Flow Cytometric Immunophenotyping of Adult T-Cell Leukemia/Lymphoma Using CD3 Gating

Hematopathology / ATLL IMMUNOPHENOTYPING USING CD3 GATING Flow Cytometric Immunophenotyping of Adult T-Cell Leukemia/Lymphoma Using CD3 Gating Taiji Yokote, MD, 1 Toshikazu Akioka, MD, 1 Satoko Oka, MD, 1 Satoshi Hara, MD, 1 Kichinosuke Kobayashi, MD, 1 Hideto Nakajima, MD, 1 Takeshi Yamano, MD, 1 Toshiyuki Ikemoto, PhD, 3 Akira Shimizu, MD, 3 Motomu Tsuji, MD, 2 and Toshiaki Hanafusa, MD 1 Key Words: Adult T-cell leukemia/lymphoma; Flow cytometry; Immunophenotyping; CD3 gating DOI: 10.1309/KEN4MXM5Y9A1GEMP Abstract Adult T-cell leukemia/lymphoma (ATLL) is a lymphoproliferative neoplasm of helper T lymphocytes caused by human T-cell leukemia virus type-1 (HTLV- 1). The disease was first described in Kyushu, in southwestern Japan, and most frequently occurs in endemic areas, such as Japan, the Caribbean basin, West Africa, Brazil, and northern Iran. ATLL is essentially a disease of adults, characterized clinically by generalized lymphadenopathy, hepatosplenomegaly, skin lesions, and hypercalcemia. The prognosis of most patients is quite poor, with a median survival time of only 13 months, even if multiagent combination chemotherapy is given. In the present study, flow cytometric immunophenotyping with CD3 gating was performed on 30 samples from 26 patients who had been given a diagnosis of ATLL. The records of these patients also were reviewed retrospectively. In 14 of the 30 samples, an abnormal CD3 low T-cell population was distinguishable from the normal T-cell populations by flow cytometric analysis. Herein we report a novel strategy for flow cytometric immunophenotyping of ATLL facilitated by CD3 low gating. Materials and Methods Case and Control Samples We retrospectively reviewed all cases of ATLL diagnosed at the Osaka Medical College, Osaka, Japan, from January 1998 to January 2004. Diagnosis of ATLL was based on the presence of an antibody positive for anti HTLV-1, histologic identification of lymphoproliferative disease of peripheral T-cell origin, and confirmation of monoclonal integration of the HTLV-1 proviral genome by Southern blot. This analysis yielded 30 samples from 26 patients, consisting of 20 peripheral blood samples, 3 bone marrow aspirate samples, 5 lymph node samples, 1 pleural effusion sample, and 1 bronchoalveolar lavage fluid (BAL) sample. The 26 patients (14 men and 12 women) had a mean ± SD age of 60.3 ± 11.8 years. Because ATLL is a clinically heterogeneous disease, the cases were classified in 4 subtypes 1 : acute-type ATLL, 19 (73%); chronic-type ATLL, 1 (4%); lymphoma-type ATLL, 3 (12%); and smoldering ATLL, 3 (12%). All cases exhibited 5% or more abnormal ATLL cells in each specimen, including cells with hyperlobated nuclei (flower cells) or irregular nucleolar contours. For comparison of CD3 fluorescence intensity of normal T cells with that of ATLL cells, 15 normal peripheral blood samples obtained from healthy volunteers were studied. The mean ± SD age of the healthy volunteers was 41.5 ± 10 years (range, 26-58 years). Flow Cytometry Sample Preparation and FACS Acquisition For preparation of the cell suspension, lymph node samples were minced with scissors and filtered through a 50-µm Am J Clin Pathol 2005;124:199-204 199 199 DOI: 10.1309/KEN4MXM5Y9A1GEMP 199

