Flow Cytometric Immunophenotyping in Posttransplant Lymphoproliferative Disorders

Transcription

1 Hematopathology / FLOW CYTOMETRIC IMMUNOPHENOTYPING IN POSTTRANSPLANT LYMPHOPROLIFERATIVE DISORDERS Flow Cytometric Immunophenotyping in Posttransplant Lymphoproliferative Disorders Cherie H. Dunphy, MD, 1 Laura J. Gardner, MD, 1 Leonard E. Grosso, MD, PhD, 2 and H. Lance Evans, MD 1 Key Words: Posttransplant lymphoproliferative disorders; Flow cytometry; Genotypic studies Abstract We studied the flow cytometric immunophenotyping (FCI) and genotypic data of 11 specimens from 10 transplant recipients and categorized them based on a scheme for posttransplant lymphoproliferative disorders (PTLDs). Specimens had been analyzed by polymerase chain reaction and/or Southern blot for T- cell and B-cell (immunoglobulin heavy chain and light chain genes) gene rearrangements (BGR). The categories for PTLDs were as follows: 1, 1; 2, 6; and 3, 4. The plasmacytic and polymorphic B-cell hyperplasias (PBCHs) revealed no monoclonal/aberrant cells by FCI or genotypic studies (GS). Three of 4 polymorphic B-cell lymphomas (PBCLs) revealed monoclonal or aberrant (no surface light chain) B cells by FCI; 1 of 3 revealed a BGR. However, the 1 case with no monoclonal/aberrant B cells by FCI revealed a BGR. Both immunoblastic lymphomas revealed monoclonal or aberrant B cells by FCI; 1 revealed a BGR. Both multiple myelomas revealed monoclonal plasma cells by FCI; 1 revealed a BGR. In the 4 PTLDs with monoclonal/aberrant B cells by FCI and no clonality detected by GS, the GS were performed on fresh and paraffin-embedded tissue samples. FCI of the plasmacytic and PBCHs supported no clonal process by GS. FCI defined a clonal process in 2 PBCLs, 1 immunoblastic lymphoma, and 1 multiple myeloma that were negative by GS. However, 1 PBCL that was polyclonal by FCI was monoclonal by GS. Thus, FCI is useful for identifying a clonal process in PTLDs with negative results by GS; FCI and GS should be performed routinely in PTLDs to detect a clonal process. Posttransplant lymphoproliferative disorders (PTLDs) represent a morphologic, immunophenotypic, and genotypic spectrum of disease. The spectrum ranges from polymorphic, polyclonal proliferations with many features of a florid viral infection at one end to monomorphic, monoclonal proliferations, usually of the B-cell type, at the other end. 1 Authors attempts to categorize PTLDs have been based on a variety of criteria, including morphologic features and analysis of clonality (predominantly by genotypic techniques). In 1981, Frizzera et al 2 described 2 major distinctive types of PTLD in renal transplant recipients: polymorphic diffuse B-cell hyperplasia (lacking necrosis and atypical immunoblasts ) and polymorphic diffuse B-cell lymphoma. In 1988, Nalesnik et al 3 described the pathology of PTLDs occurring in the setting of cyclosporine-prednisone immunosuppression. Based on morphologic features, they divided PTLDs into 3 categories: (1) polymorphic (necrosis is most pronounced with atypical immunoblasts most frequently seen), (2) monomorphic (resembles Burkitt or centroblastic lymphoma; indistinguishable from non-hodgkin lymphoma), and (3) minimal polymorphism (predominance of plasmacytoid lymphocytes and plasma cells). More recently, in 1995, Knowles et al 4 proposed dividing PTLDs into 3 distinct categories based on a combination of morphologic and genotypic studies. These categories are as follows: 1. Plasmacytic hyperplasia characteristically involves the oropharynx or nodal sites, is nearly always polyclonal, usually contains multiple Epstein-Barr virus (EBV) infection events, and lacks oncogene and tumor suppressor gene alterations. 2. Polymorphic B-cell hyperplasia and polymorphic B- cell lymphoma (PBCL) may arise nodally or extranodally, are 24 Am J Clin Pathol 2002;117:24-28 American Society for Clinical Pathology

