBP was detected. In 400 © American Society for Clinical Pathology
❚Table 2❚ Patient Population and Indications for BM Biopsiesa Characteristic Value No. of cases studied Age, median (range), y Female/male sex Patients with lymphadenopathy/mass Size of BM core biopsy specimens, range, cm No. of BM core biopsy specimens, bilateral/single Reason for bone marrow biopsy Lymphoma staging Cytopenias AML/MDS Primary myelofibrosis Plasma cell neoplasms Primary amyloidosis BPDCN Hepatitis C with cryoglobulinemia T-cell malignancy
41 73 (30-87) 17/24 10 0.6-2.9 7/34 12 9 8 1 6 1 1 1 2
AML/MDS, acute myeloid leukemia/myelodysplastic syndrome; BM, bone marrow; BPDCN, blastic plasmacytoid dendritic cell neoplasm. a Values are presented as numbers unless otherwise indicated.
(91%) of 441 cases, the MBP was consistent with a previously diagnosed lymphoma or comprised more than 5% of the total events; therefore, these cases were not included in this study. In 41 (9.3%) of 441 cases, the MBP comprised 5% or less of total cellular events (range, 0.1%-5%; median, 1%) and represented 3% to 100% of the total CD19+ B cells. These 41 cases are included in this study. All showed either k (25) or l (15) restriction except one case that was negative for both surface and cytoplasmic immunoglobulin light chains. The small MBPs were present in cases without a history of lymphoma or had a different phenotype from the previously diagnosed lymphoma. Thirteen (31.7%) of 41 cases had a phenotype (CD5+, dim/– SIg, CD10–, CD20 dim+, CD23+, FMC7 dim+/–, CD79b dim+/–) that resembled CLL (Table 3). Four other cases had a CD5+ phenotype that was not typical of CLL; two were CD23– and two were CD23+, CD79b+, and FMC7+. The MBPs in most cases (24/41 [58.5%]) were CD5–. One of the CD5– cases was CD10+. The immunophenotype of the MBP for each case, indications for bone marrow biopsy, clone size, and other clinical and histologic data are summarized in Table 3. The MBP in the nine bone marrow biopsies performed to evaluate cytopenias showed a CD5+ CLL-like immunophenotype in two cases and CD5– immunophenotype in seven cases. Four of the eight bone marrow biopsies performed to evaluate AML/MDS had an MBP with a CD5+, CLL-like immunophenotype; one had a CD5+ immunophenotype not typical of CLL; and three were CD5–. The single bone marrow involved by primary myelofibrosis had a CD5+ MBP with a CLL-like immunophenotype. The three cases with a history of PCM all had CD5–, CD10–, and l light chain–restricted small MBP. The MBP light chain (l) differed from the PCM in two cases. In the remaining case, both the myeloma cells and the MBP were l
689
Am J Clin Pathol 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
689 689
Nelson et al / Bone Marrow Biopsy Specimens With Monoclonal B-Cell Populations
❚Table 3❚ Clinical and Immunophenotypic Findings of All Cases Clone Size, Lymphoma Case Age, y/ % of Total Biopsy Organomegaly/ History No. Sex MBP Phenotype (% of B Cells) Indication Mass Lesion and Phenotype
LA in BM T Cells and/or B Cells
CD5+, CLL-like phenotype 1 72/M S&CIg–, dim CD20+, CD5+, 1 (40) ↓ Platelets No None CD10–, CD23+, dim CD79b+ 2 78/M k+, CD5+, CD10–, CD23+, 0.2 (22) ↓ Platelets No None dim FMC7+, dim CD79b+ 3 62/M Dim k+, dim CD20+, CD5+, CD10– 1 (70) Primary mye- No None lofibrosis 4 82/M l+, CD20+, CD5+, CD10– 2 (90) AML ↑ Spleen None 5 61/F k+, CD20+, CD5+, CD10–, 0.2 (15) MDS No None CD23+, dim CD79b+, FMC7– 6 76/M k+, CD20+, CD5+, CD10–, CD23+, 1 (45) MDS No NA CD79b–, FMC7– 7 73/M k+, CD20+, CD5+, CD10–, dim 1 (15) AML No None CD23+, dim CD79b+, dim FMC+ 8 62/M