Yokote et al / ATLL IMMUNOPHENOTYPING USING CD3 GATING mesh nylon filter. Peripheral blood samples, bone marrow aspirates, and pleural effusion and BAL samples were treated with EDTA or heparin for anticoagulation and processed using a lysed whole blood technique using fluorescence-activated cell sorting (FACS) lysing solution (Whole Blood Lysis Reagent, Coulter, Fullerton, CA). All samples (regardless of origin) were washed twice in phosphate-buffered saline (PBS) and resuspended in PBS at a concentration of approximately 2 10 6 cells per milliliter. Samples were stained dually with fluorescein isothiocyanate (FITC)-conjugated CD2 (MT910, DAKO, Carpinteria, CA), CD4 (T4, Coulter), CD28 (COLT2, Nichirei, Tokyo, Japan), and phycoerythrin-cyanine 5.1 (PC5)-conjugated CD3 (UCHT1, Coulter) for 30 minutes at 4 C. Negative control samples were stained with FITC-conjugated rat antihuman immunoglobulin (LODNP1, Coulter) for 30 minutes at 4 C. Case samples also were stained dually with phycoerythrinconjugated CD5 (UCHT-2, Pharmingen, San Diego, CA), CD7 (My7, Coulter), CD8 (DK25, DAKO), CD25 (2A3, Becton Dickinson, San Jose, CA), CD45RO (UCHL-1, Becton Dickinson), CD45RA (2H4, Coulter), CD62L (TQ1, Coulter), HLA-DR (L243, Pharmingen), and PC5-conjugated CD3 (UCHT1, Coulter) for 30 minutes at 4 C. Negative control samples were stained with phycoerythrin-conjugated rat antihuman immunoglobulin (LODNP1, Coulter) for 30 minutes at 4 C Image 1. All samples were rinsed twice in PBS and analyzed on a FACScan flow cytometer (EPICS XL, Beckman Coulter, Fullerton, CA) using System II software, version 3.0 (Beckman Coulter). At least 10,000 cells in total were analyzed. FACS Analysis A CD3 gating strategy was used to identify ATLL cells. All samples were analyzed in the side scatter (SS) vs CD3 histogram mode. In the abnormal cell samples, CD3 peak fluorescence intensity below 10 on the log scale was defined as CD3 low, an A 1,000 CD3-PC5 A C 0.1 0 1,024 SS B 1,000 0.1 0 1,024 SS Image 1 Representation of flow cytometric histograms for adult T-cell leukemia/lymphoma samples. A, Bivariate histogram CD3/side scatter (SS). Gate A shows the CD3 medium population. B, Bivariate histogram CD3/SS. Gate A shows the CD3 low population. PC5, phycoerythrin-cyanine 5.1. B CD3-PC5 D A C B intensity between 10 and 100 as CD3 medium, and an intensity more than 100 on the log scale as CD3 high. The cells in the CD3 low and CD3 medium population regions were acquired (gate A), and this procedure was defined as CD3 low and CD3 medium gating, respectively. Cells with overlap of less than 50% with the isotype control or with more than a 30% positive distinct population above the isotype control were defined as positive. The expression levels of T cell associated antigens (CD2, CD3, CD4, CD5, CD7, CD28, CD45RA, CD45RO, and CD62L), activated antigen (CD25), and HLA-DR in positive cells were evaluated. Immunohistochemical Analysis Lymph node slides were obtained from tissue samples fixed with 10% buffered formalin and embedded in paraffin. Peripheral blood samples and bone marrow aspirates were smeared manually. BAL samples were processed by application of Cytospin (Shandon, Pittsburgh, PA) to the fluid. Immunohistochemical analysis of the peripheral blood, bone marrow aspirate, lymph node, and BAL slides was performed using CD7 (titrated at 1:20; DK24, DAKO), CD4 (titrated at 1:20; MT310, DAKO), and CD8 (titrated at 1:20; DK25, DAKO) monoclonal antibodies, using a modified avidin-biotinperoxidase procedure with 3,3'-diaminobenzidine (Sigma Chemical, St Louis, MO) as the chromogen. A negative control experiment was performed for each sample using normal rabbit immunoglobulin fraction (DAKO), and a positive control experiment was performed using lymphocytes from peripheral blood smears from healthy volunteers, using the CD7, CD4, and CD8 monoclonal antibodies as described. The percentage of abnormal cells, such as ATLL cells, was determined from the CD7, CD4, and CD8 immunohistochemical analysis. Statistical Analysis Associations between the gating procedure and CD3 fluorescence intensity were tested using a Kruskal-Wallis test and a Bonferroni-corrected Mann-Whitney U test for unrelated variables. The mean values and their standard