2 Hematopathology / ORIGINAL ARTICLE nearly always monoclonal, usually contain a single form of EBV, and lack oncogene and tumor suppressor gene alterations. 3. Immunoblastic lymphoma and multiple myeloma represent widely disseminated disease of monoclonal origin containing a single form of EBV and alterations of 1 or more oncogene or tumor suppressor genes (ie, N-ras, p53, c-myc). In this diagnostic scheme, determination of clonality is thus important in PTLDs to categorize the process and to manage the treatment of the patient. As mentioned, analysis of clonality has been determined predominantly by genotypic techniques. Analysis of clonality in PTLDs by immunophenotyping has primarily consisted of immunohistochemical studies that have usually yielded indeterminate results. 3 Determination of clonality by flow cytometric immunophenotyping (FCI) in excised biopsy specimens of PTLD has been described by Kowal-Vern et al. 5 However, a complete immunophenotypic profile by FCI was performed in only 1 case; only 4 cases underwent FCI analysis of B-cell markers (CD19 and CD20). Thus, the purposes of the present study were to describe the morphologic, immunohistochemical, and FCI findings based on a comprehensive analysis of FCI markers in PTLDs and systematically compare the results with those of genotypic studies. Materials and Methods Tissue Specimens The files of the Division of Hematopathology, St Louis University Health Sciences Center, St Louis, MO, were reviewed from July 1993 to December 1999 for cases diagnosed as PTLD. Eleven specimens (2 nodal, 9 extranodal) from 10 patients (8 adults, 2 children; recipients of 6 heart, 3 liver, and 1 kidney transplant) were identified. The tissue specimens were reviewed morphologically with the corresponding immunohistochemical stains (6 specimens), FCI data (all specimens), and results of genotypic studies (all specimens). The PTLDs were categorized based on the scheme of Knowles et al 4 as outlined. The results of analyses on 3 of the cases (2 PBCL and 1 immunoblastic lymphoma) have been described. 6 Immunohistochemical Analysis The following immunohistochemical stains were variably performed on formalin-fixed (3 specimens) and/or B- 5 fixed (4 specimens) tissue on the DAKO Autostainer (DAKO, Carpinteria, CA) universal staining system using the DAKO labeled streptavidin biotin horseradish peroxidase system, DAKO 3,3'-diaminobenzidine substrate, and the following antibodies: CD3, CD45RO (UCHL-1), CD20 (L- 26), CD45 (LCA), CD68 (KP-1), kappa, epithelial membrane antigen, and bcl-2 (DAKO); CD43 (MT-1), lambda, and MB-2 (Biogenex, San Ramon, CA); CD30 (Ber-H2) (Novacastra, Vector Laboratories, Burlingame, CA); and CD15 (Leu-M1) (Becton Dickinson, San Jose, CA). The following immunohistochemical stains were performed using antigen retrieval with Citra Buffer Retrieval Solution (DAKO): CD3, CD68, CD23, bcl-2, and CD30. Antigen retrieval was not used with the remaining immunohistochemical stains. Control tissue samples were tonsil for CD3, CD45RO, CD20, CD45, CD68, CD23, kappa, bcl-2, CD43, lambda, and MB-2; carcinoma for epithelial membrane antigen; anaplastic CD30+ large cell lymphoma for CD30; and Hodgkin lymphoma for CD15. Flow Cytometric Immunophenotyping Touch imprints of a portion of each tissue sample submitted for flow cytometry were prepared and Wright stained. The tissue sample then was suspended into RPMI 1640 medium (Mediatech Cellgro Tissue Culture Media, Fisher-Scientific, Pittsburgh, PA). The single cell suspension was then analyzed on a FACScan (Becton Dickinson) or an Ortho Cytoronabsolute (Ortho Diagnostic Systems, Raritan, NJ) flow cytometer for various antigens, using standard techniques and the following commercially available monoclonal antibodies: CD1, CD4, CD5, CD8, CD10, and HLA-DR (Ortho); CD2, CD13, CD14, CD19, CD24, and CD56 (Coulter Clone, Coulter