Antibodies to CD10, the common acute lymphoblastic

Transcription

1 Immunoperoxidase Detection of CD10 in Precursor T-Lymphoblastic Lymphoma/Leukemia A Clinicopathologic Study of 24 Cases Daniel A. Conde-Sterling, MD; Nadine S. I. Aguilera, MD; Meenakshi A. Nandedkar, MD; Susan L. Abbondanzo, MD Context. CD10 was originally reported in non T-cell lymphoblastic lymphomas/leukemias. It has since been identified, however, in a minority of cases of T-lymphoblastic lymphoma/leukemia and other hematopoietic and nonhematopoietic entities. The usual method for the detection of CD10 previously required fresh tissue. A new antibody for CD10 (56C6) in paraffin embedded tissue sections, however, has recently become available. Objective. To study the expression of CD10 in paraffin sections of T-lymphoblastic lymphoma/leukemia using monoclonal antibody 56C6. Design. Twenty-four cases of T-lymphoblastic lymphoma/leukemia in various anatomic sites were studied. Immunohistochemical analysis with CD10 and a panel of other hematolymphoid antibodies was performed in all 24 cases. Gene rearrangement studies for the T-cell receptor by the polymerase chain reaction were performed in 18 of 24 cases. Results. All cases were positive with at least 2 T-cell markers. In 15 (63%) of 24 cases CD10 was positive. T- cell receptor gene rearrangement was detected in 10 of 18 cases. Conclusions. Immunodetection of CD10 in T-lymphoblastic lymphoma/leukemia using monoclonal antibody 56C6 is common. This finding is useful in the evaluation of T-cell neoplasms. (Arch Pathol Lab Med. 2000;124: ) Antibodies to CD10, the common acute lymphoblastic leukemia antigen (CALLA), recognize a single 100- kd glycosylated polypeptide, 1 which is identical to that of human neutral endopeptidase. 2 This antigen was originally believed to be present only in acute lymphoblastic leukemia (ALL) of the common non-b, non-t and B-cell types. 1 Later studies, however, demonstrated that CD10 expression is not restricted to ALL cells and is not lineage specific within ALL. Immunodetection of CD10 has been identified in chronic myelogenous leukemia in lymphoid blast crisis, 3 follicle center lymphoma, 4 Burkitt lymphoma, 5 and nonhematopoietic tumors. 6,7 CD10 is expressed in lymphoid progenitor cells and has been found to play a specific role in promoting early T-cell development A review of the literature shows that immunodetection of CD10 in T-lymphoblastic leukemia/lymphoma (LBL/L) using fresh tissue (flow cytometric suspensions or frozen sections) varies from 15% to 40%. 1,3,5,11 Although the absence of CD10 expression in B-ALL is reportedly associated with poor outcome, 12 its expression in T-ALL in relation to the clinical outcome is more uncertain. Accepted for publication October 11, From the Armed Forces Institute of Pathology, Department of Hematopathology, Washington, DC. The opinions or statements herein contained are the private views of the authors and are not to be construed as representing the views of the Department of the Navy, the Department of the Air Force, or the Department of Defense. Reprints: Susan L. Abbondanzo, MD, Department of Hematopathology, Armed Forces Institute of Pathology, th St NW, Bldg 54, Room G124A, Washington, DC Monoclonal antibodies VIL-A1, J-5, and BA-3 recognize CD10, 13,14 but they do not work in formalin-fixed paraffinembedded (FFPE) tissue sections. This had previously restricted the immunodetection of CD10 to frozen sections and flow cytometric cell suspensions. However, 2 recently developed monoclonal antibodies that work in FFPE tissue sections are monoclonal antibodies 56C6 and NCL-CD ,16 We used monoclonal antibody 56C6 to investigate the immunoreactivity of CD10 in 24 cases of precursor T- LBL/L and correlated this finding with the clinical presentation. MATERIAL AND METHODS Formalin-fixed, paraffin-embedded tissue sections from 24 cases of T-LBL/L were retrieved from the Hematopathology Registry of the Armed Forces Institute of Pathology. Clinical histories, hematoxylin-eosin slides, and paraffin blocks (or unstained slides) were available in all cases. In cases with concomitant tissue and bone marrow involvement, leukemia was defined as the presence of more than 25% lymphoblasts in the bone marrow or at least 10% lymphoblasts in the peripheral blood. The clinical information, including extent of disease at the time of biopsy, was obtained in most cases from the contributing pathologist requisition form. Immunohistochemical analysis was performed in all 24 cases according to standardized methods 17 for CD45RB, CD3, CD43, CD45RO, F-1 receptor, myeloperoxidase, terminal deoxynucleotidyl transferase (TdT), CD34, CD99, CD20, CD79a, and CD10 (56C6) (Table 1). A modified heat-induced epitope retrieval (HIER) procedure was performed before incubation with monoclonal antibody 56C6 for CD10 and CD79a. 18 Briefly, the sections were immersed in 1 mm EDTA buffer at a ph of 8 and heat in a 800-W microwave oven for 5 minutes at high power and for an additional 5 minutes at medium-low power. 