deviations, the median, 25th and 75th percentiles, and range were calculated for each variable. P values of.017 or less were considered statistically significant. Associations between the gating procedure and immunophenotyping also were tested using a Fisher exact test, as determined by expected frequencies of variables in 2 2 tables. A P value of less than.05 was used as the criterion for statistical significance. Statistical analyses were performed using SPSS 12J software (SPSS, Tokyo, Japan). Results CD3 Populations CD3 expression was positive in abnormal cells, but the CD3 fluorescence intensity showed 2 patterns. Of the 30 samples 200 Am J Clin Pathol 2005;124:199-204 200 DOI: 10.1309/KEN4MXM5Y9A1GEMP

Hematopathology / ORIGINAL ARTICLE from 26 patients, 14 samples contained a CD3 low population and 16 contained a CD3 medium population. Figure 1 shows the fluorescence intensity for the CD3 low (3.94 ± 2.6) and CD3 medium (26 ± 12) populations and for the normal T-cell population from healthy volunteers (15.6 ± 2.8, control population). Statistically significant differences in the mean fluorescence intensity were observed between the CD3 low population and either the CD3 medium or control population (P <.001). The mean fluorescence intensity of the CD3 low population was significantly lower than that of the control population (P <.001), but no statistically significant difference was observed between the mean fluorescence intensities of the control and CD3 medium populations (P =.037). CD3 Gating The histogram containing the CD3 medium population was separated into 3 categories: CD3 medium population (gate A), granulomonocytes (gate B), and mature B lymphocytes and precursor cells (gate C). The CD3 medium population showed the highest CD3 fluorescence intensity and a low SS signal. Granulomonocytes showed negative CD3 fluorescence intensity and the highest SS signal, and the mature B lymphocytes and precursors showed negative CD3 fluorescence intensity and a low SS signal. Hence, the CD3 medium population was easily distinguishable from granulomonocytes and mature B lymphocytes and precursor cells based on the CD3 and SS signals. By using this approach, the CD3 medium population was identified by gating, and immunophenotyping was performed. In contrast, the histogram containing the CD3 low population was separated clearly into 4 categories: the CD3 low population (gate A), granulomonocytes (gate B), mature B lymphocytes and precursor cells (gate C), and T lymphoblasts and T lymphocytes (gate D). The CD3 low population showed low CD3 fluorescence intensity and a low SS signal, the granulomonocytes showed negative CD3 fluorescence intensity and the highest SS signal, the mature B lymphocytes and precursor cells showed negative CD3 fluorescence intensity and a low SS signal, and the mature T lymphocytes and T lymphoblasts showed the highest CD3 fluorescence intensity and a low SS signal. Hence, similar to the CD3 medium population, the CD3 low population was easily distinguishable from other cells based on the CD3 and SS signals. Following identification, immunophenotyping of the CD3 low population was performed. Flow Cytometric Analysis Table 1 and Table 2 give the results of flow cytometric analysis of each specimen. This analysis showed aberrant T- cell antigen loss or reduced expression in 6 cases (38%) in the CD3 medium population and in 14 cases (100%) in the CD3 low population. A reactive or inconclusive phenotype was apparent in 12 cases (75%) in the CD3 medium population and in 1 case (7%) in the CD3 low population. Loss of CD7 was the most FL 50.0 40.0 30.0 20.0 10.0 0.0 CD3 low Gating (n = 14) * Control (n = 15) CD3 medium Gating (n = 16) Figure 1 The abscissa shows the fluorescence intensity (FL), and the ordinate