CD5+, CD10–, CD20+, CD23+ k, 1 (22) DLBCL No CD5–, CD10–, CD19+, CD22+, FMC7–, CD79b– bcl–6+, MUM-1+, l 9 70/F k, CD5+, CD10–, CD23+, CD19+, 1 (27) DLBCL No CD5–, CD20+ FMC7–, CD20+, dim CD22+, CD79b– 10 75/M Dim k, CD5+, CD10–, CD23+, 1 (97) FL No k, CD10+, CD5–, dim CD20+,CD22+, CD25–, CD20+, CD22+, FMC7–, CD79b– CD25–, CD52+ 11 69/M CD19+, CD20+, CD22+, k+, CD5+, 5 (79) FL No IHC: CD20+, CD5– CD23+, CD10–, FMC7–, CD79b– 12 63/M k+, CD20+, CD5+, CD10–, CD23+, 1 (50) MG No IgM K FMC7–, dim CD79b+ 13 59/M l+, CD20+, CD5+, CD10–, CD23+, 0.1 (10) MCL No CD20+, CD5+, cyclin D1– CD10–, CD23–, BCL-6–, cyclin D1+ CD5– phenotype 14 87/F k+, CD20+, sCD22+, CD5–, CD10– 2 (60) ↓ Platelets No None 15 52/M k+, CD20+, CD5–, CD10–, CD23+, 5 (100) Anemia No None CD22+, FMC7+, CD79b– 16 30/M l+, CD20+, CD5–, CD10– 1 lymphoid aggregate
13 (31.7) 6 1 1 4 6
Characterization of Tissue Findings in Bone Marrow Biopsy Specimens With Small Monoclonal B-Cell Populations Beverly P. Nelson, MD,1 Anmaar Abdul-Nabi, MD,1 Charles Goolsby, PhD,1 Jane Winter, MD,2 and LoAnn Peterson, MD1 From the 1Department of Pathology and 2Department of Medicine, Division of Hematology Oncology, Northwestern University, Feinberg Medical School, Chicago, IL. Key Words: Hematopathology; Flow cytometry; Immunopathology Am J Clin Pathol May 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
ABSTRACT Objectives: Bone marrows (BMs) with incidentally identified, small monotypic B-cell populations (MBPs) were evaluated. Methods: BM aspirates with MBPs representing 5% or less of total events by flow cytometry, less than 5.0 × 109/L B cells in blood, and no history of lymphoma or MBP with a different phenotype from prior lymphoma were selected. Clinical, immunophenotypic, and histologic findings were evaluated. Results: Forty-one of 3,052 BMs had MBPs at 5% or less of total events (median, 1%); 17 were females and 24 were males aged 30 to 87 years (median, 73 years). The MBPs were CD5– in 24, CD5+ resembling chronic lymphocytic leukemia (CLL) in 13, and CD5+ unlike CLL in four. Eighteen of 40 had lymphoid aggregates (LAs) with mostly T cells or a mixture of B and T cells, but three cases had B-cell–rich LAs. Conclusions: Unlike monoclonal B lymphocytosis in blood, MBPs in BMs were more commonly CD5–. Fortyfive percent of BMs had LAs; none were interpreted as lymphoma, although three were suspicious for B-cell lymphoma.
Monoclonal B lymphocytosis (MBL) is defined as the presence of less than 5.0 × 109/L monoclonal B cells in the blood of individuals who do not have extramedullary tissue involvement, lymphadenopathy/organomegaly, autoimmune disorders, or a B-cell lymphoma/leukemia.1-3 These small populations of monoclonal B cells are more common with increasing age and in males.4,5 MBL exhibits different phenotypes but is typically CD5+, resembling chronic lymphocytic leukemia (CLL), and less commonly CD5–.5,6 Small monotypic B-cell populations (MBPs) also have been reported in bone marrow aspirates; these MBPs have been interpreted either as of unknown significance or as evidence of lymphoma even when no morphologic evidence of lymphoma was found.7,8 Therefore, we performed this retrospective study to evaluate the clinical, immunophenotypic, and histologic findings of small MBPs identified in bone marrow biopsy specimens obtained for common indications such as unexplained cytopenias, myeloid malignancies, plasma cell neoplasms, and lymphoma staging. Cases with previously diagnosed lymphoma were included in this study only when the immunophenotype of the MBPs differed from the lymphoma.