Immunology, Hialeah, FL); CD3, CD7, and CD20 (Becton Dickinson); CD23, CD25, and CD30 (DAKO); CD45 (Caltag, Burlingame, CA); IgA, IgD, IgG, IgM, and kappa and lambda light chains (Kallestad, Chaska, MN); and CD138 (Biotest, Denville, NJ). Dual staining was performed as follows: CD3/4, CD8/56, CD19/5, CD20/HLA-DR, CD45/10, CD14/23, CD13/14, CD2/24, and kappa/lambda. CD14/23 were dually stained because CD23 may nonspecifically stick to monocytes. The remaining antibodies were analyzed individually. Regions analyzed included lymphocyte and large cell regions that were gated based on their forward and side light-scatter properties. The antigens analyzed in a particular case were based on cytomorphologic examination of the touch imprints. The kappa/lambda monoclonality was determined by determining the percentage of cells staining only with kappa vs the percentage of cells staining only with lambda. Monoclonality was considered present if the kappa/lambda ratio was greater than 3:1 or less than 1:2 as previously defined. 7-9 An aberrant T-cell immunophenotype was defined as loss of expression of one of the T-cell antigens (ie, CD2, CD3, CD5, or CD7) and/or coexpression of CD4 and CD8 and/or CD1. An aberrant B-cell immunophenotype was defined as American Society for Clinical Pathology Am J Clin Pathol 2002;117:

3 Dunphy et al / FLOW CYTOMETRIC IMMUNOPHENOTYPING IN POSTTRANSPLANT LYMPHOPROLIFERATIVE DISORDERS Table 1 Diagnostic Categories of PTLDs With Corresponding Data From Immunohistochemical, FCI, and Genotypic Studies Clonality Immunohistochemical Data * FCI Data Genotypic Data Category 1 Plasmacytic hyperplasia Negative with L-26,MB-2, MT-1, Polyclonal Polyclonal UCHL-1, LCA, kappa, lambda, CD30 Category 2 Polymorphic B-cell hyperplasia ND Polyclonal Polyclonal Polymorphic B-cell hyperplasia ND Polyclonal Polyclonal Polymorphic B-cell lymphoma ND Monoclonal Monoclonal Polymorphic B-cell lymphoma L- 26+ Monoclonal Polyclonal Polymorphic B-cell lymphoma L- 26+, bcl-2+, faint CD3+, subset Aberrant B cells Polyclonal monoclonal cytoplasmic kappa, (no surface light chains) CD30+, EMA /+, LMP /+, Leu-M1 Polymorphic B-cell lymphoma Negative with L-26, MB-2, UCHL-1, Polyclonal Monoclonal MT-1, CD3, Leu-M1, LCA, KP-1; CD30+, EMA /+ Category 3 Immunoblastic lymphoma ND Monoclonal Monoclonal Immunoblastic lymphoma L-26+, CD3, LCA Aberrant B cells Polyclonal (no surface light chains) Multiple myeloma ND Monoclonal Monoclonal Multiple myeloma Monoclonal cytoplasmic lambda Monoclonal Polyclonal EMA, epithelial membrane antigen; FCI, flow cytometric immunophenotyping; LCA, leukocyte common antigen; LMP, latent membrane protein; ND, not done; PTLDs, posttransplant lymphoproliferative disorders; +, positive;, negative. * See the Materials and Methods section for further description and sources of the antibodies. loss of or no associated expression of surface light chains in a B-cell proliferation. The diagnostic interpretations were based on numeric values (ie, the percentage of cells positive for each antibody used). Graphic displays also were reviewed for pattern recognition of cell populations (eg, cells dually staining with CD19/CD5, cells selectively staining with kappa vs lambda, intensity of expression of surface antigens). Genotypic Studies All specimens (9 fresh and 2 paraffin-embedded tissue samples) were analyzed by polymerase chain reaction (PCR; 8 specimens), Southern blot analysis (5 specimens), or both for B-cell with or without analysis of T-cell gene rearrangements. DNA was recovered by standard techniques previously reported. 10,11 Southern blot analysis using probes for the joining region of the immunoglobulin heavy chain gene, constant regions of both kappa and lambda light chains, and T-cell receptor-beta (TCR-beta) were performed as previously described. 10 PCR to detect a monoclonal immunoglobulin heavy chain and TCR-beta rearrangements were performed as previously described. 