indicates the gating procedure. The mean FL of the CD3 low, CD3 medium, and healthy volunteer (control) populations are shown. The box for each population extends from the 25th to the 75th percentile, and the line in the middle of the box represents the median value. No statistically significant difference in the mean FL was observed between the control and CD3 medium populations. The mean FL for the CD3 low population was significantly lower than that for the control population. * P <.001. P =.037. common single T-cell antigen abnormality, occurring in 6 cases (38%) in the CD3 medium population and in all cases (100%) in the CD3 low population. Coexpression of CD4 and CD8 occurred in 8 cases (50%) in the CD3 medium population but did not occur in the CD3 low population. Coexpression of CD45RO and CD45RA occurred in 6 cases (43%) in the CD3 medium population and 1 case (7%) in the CD3 low population. Regarding differences between the CD3 medium and CD3 low populations, there was no significant difference in the expression of CD2, CD4, CD5, CD28, CD62L, CD45RA, CD45RO, and HLA-DR. However, the expression levels of CD7 and CD8 in the CD3 low population were significantly lower than those in the CD3 medium population (P <.001 and P <.04, respectively) and expression of CD25 in the CD3 low population was significantly higher than that in the CD3 medium population (P <.01) Figure 2. Am J Clin Pathol 2005;124:199-204 201 201 DOI: 10.1309/KEN4MXM5Y9A1GEMP 201

Yokote et al / ATLL IMMUNOPHENOTYPING USING CD3 GATING Table 1 Immunophenotyping of Adult T-Cell Leukemia/Lymphoma Specimens * Case No. Sample Type CD7 CD5 CD2 CD4 CD8 CD25 CD28 CD62L CD45RA CD45RO HLA-DR 1 PB + + + + + + + 2 PB + + + + + + + + + 3 PB + + + + + + + + 4 BM + + + + + 5 BAL ND + + ND ND ND ND 6 LN + + + + ND ND ND ND 7 LN + + + + + + 8 LN + + + + + + + 9 PB + + + + + + + + + 10 PB + + + + + + + + + 11 PB + + + + + + + + + + 12 PB + + + + + + + + 13 PB + + + + + + + + 14 BM + + + + + + + + 15 BM + + + + + + + + + + 16 PL + + + + + + + + + 17 PB + + ND ND + + + + 18 LN + + + + + + + 19 LN + + + + + + + + 20 PB + + + + + + + + 21 PB + + + + + + 22 PB + + + + + + + 23 PB + + + + + + + 24 PB + + + + + + 25 PB + + + + 26 PB + + + + + + + + + 27 PB + + + + + + + 28 PB + + + + + + 29 PB + + + + + + + 30 PB + + + + + + + BAL, bronchoalveolar lavage fluid; BM, bone marrow; LN, lymph node; ND, not done; PB, peripheral blood; PL, pleural effusion; +, positive;, negative. * The CD3 peak fluorescence intensity below 10 on the log scale is defined as CD3 low, an intensity between 10 and 100 as CD3 medium. Samples 1-16 were analyzed using CD3 medium gating, and samples 17-30 were analyzed using CD3 low gating. Table 2 Summary of Immunophenotyping Using CD3 low and CD3 medium Gating * Immunophenotype CD3 low Gating CD3 medium Gating CD7 0/14 (0) 10/16 (63) CD5 13/14 (93) 15/15 (100) CD2 13/14 (93) 15/15 (100) CD4 11/12 (92) 16/16 (100) CD8 1/12 (8) 8/16 (50) CD25 10/14 (71) 3/16 (19) CD28 13/14 (93) 10/14 (71) CD62L 14/14 (100) 12/15 (80) CD45RA 3/14 (21) 6/14 (43) CD45RO 11/14 (79) 13/14 (93) HLA-DR 5/14 (36) 10/15 (67) * Data are given as number of positive cases/total number of cases (percentage of positive cases). Immunohistochemical Analysis Immunohistochemical analysis using CD7, CD4, and CD8 was performed for all samples. The ATLL cells in all specimens were not immunoreactive with CD7. In all specimens but one, the ATLL cells were immunoreactive with CD4 and were not immunoreactive with CD8. In the 1 aberrant case, for which flow cytometric analysis indicated the presence of a CD3 low population, the ATLL cells were immunoreactive with CD8 and were not immunoreactive with CD4. Because CD8+ and CD4 ATLL has been reported, 2,3 we did not conclude that this case should be categorized