704 Arch Pathol Lab Med Vol 124, May 2000 Detection of CD10 Conde-Sterling et al

2 Table 1. Immunohistochemistry Antibody Panel Antibody Company Dilution Antigen Retrieval* Secondary Antibody CD45RB (LCA) CD20 (L26) CD3 CD45RO(UCHL-1) ßF-1 (8A3) CD10 (56C6) Myeloperoxidase CD34 (Q Bend) CD99 (MIC2 [12E7]) TdT CD43 (MT-1) CD79a (HM57) * HIER indicates heat-induced epitope retrieval., Carpinteria, Calif Endogen, Woburn, Mass Vector, Burlingame, Calif Biotest, Denville, NJ Immunotech, Miami, Fla BioGenex, San Ramon, Calif Biotest 0 1:600 Protein digestion Protein digestion HIER HIER Rabbit Rabbit Table 2. Clinical and Immunophenotypic Results* Case Sex/Age Anatomic Site(s) Studied CD10 TdT CD34 CD99 CD79a F/6 M/20 M/15 F/22 M/38 M/37 M/16 M/39 M/26 F/3 M/15 F/18 M/71 M/6 M/6 M/30 M/16 F/18 M/27 F/67 Bone marrow Supraclavicular Axillary, peripheral blood Bone marrow, axillary Bone marrow, spleen Submandibular, cervical Mediastinal, axillary Tonsil, peripheral blood Bone marrow, cervical Inguinal Inguinal, bone marrow Mediastinal * TdT indicates terminal deoxynucleotidyl transferase;, not performed. T-cell receptor (TCR) gene rearrangement studies were performed in 18 of 24 cases. Briefly, sections were deparaffinized, digested with proteinase K, and stored at 4C before performing polymerase chain reaction (PCR) using consensus primers for VJ1, VJ2, D1J2, and D2J2 genes, according to the method of Abruzzo and colleagues. 19 The phosphorus 32 labeled PCR products were resolved by high-resolution 6% denaturing polyacrylamide sequencing gel electrophoresis and detected by autoradiography. 19 RESULTS Clinical Findings There were 18 males and 6 females with a ratio of 3:1. The ages ranged from 3 to 71 years (mean, 25 years). Seven of the 24 cases had bone marrow or peripheral blood involvement. Four of these 7 cases, which showed more than 25% lymphoblasts in the bone marrow or at least 10% lymphoblasts in the peripheral blood, were classified as leukemias. These 4 cases had clinical evidence of widespread disease, including involvement of the spleen (2/4), lung (1/4), lymph node (1/4), and skin (1/4). The anatomic sites of the tissue-based lymphoblastic lymphomas included 17 lymph nodes (12, cervical; 1, supraclavicular; 1, axillary; 2, inguinal; and 1, submandibular), 2 mediastinal masses, and 1 tonsil (Table 2). In 9 of these 20 cases with lymphoblastic lymphoma, there was localized disease at the time of diagnosis. Morphology The bone marrow aspirate smear from the cases with lymphoblastic leukemia demonstrated typical L2 morphology, characterized by large lymphoblasts with conspicuous nucleoli and moderately abundant cytoplasm. Occasional cytoplasmic vacuoles were identified. The corresponding core biopsy specimens showed hypercellularity with diffuse infiltrates of large to intermediate-sized cells with blastic nuclear chromatin. The tissue biopsy specimens from the cases with lymphoblastic lymphoma (n 20) showed architectural effacement by small to intermediate-sized cells with scanty cytoplasm and fine blastic nuclear chromatin (Figure 1). In 11 of these cases, the neoplastic cells were convoluted, whereas in 9 cases, they were of the nonconvoluted type. Arch Pathol Lab Med Vol 124, May 2000 Detection of CD10 Conde-Sterling et al 705

3 706 Arch Pathol Lab Med Vol 124, May 2000 Detection of CD10 Conde-Sterling et al