6 The limits of detection of the Southern blot analysis and PCR-based assays were 5% and 1%, respectively. The PCR analysis evaluated the presence or absence of a monoclonal rearrangement of the immunoglobulin heavy chain in all 8 cases and a monoclonal rearrangement of the TCR-beta gene in 6 of 8 specimens. Southern blot analysis evaluated the presence or absence of a monoclonal rearrangement of the immunoglobulin heavy chain gene in 4 of 5 specimens and monoclonal rearrangements of the light chain (kappa and lambda) genes and the TCR-beta gene in all 5 specimens. Results Tissue Specimens The PTLDs were categorized as outlined in Table 1. Immunohistochemical Analysis The immunohistochemical staining results are outlined in Table 1. In the 6 specimens in which immunohistochemical stains were performed, they contributed substantially to the diagnosis in 4 specimens. In 2 cases, they supported the B-cell origin of the PTLD (1 immunoblastic lymphoma with aberrant B-cells by FCI and no B-cell gene rearrangement by genotypic analysis and 1 PBCL with monoclonal B cells by FCI and no B-cell gene rearrangement by genotypic analysis). In the other 2 cases, the immunohistochemical stains revealed monoclonal cytoplasmic light chain (1 multiple myeloma with monoclonal plasma cells by FCI and no B-cell gene rearrangement by genotypic analysis and 1 PBCL with aberrant B cells by FCI and no B-cell gene 26 Am J Clin Pathol 2002;117:24-28 American Society for Clinical Pathology

4 Hematopathology / ORIGINAL ARTICLE Image 1 A polymorphic B-cell lymphoma, which was negative by flow cytometric immunophenotyping and revealed a B-cell gene rearrangement, is composed morphologically of large atypical lymphocytes with nucleoli in a background of cellular debris and necrosis (H&E, 400). rearrangement by genotypic analysis). However, there were no specimens in which the immunohistochemical stain results independently identified the cellular origin of the PTLD and established monoclonality. FCI and Genotypic Studies The FCI data and genotypic results are outlined in Table 1 with the corresponding diagnostic categories. In summary, FCI of the plasmacytic hyperplasia and polymorphic B-cell hyperplasias identified no clonal process. These cases were polyclonal by genotypic studies performed by Southern blot analysis of fresh tissue samples. The FCI data identified a clonal proliferation in 2 polymorphic B-cell lymphomas, 1 immunoblastic lymphoma, and 1 multiple myeloma Image 1 that were negative by genotypic studies. The genotypic studies in these negative cases all were performed by PCR analysis in fresh tissue samples (1 PBCL and 1 multiple myeloma). However, 1 PBCL that was polyclonal by FCI was monoclonal by genotypic analysis by PCR and Southern blot analysis of a fresh tissue sample (Image 1). Discussion PTLDs develop secondary to the immunosuppression treatment given to organ transplant recipients and seem related to the degree of immunosuppression. Clinical management of PTLD generally has been based on the results of clonality studies. The cases that are polyclonal by all methods tend to regress with reduced immunosuppression. Thus, lack of monoclonality by genotypic analysis generally has accurately predicted the cases that will respond to reduced immunosuppression. 12 The cases with clonality may progress (ie, more intense rearranged bands) or regress (ie, weaker rearranged bands). 1 Response to reduced immunosuppression usually is observed within 2 weeks and may be seen within a few days. 12 While determining clonality in these cases is of clinical importance, it may be determined by various modalities. While immunohistochemical studies identify the lineage of the proliferation, review of the literature demonstrates that immunohistochemical studies of clonality in PTLDs most often have yielded indeterminate results. 