as having an inconclusive phenotype. The results of immunohistochemical analysis with CD7, CD4, and CD8 were consistent with immunophenotyping performed by flow cytometric analysis with CD3 low gating. In contrast, immunohistochemical analysis performed with CD7, CD4, and CD8 was inconsistent with the flow cytometric analysis using the CD3 medium gating procedure, suggesting that this procedure might not exclude normal T cells from ATLL cells. Discussion ATLL is a distinct clinicopathologic syndrome characterized by a mature T-cell surface marker profile, an association with HTLV-1, the presence of abnormal lymphocytosis with pleomorphic nuclei, 4-6 and clinical and cytogenetic diversity. 4,7-9 The abnormal lymphocytes typically express CD2, CD3, CD4, CD5, and CD25; rarely express CD8; and lack CD7 expression in immunophenotypic analysis. 10,11 Because the morphologic 202 Am J Clin Pathol 2005;124:199-204 202 DOI: 10.1309/KEN4MXM5Y9A1GEMP

Hematopathology / ORIGINAL ARTICLE * 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% CD7 CD5 CD2 CD4 CD8 CD25 CD28 CD62L CD45RA CD45RO HLA-DR Figure 2 There was no significant difference in the expression of CD2, CD4, CD5, CD28, CD62L, CD45RA, CD45RO, and HLA- DR. The expression of CD7 and CD8 in the CD3 low gating population (black bars) was significantly lower than that in the CD3 medium gating population (white bars), and CD25 expression in the CD3 low gating population was significantly higher than that in the CD3 medium gating population. * P <.001. P <.04. P <.01. features of ATLL are diverse, confirmation of the presence of the disease is obtained through use of a specific antibody panel for immunocytochemical analysis, and flow cytometry can be used to help distinguish ATLL from other peripheral T-cell lymphoproliferative diseases, such as T-cell prolymphocytic leukemia, certain kinds of peripheral T-cell lymphoma, and mycosis fungoides/sézary syndrome. 10 The expression of CD2, CD3, CD4, CD8, CD25, CD28, CD45RA, CD45RO, and HLA-DR on ATLL cells has been reported, but these data are based on immunophenotyping combined with flow cytometric analysis using a conventional gating step for forward scatter and SS or on immunohistochemical analysis. 12-16 Flow cytometric analysis has been introduced to study hematologic malignancy based on surface antigen levels, because malignant cells often display an aberrant phenotype, which may involve antigen loss or changes in combinations of surface markers, compared with normal hematologic cells. However, most T-cell antigens are expressed on normal and abnormal T lymphocytes. Hence, it is difficult to discriminate between normal and abnormal T cells by a conventional immunophenotyping strategy 17 because the conventional gating step for forward scatter and SS does not discriminate well between such cells. In acute myeloid leukemia, good discrimination can be achieved between the blast cell population and normal cells through the systematic use of leukocyte common antigen (CD45) expression and SS. 18 This discrimination is based on the fact that the precursor cells express low and intermediate levels of CD45, whereas mature lymphocytes express a high level of CD45. 19 However, because ATLL is a lymphoproliferative disease of peripheral T-cell origin, the ATLL cells partially overlap with the gated normal lymphocytes, and, therefore, this procedure cannot discriminate between normal and ATLL cells. CD3 is part of the T-cell receptor complex and is involved in transducing stimulatory signals after antigen-specific recognition. 20 It is expressed in the membrane at the late stage of thymic differentiation and in mature (peripheral) T lymphocytes, 21 after which CD3 can be detected only in the cytoplasm at the earliest stages of T-cell differentiation. The expression of CD3 is down-regulated after antigen recognition and activation. 