4 Figure 3. Left, Immunohistochemical staining for CD10 in lymphoblastic lymphoma (immunoperoxidase technique, original magnification 130). Right, Immunohistochemical staining for terminal deoxynucleotidyl transferase in lymphoblastic lymphoma (immunoperoxidase technique, original magnification 130). There was a high mitotic rate in all cases, ranging from 2 to 10 per high-power field. Seventeen of the 20 cases of lymphoblastic lymphoma showed capsular infiltrates and/or extension to the surrounding adipose tissue. The infiltrating malignant cells tended to arrange in a single file pattern (Figure 2). Interspersed tingible body macrophages conferred a prominent starry sky pattern in 8 of 20 cases of lymphoblastic lymphoma. Immunophenotype In each case, there was immunoreaction with 1 or more T-cell markers (CD3 in 21 of 24, CD45RO in 11 of 24, CD43 in 21 of 24, F-1 in 8 of 22), supporting the T-cell phenotype. Coexpression with the B-cell marker CD79a was seen in 2 (11%) of 18 cases. All the cases were negative for CD20 and myeloperoxidase. Immunoreactivity for CD10 was noted in 63% (15/24) of the cases in a primarily cytoplasmic pattern (Figure 3, left). There was weak immunoreaction and tissue fragmentation in cases that were poorly fixed, most likely as a result of the HIER process. The decalcification process did not affect the immunodetection with CD10 since positive reactivity was noted in decalcified bone marrow core biopsies. The CD10-positive cases included 3 of 4 cases with primary bone marrow involvement (leukemia) and 12 of 20 cases with predominantly tissue-based lymphomatous disease. All the cases that presented with localized disease were immunoreactive for CD10 (9/9), while only 3 of 10 cases with disseminated tissue-based disease at the time of diagnosis were positive for CD10. Using the Fisher exact test, we found a statistically significant difference in rates of CD10 positivity between localized and disseminated disease at presentation (P.001). TdT was positive in 21 of 24 cases (Figure 3, right), while CD34 was positive in 7 of 18 cases. CD99 was positive in 11 of 15 cases. TCR gene rearrangement was demonstrated by PCR in 10 of 18 cases studied. COMMENT The cases of lymphoblastic lymphoma/leukemia included in this study reflect the proportionally higher number Figure 1. Histologic appearance of lymphoblastic lymphoma (B-5 fixed). Intermediate-sized cells with scanty cytoplasm, finely dispersed or blastic nuclear chromatin, and high mitotic rate (hematoxylin-eosin, original magnification 130). Figure 2. Lymphoblastic lymphoma infiltrating extranodal adipose tissue with single-file arrangement of neoplastic cells (hematoxylin-eosin, original magnification 160). Arch Pathol Lab Med Vol 124, May 2000 Detection of CD10 Conde-Sterling et al 707

5 of tissue-based (rather than bone marrow) lesions that are seen in consultation at the Armed Forces Institute of Pathology. Because T-lymphoblastic lymphomas have a high incidence of bone marrow involvement, they are classified as lymphoblastic lymphoma/leukemia in the Revised European-American Classification of Lymphoid Neoplasms. 20 Since the cytologic and ultrastructural characteristics of T- lymphoblastic leukemia and lymphoma are identical, 21 they are considered to be different manifestations of the same disease entity. The differential diagnosis of lymphoblastic lymphoma includes nonhematopoietic small cell tumors of childhood. Overlapping immunohistochemical findings sometimes complicate the assessment of small round cell tumors. Seventy-three percent of our cases of T-lymphoblastic lymphoma demonstrated expression of the MIC2 gene product (CD99), the so-called Ewing sarcoma marker. Moreover, small round cell tumors have also been reported to express CD10 (CALLA). 6 Therefore, a panel of immunohistochemical markers is necessary to differentiate these entities. Our study demonstrated that CD10 is frequently expressed in precursor T-lymphoblastic lymphoma/leukemia. The expression of CD10 in lymphoblastic lymphoma does not therefore imply a B-cell lineage. In addition, CD79a (a wide-spectrum B-cell marker) has been reported in approximately 40% of T-lymphoblastic lymphomas using the monoclonal antibody JCB In our study, the coexpression of CD79a using monoclonal antibody HM57 was 11% (2/18). In these 2 cases, there was reactivity with CD79a in less than 50% of the neoplastic cells, whereas reactivity with at least 2 T-cell markers was strong and diffuse. Although the percentage and intensity of CD79a reactivity are lower in our study than in other reported series, they still show the nonspecificity of CD79a. It is important to use an antibody panel to determine phenotype or cell lineage in lymphoblastic lymphomas. Approximately 80% of lymphoblastic lymphomas are derived from precursor T cells. 