Materials and Methods Case Selection This study was approved by the Institutional Review Board of Northwestern Memorial Hospital (Chicago, IL). In a retrospective review from January 2008 through December 2011 at Northwestern Memorial Hospital, we © American Society for Clinical Pathology
687
Am J Clin Pathol 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
687 687
Nelson et al / Bone Marrow Biopsy Specimens With Monoclonal B-Cell Populations
identified MBPs in 441 of 3,052 bone marrow aspirates, on which four- to eight-color flow cytometry immunophenotyping was done. Cases in which the MBPs comprised 5% or less of total events, the B-cell lymphocyte count in peripheral blood was less than 5.0 × 109/L, and there was either no history of lymphoma or the immunophenotype of the small MBPs differed from the established lymphoma were selected. These selection criteria identified 41 cases. The clinical findings, characteristics of the MBPs, and morphologic findings in the bone marrow biopsy specimens of these cases were further evaluated. Flow Cytometry Analysis Data collection was performed on a Becton Dickinson Biosciences (San Jose, CA) LSR II flow cytometer employing 488-nm and 635-nm laser excitation and using a 505-nm long pass (LP), a 550-nm dichroic LP (DCLP), a 530-nm band pass (BP), a 575-nm BP, a 685-nm DCLP, a 695-nm BP, a 735-nm DCLP, and a 780-nm BP for the 488-nm excitation source and a 735-nm DCLP, a 660-nm BP, and a 780-nm BP for the 635-nm excitation source. Negative and positive staining was determined by comparison to internal nonreactive (negative) and reactive (positive) small lymphocytes in the sample. Data analysis was performed using
FACSDiva software (Becton Dickinson Biosciences). The primary antibodies used and their combinations are listed in ❚Table 1❚. The samples were prepared and stained using a modified QPrep (Beckman Coulter, Miami, FL) procedure as previously described.9 For each specimen, greater than 100,000 events with a goal of a minimum of 2,000 B cells were collected. Any B-cell population or subset of B cells identified by heterogeneous antigen expression with a k/l ratio below 0.5 or above 5.0 was regarded as monotypic. The MBP cases were divided into three groups according to their phenotypes: CLL-like, CD5+ with phenotype not typical for CLL, and CD5–. Cases with the following phenotype were defined as CLL-like: CD5+, surface immunoglobulin light chain (SIg) dim/–, CD10–, CD20 dim+, CD23+, FMC7 dim+/–, and CD79b dim+/–. The CD5+ group with a phenotype not typical for CLL was defined as follows: CD5+, SIg dim/bright, CD20 bright+, CD10–, CD23+/–, FMC7 dim+/–, and CD79b+/–, while the CD5– group was defined as CD5– with any combination of staining patterns for the other markers. Cases were defined as dim+ if staining intensity overlapped with the nonreactive lymphocytes in the sample and bright if there was no overlap with nonreactive cells. Not all antigens were tested for each case.
❚Table 1❚ Antibodiesa Cluster Designation
Conjugate
CD2 PE-Cy7 CD3 APC CD4 PerCP-Cy5.5 CD5 FITC CD5 PE-Cy7 CD7 PE CD8 APC-H7 CD10 APC CD19 PerCP-Cy5.5 CD20 PE-Cy7 CD20 APC-H7 CD22 PE CD23 PE CD23 PE CD25 APC-H7 CD38 APC CD45 APC-H7 CD56 FITC CD57 FITC CD71 PE k APC k FITC l PE l FITC FMC7 FITC CD79b FITC CD52 APC a
Clone
Source
39C1.5 Beckman Coulter, Miami, FL UCHT1 Beckman Coulter SK3 BD Biosciences, San Jose, CA L17F12 BD Biosciences BL1a Beckman Coulter 8H8.1 Beckman Coulter SK1 BD Biosciences ALB1 Beckman Coulter SJ25C1 BD Biosciences B9E9 Beckman Coulter L27 BD Biosciences SJ10.1H11 Beckman Coulter HD50 Beckman Coulter EBVCS-5 BD Biosciences M-A251 BD Biosciences HB7 BD Biosciences 2D1 BD Biosciences NCAM16.2 BD Biosciences NC1 Beckman Coulter YDJ1.2.2 Beckman Coulter MHK-49 Biolegend, San Diego, CA Polyclonal Dako, Carpinteria, CA Polyclonal Dako MHL-38 Biolegend FMC7 Beckman Coulter 3A2-2E7 BD Bioscience HI186 Biolegend
Antibody combinations: CD5/FITC, CD7/PE, CD4/PerCPCy5.5, CD2/PECY7, CD3/APC, CD8/APCH7; CD56/57/FITC, CD22/PE, CD19/PerCPCy5.5, CD5/PECY7, CD52/ APC, CD45/APCH7; k/FITC, l/PE, CD19/PerCPCy5.5, CD20/PECY7, CD10/APC, CD45/APCH7; l/FITC, CD23/PE, CD19/PerCPCy5.5, CD5/PECY7, k/APC, CD25/ APCH7; FMC7/FITC, CD23/BD, CD19/PerCPCy5.5, CD5/PECY7, CD38/APC, CD20/APCH7; CD79b/FITC, CD71/PE, CD19/PerCPCy5.5, CD5/PECY7, CD38/APC, CD45/APCH7.