22 Similarly, ATLL cells are activated T cells that mostly express interleukin-2 receptor (CD25), and it has been suggested that CD3 is down-regulated in ATLL cells after activation and infection with HTLV-1. 23,24 Therefore, we anticipated that good discrimination between normal T cells and ATLL cells might be achieved through the use of CD3 expression. The revised European-American Lymphoma (REAL) classification is in widespread use. This system emphasizes immunophenotyping analysis with clinical features and morphologic criteria. In the REAL classification, ATLL is placed in the category of peripheral T-cell and natural killer cell neoplasms. Tumor cells typically express CD2, CD3, CD4, CD5, and CD25; rarely express CD8; and usually lack CD7, based on immunophenotypic analysis. 10,11 Immunophenotyping performed by flow cytometric analysis with CD3 low gating is largely consistent with the immunophenotyping criteria suggested in the REAL classification, 10 as previous reports also have stated. 12-16 We studied the immunophenotyping of ATLL by flow cytometric analysis with CD3 gating. Because statistically significant differences in immunophenotyping were apparent between CD3 low and CD3 medium gating, we compared the results Am J Clin Pathol 2005;124:199-204 203 203 DOI: 10.1309/KEN4MXM5Y9A1GEMP 203

Yokote et al / ATLL IMMUNOPHENOTYPING USING CD3 GATING of flow cytometric analysis with immunohistochemical data for the expression of CD7, CD4, and CD8. Flow cytometric analysis is easy to perform, reliable, and inexpensive, and clinically relevant data can be obtained soon after completion of flow cytometry measurements. The CD3 low gating procedure is more sensitive than conventional or CD45/SS gating and seems to be an appropriate method of choice. The clinical course of ATLL is very aggressive, 25,26 but recently it has been suggested that the disease might be cured by allogeneic stem cell transplantation. 25 We also believe that minimal residual disease monitoring is clinically useful during follow-up after therapy. If specific surface markers in ATLL cells are determined, flow cytometric assays can efficiently complement techniques for minimal residual disease monitoring, particularly in combination with studies of T-cell receptor gene rearrangement. From the 1 First Department of Internal Medicine, 2 Division of Surgical Pathology, and 3 Department of Clinical Central Laboratory, Osaka Medical College, Osaka, Japan. Address reprint requests to Dr Yokote: First Department of Internal Medicine, Osaka Medical College, 2-7 Daigakumachi, Takatsuki City, Osaka 569-0801, Japan. References 1. Shimoyama M. Diagnostic criteria and classification of clinical subtypes of adult T-cell leukaemia-lymphoma: a report from the Lymphoma Study Group (1984-87). Br J Haematol. 1991;79:428-437. 2. Matsushita K, Arima N, Hidaka S, et al. CD8-positive adult T-cell leukemia cells with an integrated defective HTLV-I genome show a paracrine growth to IL-2. Am J Hematol. 1994;47:123-128. 3. Lorand-Metze I, Pombo-de-Oliveira M. Adult T-cell leukemia (ATL) with unusual immunophenotype and a high cellular proliferation rate. Leuk Lymphoma. 1996;22:523-526. 4. Uchiyama T, Yodoi J, Sasagawa K, et al. Adult T-cell leukemia: clinical and hematologic features of 16 cases. Blood. 1977;50:481-492. 5. Hinuma Y, Nagata K, Hanaoka M, et al. Adult T-cell leukemia: antigen in an ATL cell line and detection of antibodies to the antigen in human sera. Proc Natl Acad Sci U S A. 1981;78:6476-6480. 6. Yoshida M, Seiki M, Yamaguchi K, et al. Monoclonal integration of human T-cell leukemia provirus in all primary tumors of adult T-cell leukemia suggests causative role of human T-cell leukemia virus in the disease. Proc Natl Acad Sci U S A. 1984;81:2534-2537. 