23 The expression of CD10 in lymphomas of T-cell phenotype appears restricted to T- LBL/L. 15 Therefore the coexpression of a T-cell marker and CD10 is highly indicative of immature T cells, making CD10 a valuable tool in the assessment of T-cell lesions. Moreover, our results showed a higher expression of CD10 in T-LBL/L than previous flow cytometric and frozen-section studies. 3,5,11 In summary, the development of a CD10 antibody for paraffin-embedded tissue sections has facilitated the study of this important marker in T-LBL/L and has helped in the differential diagnosis of T-cell neoplasms. References 1. Greaves MF, Brown G, Rapson NT, Lister TA. Antisera to acute lympho- 343blastic leukemia cells. Clin Immunol Immunopathol. 1975;4: Foon KA, Gale RP, Todd RF. Recent advances in the immunologic classification of leukemia. Semin Hematol. 1986;23: Ritz J, Pesando JM, Notis-McConarty J, Lazarus H, Schlossman SF. A monoclonal antibody to human acute lymphoblastic leukaemia antigen. Nature. 1980; 283: Greaves MF, Hariri G, Newman RA, Sutherland DR, Ritter MA, Ritz J. Selective expression of the common acute lymphoblastic leukemia (gp 100) antigen on immature lymphoid cells and their malignant counterparts. Blood. 1983;61: Ritz J, Nadler LM, Bhan AK, Notis-McConarty J, Pesando JM, Schlossman SF. Expression of common acute lymphoblastic leukemia antigen (CALLA) by lymphomas of B-cell and T-cell lineage. Blood. 1981;58: Pilkington GR, Pallesen G. Phenotypic characterization of nonhaemopoietic small cell tumours of childhood with monoclonal antibodies to leucocytes, epithelial cells and cytoskeletal proteins. Histopathology. 1989;14: Durie BG, Grogan TM. CALLA-positive myeloma: an aggressive subtype with poor survival. Blood. 1985;66: Guerin S, Mari B, Maulon L, Belhacene N, Marguet D, Auberger P. CD10 plays a specific role in early thymic development. FASEB J. 1997;11: Tjonnfjord GE, Steen R, Veiby OP, Morkrid L, Egeland T. Haemopoietic progenitor cell differentiation: flow cytometric assessment in bone marrow and thymus. Br J Haematol. 1995; Mari B, Breittmayer JP, Guerin S, et al. High levels of functional endopeptidase (CD10) activity on human thymocytes: preferential expression on immature subsets. Immunology. 1994;82: Bernard A, Boumsell L, Reinherz EL, et al. Cell surface characterization of malignant T cells from lymphoblastic lymphoma using monoclonal antibodies: evidence for phenotypic differences between malignant T cells from patients with acute lymphoblastic leukemia and lymphoblastic lymphoma. Blood. 1981;57: Donadieu J, Auclerc MF, Baruchel A, et al, French Acute Lymphoblastic Leukaemia Study Group. Critical study of prognostic factors in childhood acute lymphoblastic leukaemia: differences in outcome are poorly explained by the most significant prognostic variables. Br J Haematol. 1998;102: Cossman J, Neckers LM, Leonard WJ, Greene WC. Polymorphonuclear neutrophils express the common acute lymphoblastic leukemia antigen. J Exp Med. 1983;157: Knapp W, Majdic O, Bettelheim P, Liszka K. VIL-A1, a monoclonal antibody reactive with common acute lymphatic leukemia cells. Leuk Res. 1982;6: Kaufmann O, Flath B, Spath-Schwalbe E, Possinger K, Dietel M. Immunohistochemical detection of CD10 with monoclonal antibody 56C6 on paraffin sections. Am J Clin Pathol. 1999;111: McIntosh GG, Lodge JA, Watson P, et al. NCL-CD10-270: a new monoclonal antibody recognizing CD10 in paraffin-embedded tissue. Am J Pathol. 1999;154: Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem. 1981;29: Shi SR, Key ME, Kalra KL. Antigen retrieval in formalin-fixed, paraffinembedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections. J Histochem Cytochem. 1991;39: Abruzzo LV, Griffith LM, Nandedkar M, et al. Histologically discordant lymphomas with B- cell and T-cell components. Am J Clin Pathol. 1997;108: Harris NL, Jaffe ES, Stein H, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood. 1994;84: Nathwani BN, Kim H, Rappaport H. Malignant lymphoma, lymphoblastic. Cancer. 1976;38: Pilozzi E, Pulford K, Jones M, et al. Co-expression of CD79a (JCB117) and CD3 by lymphoblastic lymphoma. J Pathol. 1998;186: Cossman J, Chused TM, Fisher RI, Magrath I, Bollum F, Jaffe ES. Diversity of immunological phenotypes of lymphoblastic lymphoma. Cancer Res. 1983; 43: Arch Pathol Lab Med Vol 124, May 2000 Detection of CD10 Conde-Sterling et al