688 688
Am J Clin Pathol 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
© American Society for Clinical Pathology
AJCP / Original Article
Morphologic and Immunohistochemical Assessment All H&E- and immunohistochemical-stained sections performed on the bone marrow core biopsy specimens or clot sections as well as concurrent bone marrow aspirate and peripheral blood smears were reviewed by three observers (B.P.N., A.A.-N., and L.P.). All cases were evaluated for the presence of lymphoid aggregates, as well as their extent and patterns (paratrabecular, nonparatrabecular, or interstitial) of bone marrow involvement. The cellular composition of all lymphoid infiltrates was assessed by immunohistochemical stains for CD20 and/or PAX-5 and CD3. A cyclin D1 immunohistochemical stain was performed on all cases with a CD5+ phenotype that was not typical for CLL as well as both cases with a history of mantle cell lymphoma (MCL). The following are the primary antibodies that were used: monoclonal anti-CD20 at 1:800 dilution (clone LC6; Dako, Carpinteria, CA), ready-to-use monoclonal anti-CD3 (clone: PS1; Dako), monoclonal anti-PAX5 at 1:10 dilution (clone DAK. PAX5; Dako), and polyclonal anti–cyclin D1 at 1:40 dilution (clone SP4; Leica, Buffalo Grove, IL). Antigen retrieval and immunohistochemical stains were performed automatically on the BenchMark XT (Ventana Medical Systems, Tucson, AZ) using the Advanced HRP detection system. Tissues from reactive tonsil and normal lymph node were used as controls.
Results Clinical Characteristics Among the 41 patients, 17 were females and 24 were males ranging in age from 30 to 87 years (median, 73 years) ❚Table 2❚. One patient with T-cell prolymphocytic leukemia (T-PLL) had T lymphocytosis in the blood; the lymphocyte count was normal in all other cases. Indications for the bone marrow biopsies are summarized in Table 2. Lymphoma staging occurred in 12 cases (six diffuse large B-cell lymphomas [DLBCLs], three follicular lymphomas [FLs] , one marginal zone lymphoma [MZL], and two MCLs); other diagnoses included unexplained cytopenia (n = 9), acute myeloid leukemia (AML; n = 2), myelodysplastic syndrome (MDS; n = 6), blastic plasmacytoid dendritic cell neoplasm (n = 1), plasma cell neoplasms (n = 6 [plasma cell myeloma (PCM), n = 3; history of serum monoclonal protein, n = 3]), hepatitis C and cryoglobulinemia (n = 1), T-cell malignancy (n = 2 [T-cell large granular lymphocytic leukemia, n = 1; T-PLL, n = 1]), and primary myelofibrosis (n = 1). Organomegaly and/or a mass lesion were detected in 10 cases but were related to the previously diagnosed underlying diseases in all cases ❚Table 3❚. Flow Cytometric Immunophenotyping In 441 of the 3,052 bone marrow specimens in which flow cytometry was performed, an MBP was detected. In 400 © American Society for Clinical Pathology
❚Table 2❚ Patient Population and Indications for BM Biopsiesa Characteristic Value No. of cases studied Age, median (range), y Female/male sex Patients with lymphadenopathy/mass Size of BM core biopsy specimens, range, cm No. of BM core biopsy specimens, bilateral/single Reason for bone marrow biopsy Lymphoma staging Cytopenias AML/MDS Primary myelofibrosis Plasma cell neoplasms Primary amyloidosis BPDCN Hepatitis C with cryoglobulinemia T-cell malignancy
41 73 (30-87) 17/24 10 0.6-2.9 7/34 12 9 8 1 6 1 1 1 2
AML/MDS, acute myeloid leukemia/myelodysplastic syndrome; BM, bone marrow; BPDCN, blastic plasmacytoid dendritic cell neoplasm. a Values are presented as numbers unless otherwise indicated.