7. Kikuchi M, Mitsui T, Takeshita M, et al. Virus-associated adult T-cell leukemia (ATL) in Japan: clinical, histological and immunological studies. Hematol Oncol. 1986;4:67-81. 8. Duggan DB, Ehrlich GD, Davey FP, et al. HTLV-I-induced lymphoma mimicking Hodgkin s disease; diagnosis by polymerase chain reaction amplification of specific HTLV-I sequences in tumor DNA. Blood. 1988;71:1027-1032. 9. Itoyama T, Chaganti RS, Yamada Y, et al. Cytogenetic analysis and clinical significance in adult T-cell leukemia/lymphoma: a study of 50 cases from the human T-cell leukemia virus type-1 endemic area, Nagasaki. Blood. 2001;97:3612-3620. 10. 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;84:1361-1392. 11. Dahmoush L, Hijazi Y, Barnes E, et al. Adult T-cell leukemia/ lymphoma: a cytopathologic, immunocytochemical, and flow cytometric study. Cancer. 2002;96:110-116. 12. Suzuki M, Uno H, Yamashita K, et al. Clinical significance of CD45RO expression on peripheral blood mononuclear cells in HTLV-I infected individuals. Br J Haematol. 1996;92:401-409. 13. Worner I, Matutes E, Beverley PC, et al. The distribution of CD45R, CD29 and CD45RO (UCHL1) antigens in mature CD4 positive T-cell leukaemias. Br J Haematol. 1990;74:439-444. 14. Richardson JH, Edwards AJ, Cruickshank JK, et al. In vivo cellular tropism of human T-cell leukemia virus type 1. J Virol. 1990;64:5682-5687. 15. Nagatani T, Matsuzaki T, Iemoto G, et al. Comparative study of cutaneous T-cell lymphoma and adult T-cell leukemia/lymphoma: clinical, histopathologic, and immunohistochemical analyses. Cancer. 1990;66:2380-2386. 16. Shirono K, Hattori T, Hata H, et al. Profiles of expression of activated cell antigens on peripheral blood and lymph node cells from different clinical stages of adult T-cell leukemia. Blood. 1989;73:1664-1671. 17. Ginaldi L, Matutes E, Farahat N, et al. Differential expression of CD3 and CD7 in T-cell malignancies: a quantitative study by flow cytometry. Br J Haematol. 1996;93:921-927. 18. Borowitz MJ, Guenther KL, Shults KE, et al. Immunophenotyping of acute leukemia by flow cytometric analysis: use of CD45 and right-angle light scatter to gate on leukemic blasts in three-color analysis. Am J Clin Pathol. 1993;100:534-540. 19. Shah VO, Civin CI, Loken MR. Flow cytometric analysis of human bone marrow, IV: differential quantitative expression of T-200 common leukocyte antigen during normal hemopoiesis. J Immunol. 1988;140:1861-1867. 20. Meuer SC, Acuto O, Hussey RE, et al. Evidence for the T3- associated 90K heterodimer as the T-cell antigen receptor. Nature. 1983;303:808-810. 21. Reinherz EL, Kung PC, Goldstein G, et al. Discrete stages of human intrathymic differentiation: analysis of normal thymocytes and leukemic lymphoblasts of T-cell lineage. Proc Natl Acad Sci U S A. 1980;77:1588-1592. 22. Beverley P. The importance of T3 in the activation of T lymphocytes. Nature. 1983;304:398-399. 23. Yoshida M, Miyoshi I, Hinuma Y. Isolation and characterization of retrovirus from cell lines of human adult T-cell leukemia and its implication in the disease. Proc Natl Acad Sci U S A. 1982;79:2031-2035. 24. Uchiyama T, Hori T, Tsudo M, et al. Interleukin-2 receptor (Tac antigen) expressed on adult T cell leukemia cells. J Clin Invest. 1985;76:446-453. 25. Utsunomiya A, Miyazaki Y, Takatsuka Y, et al. Improved outcome of adult T-cell leukemia/lymphoma with allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant. 2001;27:15-20. 26. Yamada Y, Tomonaga M, Fukunaga H, et al. A new G- CSF supported combination chemotherapy, LSG15, for adult T-cell leukemia-lymphoma: Japan Clinical Oncology Group Study 9303. Br J Haematol. 2001;113:375-382. 204 Am J Clin Pathol 2005;124:199-204 204 DOI: 10.1309/KEN4MXM5Y9A1GEMP