3 Most genotypic studies have used Southern blot analysis with probes for the immunoglobulin heavy and light chain genes. Routine Southern blot analysis has demonstrated monoclonal B-cell populations in almost all cases with immunophenotypically monoclonal B cells and in one third of all immunophenotypically polyclonal or indeterminate cases. 4,13-15 The reason for this finding is that genotypic analysis can detect minor monoclonal populations within an extensive polyclonal background, as well as monoclonal cells with no surface expression of kappa or lambda. 13,16,17 Thus, genotypic studies are clearly preferable to immunohistochemical immunophenotypic studies in demonstrating monoclonality in PTLDs. However, to our knowledge, comprehensive flow cytometric immunophenotyping has not been described and systematically compared with genotypic analysis results in such a large study. We compared such results in 11 specimens and also compared FCI data and immunohistochemical stain results in 6 available specimens. Our study demonstrated that plasmacytic and polyclonal B-cell hyperplasias were polyclonal by FCI and genotypic data and that polymorphic B-cell and immunoblastic lymphomas, as well as multiple myeloma, were monoclonal by FCI and/or genotypic data. In addition, our analysis demonstrated the usefulness of FCI in identifying a clonal process in 4 cases of PTLD with negative genotypic study results. The genotypic analysis in all of these negative cases was performed by PCR (2 fresh tissue; 2 paraffinembedded tissue). Thus, our study indicates a combined FCI and genotypic analytic approach is useful for detecting a clonal process in PTLDs. Five cases of PTLD had discrepancies in the analysis of monoclonality by genotyping and FCI. Presumably because of detection of a minor population of monoclonal cells by genotypic analysis, 1 PTLD was polyclonal by FCI but revealed a B-cell gene rearrangement by genotypic analysis. American Society for Clinical Pathology Am J Clin Pathol 2002;117:

5 Dunphy et al / FLOW CYTOMETRIC IMMUNOPHENOTYPING IN POSTTRANSPLANT LYMPHOPROLIFERATIVE DISORDERS Four cases that were monoclonal or revealed an aberrant B- cell population by FCI without a genotypically detectable monoclonal rearrangement present an interesting dilemma. While enhancing the limit of detection and allowing for the use of smaller tissue specimens, as well as formalin-fixed, paraffin-embedded tissue in genotypic studies, the falsenegative rate of PCR-based assays must be discussed. As previously documented, the PCR method used to assess immunoglobulin gene rearrangement will detect approximately 70% to 80% of all B-lineage neoplasms compared with Southern blot analysis. However, the rate of detection varies greatly with the type of neoplasm; chronic lymphocytic leukemia/small lymphocytic lymphoma and B-cell acute lymphoblastic leukemia have the highest detection rate. Explanations, including deletion or alteration of the oligonucleotide primer binding sites, have been proposed to account for the differing detection rates Comparisons of immunohistochemical stain results and FCI data revealed that the immunohistochemical stains supported the findings by FCI. Of interest, in 1 PBCL, CD20 was negative by FCI but detected by immunohistochemical staining. Thus, the role of immunohistochemical stains seems to be selective; immunohistochemical stains may be helpful in cases with equivocal or negative results on FCI and genotypic studies. From the Divisions of 1 Hematopathology and 2 Molecular Pathology, Department of Pathology, St Louis University Health Sciences Center, St Louis, MO. Address reprint requests to Dr Dunphy: Dept of Pathology and Laboratory Medicine, University of North Carolina, CB# 7525, Chapel Hill, NC Acknowledgments: We acknowledge Patty Jenkins and Linda Sheahan for secretarial assistance. References 1. Swerdlow SH. Post-transplant lymphoproliferative disorders: a morphologic, phenotypic, and genotypic spectrum of disease. Histopathology. 