(91%) of 441 cases, the MBP was consistent with a previously diagnosed lymphoma or comprised more than 5% of the total events; therefore, these cases were not included in this study. In 41 (9.3%) of 441 cases, the MBP comprised 5% or less of total cellular events (range, 0.1%-5%; median, 1%) and represented 3% to 100% of the total CD19+ B cells. These 41 cases are included in this study. All showed either k (25) or l (15) restriction except one case that was negative for both surface and cytoplasmic immunoglobulin light chains. The small MBPs were present in cases without a history of lymphoma or had a different phenotype from the previously diagnosed lymphoma. Thirteen (31.7%) of 41 cases had a phenotype (CD5+, dim/– SIg, CD10–, CD20 dim+, CD23+, FMC7 dim+/–, CD79b dim+/–) that resembled CLL (Table 3). Four other cases had a CD5+ phenotype that was not typical of CLL; two were CD23– and two were CD23+, CD79b+, and FMC7+. The MBPs in most cases (24/41 [58.5%]) were CD5–. One of the CD5– cases was CD10+. The immunophenotype of the MBP for each case, indications for bone marrow biopsy, clone size, and other clinical and histologic data are summarized in Table 3. The MBP in the nine bone marrow biopsies performed to evaluate cytopenias showed a CD5+ CLL-like immunophenotype in two cases and CD5– immunophenotype in seven cases. Four of the eight bone marrow biopsies performed to evaluate AML/MDS had an MBP with a CD5+, CLL-like immunophenotype; one had a CD5+ immunophenotype not typical of CLL; and three were CD5–. The single bone marrow involved by primary myelofibrosis had a CD5+ MBP with a CLL-like immunophenotype. The three cases with a history of PCM all had CD5–, CD10–, and l light chain–restricted small MBP. The MBP light chain (l) differed from the PCM in two cases. In the remaining case, both the myeloma cells and the MBP were l
689
Am J Clin Pathol 2014;141:687-696 DOI: 10.1309/AJCPD5TOBDJU4GKO
689 689
Nelson et al / Bone Marrow Biopsy Specimens With Monoclonal B-Cell Populations
❚Table 3❚ Clinical and Immunophenotypic Findings of All Cases Clone Size, Lymphoma Case Age, y/ % of Total Biopsy Organomegaly/ History No. Sex MBP Phenotype (% of B Cells) Indication Mass Lesion and Phenotype
LA in BM T Cells and/or B Cells
CD5+, CLL-like phenotype 1 72/M S&CIg–, dim CD20+, CD5+, 1 (40) ↓ Platelets No None CD10–, CD23+, dim CD79b+ 2 78/M k+, CD5+, CD10–, CD23+, 0.2 (22) ↓ Platelets No None dim FMC7+, dim CD79b+ 3 62/M Dim k+, dim CD20+, CD5+, CD10– 1 (70) Primary mye- No None lofibrosis 4 82/M l+, CD20+, CD5+, CD10– 2 (90) AML ↑ Spleen None 5 61/F k+, CD20+, CD5+, CD10–, 0.2 (15) MDS No None CD23+, dim CD79b+, FMC7– 6 76/M k+, CD20+, CD5+, CD10–, CD23+, 1 (45) MDS No NA CD79b–, FMC7– 7 73/M k+, CD20+, CD5+, CD10–, dim 1 (15) AML No None CD23+, dim CD79b+, dim FMC+ 8 62/M CD5+, CD10–, CD20+, CD23+ k, 1 (22) DLBCL No CD5–, CD10–, CD19+, CD22+, FMC7–, CD79b– bcl–6+, MUM-1+, l 9 70/F k, CD5+, CD10–, CD23+, CD19+, 1 (27) DLBCL No CD5–, CD20+ FMC7–, CD20+, dim CD22+, CD79b– 10 75/M Dim k, CD5+, CD10–, CD23+, 1 (97) FL No k, CD10+, CD5–, dim CD20+,CD22+, CD25–, CD20+, CD22+, FMC7–, CD79b– CD25–, CD52+ 11 69/M CD19+, CD20+, CD22+, k+, CD5+, 5 (79) FL No IHC: CD20+, CD5– CD23+, CD10–, FMC7–, CD79b– 12 63/M k+, CD20+, CD5+, CD10–, CD23+, 1 (50) MG No IgM K FMC7–, dim CD79b+ 13 59/M l+, CD20+, CD5+, CD10–, CD23+, 0.1 (10) MCL No CD20+, CD5+, cyclin D1– CD10–, CD23–, BCL-6–, cyclin D1+ CD5– phenotype 14 87/F k+, CD20+, sCD22+, CD5–, CD10– 2 (60) ↓ Platelets No None 15 52/M k+, CD20+, CD5–, CD10–, CD23+, 5 (100) Anemia No None CD22+, FMC7+, CD79b– 16 30/M l+, CD20+, CD5–, CD10– 1 lymphoid aggregate
13 (31.7) 6 1 1 4 6