1992;20: Frizzera G, Hanto DW, Gajl-Peczalska KJ, et al. Polymorphic diffuse B-cell hyperplasias and lymphomas in renal transplant recipients. Cancer Res. 1981;41: Nalesnik MA, Jaffe R, Starzl TE, et al. The pathology of posttransplant lymphoproliferative disorders occurring in the setting of cyclosporine A prednisone immunosuppression. Am J Pathol. 1988;133: Knowles DM, Cesarman E, Chadburn A, et al. Correlative morphologic and molecular genetic analysis demonstrates three distinct categories of post-transplantation lymphoproliferative disorders. Blood. 1995;85: Kowal-Vern A, Swinnen L, Pyle J, et al. Characterization of post-cardiac transplant lymphomas: histology, immunophenotyping, immunohistochemistry, and gene rearrangement. Arch Pathol Lab Med. 1996;120: Collins BT, Ramos RR, Grosso LE. Combined fine needle aspiration biopsy, and immunophenotypic and genotypic approach to posttransplantation lymphoproliferative disorders. Acta Cytol. 1998;42: Segal GH, Edinger MG, Owen M, et al. Concomitant delineation of surface Ig, B-cell differentiation antigens, and HLA-DR on lymphoid proliferations using three-color immunocytochemistry. Cytometry. 1991;12: Kelly-Williams S, Zmijewski CM, Tomaszewski JE. Lymphocyte subpopulations in normal lymph nodes harvested from cadavers. Lab Med. 1989;7: Witzig TE, Banks PM, Stenson MJ, et al. Rapid immunophenotyping of B cell non-hodgkin s lymphoma by flow cytometry: a comparison with the standard frozen-section method. Am J Clin Pathol. 1990;94: Havlioglu NH, Manepalli A, Galindo L, et al. Primary Ki-1 (anaplastic large cell) lymphoma of the brain and spinal cord. Am J Clin Pathol. 1995;103: Dunphy CH, Grosso LE, Rodriguez JJ, et al. Bilateral mucosaassociated lymphoid tissue lymphomas of parotid glands: a 13-year interval. Mod Pathol. 1996;9: Nalesnik MA, Makowka L, Starzl TE. The diagnosis and treatment of post-transplant lymphoproliferative disorders. Curr Probl Surg. 1988;25: Zutter MM, Martin PJ, Sale GE, et al. Epstein-Barr virus lymphoproliferation after bone marrow transplantation. Blood. 1988;72: Griffith RC, Saha BK, Janney CM, et al. Immunoblastic lymphoma of T-cell type in a chronically immunosuppressed renal transplant recipient. Am J Clin Pathol. 1990;93: Shapiro RS, McClain K, Frizzera G, et al. Epstein-Barr virus associated B cell lymphoproliferative disorders following bone marrow transplantation. Blood. 1988;71: Cleary ML, Warnke R, Sklar J. Monoclonality of lymphoproliferative lesions in cardiac transplant recipients: clonal analysis based on immunoglobulin-gene rearrangements. N Engl J Med. 1984;310: Hanto DW, Birkenbach M, Frizzera G, et al. Confirmation of the heterogeneity of post-transplant Epstein-Barr virus associated B cell proliferations by immunoglobulin gene rearrangement analyses. Transplantation. 1989;47: Lehman CM, Sarago C, Nasim S, et al. Comparison of PCR with Southern hybridization for the routine detection of immunoglobulin heavy chain gene rearrangements. Am J Clin Pathol. 1995;103: Segal GH, Jorgensen T, Masih AS, et al. Optimal primer reaction analysis, I: low-grade B-cell lymphoproliferative disorders of non-follicular origin. Hum Pathol. 1994;25: Segal GH, Jorgensen T, Scott M, et al. Optimal primer reaction analysis, II: follicular lymphomas. Hum Pathol. 1994;25: Lombardo JF, Hwang TS, Maiese RL, et al. Optimal primer reaction analysis, III: intermediate and high-grade B-cell neoplasms. Hum Pathol. 1996;27: Segal GH, Wittwer CT, Fishleder AJ, et al. Identification of monoclonal B-cell populations by rapid cycle polymerase chain reaction: a practical screening method for the detection of immunoglobulin gene rearrangements. Am J Pathol. 1992;14: Am J Clin Pathol 2002;117:24-28 American Society for Clinical Pathology