Summary. Conclusions Our results show a close correlation between clinicopathological features of HTLV-I-associated cutaneous lesions and prognosis.

Transcription

1 CLINICAL AND LABORATORY INVESTIGATIONS DOI /J X Clinicopathological features of cutaneous lesions of adult T-cell leukaemia/ lymphoma T. Yamaguchi,* K. Ohshima,* K. Karube,* T. Tutiya,* R. Kawano,* H. Suefuji,* A. Shimizu, J. Nakayama, J. Suzumiya,à Y. Moroi, K. Urabe, M. Furue, T. Koga and M. Kikuchi* Departments of *Pathology, Dermatology and àinternal Medicine, School of Medicine, Fukuoka University, Nanakuma, Jonan-ku, Fukuoka 814-1, Japan Department of Dermatology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Division of Dermatology, Fukuoka Red Cross Hospital, Fukuoka, Japan Summary Correspondence Takahiro Yamaguchi. Accepted for publication 1 June 24 Key words: adult T-cell leukaemia lymphoma, cutaneous lesions, prognosis, proviral DNA integration Background Adult T-cell leukaemia lymphoma (ATLL) is a human malignancy associated with human T-cell leukaemia virus type I (HTLV-I). ATLL frequently involves the skin. Objectives To correlate the clinicopathological features and prognosis in patients with ATLL and cutaneous lesions. Methods We examined the HTLV-I proviral state and the clinicopathological features of the cutaneous lesions in patients with serum anti-atl antibody, to clarify the correlation between macroscopic histopathological findings and prognosis. Southern blot analysis was performed in all cases to detect monoclonal HTLV-I proviral DNA integration. Results The cutaneous lesions of 46 patients were positive for proviral DNA integration. The median survival time of patients with monoclonal proviral DNA integration in cutaneous lesions was 14 months, which was markedly shorter than that of patients negative for proviral DNA integration (72 months). Of the 46 patients with proviral DNA, 21 had solitary or multiple red nodules (including three with subcutaneous induration), eight had multiple red papules and 17 had erythema. Patients with papules and nodules had poorer prognosis than those with erythema. Histopathologically, the prognosis was poorer in patients with nodular or diffuse infiltration of medium-sized to large lymphoma cells, compared with those with perivascular infiltration of small to medium-sized lymphoma cells. Conclusions Our results show a close correlation between clinicopathological features of HTLV-I-associated cutaneous lesions and prognosis. Adult T-cell leukaemia lymphoma (ATLL) is a human malignancy associated with human T-cell leukaemia virus type I (HTLV-I). 1 ATLL can be diagnosed based on clinicopathological findings and the presence of monoclonal integrated HTLV- I provirus in the DNA of tumour cells. 2,3 Prototypic ATLL has a rapid course and is characterized by the presence of atypical peripheral blood cells with markedly indented nuclei. Hypercalcaemia, hepatosplenomegaly and lymphadenopathy are frequently associated with this disease. 1 The mode of HTLV-I proviral DNA integration changes from undetectable to polyclonal, and then further to monoclonal upon malignant transformation. Analysis of proviral DNA integrated in cellular DNA showed that leukaemic cells were always infected with HTLV-I. 2,3 These cells were also consistently monoclonal with respect to proviral integration, which indicated that they originated from a single cell infected with HTLV-I. 2,3 Based on the clinical features, a diversity of clinical and prognostic features has been noted and several subgroupings of ATLL have been proposed, mainly based on haematological and serological features. The subgroups are referred to as acute, chronic, lymphoma and smouldering types of ATLL. 4 ATLL commonly involves the skin as well as the peripheral blood and lymph nodes. Cutaneous lesions of ATLL are polymorphous in appearance; skin lesions are frequently observed in all clinical subtypes and their incidence was reported to vary from 43% to 72%. 5 In addition, a cutaneous type of ATLL (catll) has been proposed as a clinical subtype with 76

2 Cutaneous lesions in ATLL, T. Yamaguchi et al. 77 cutaneous lesions persisting throughout the entire course, not easily developing into lymphoma or nodal lymphoma. 6 9 To our knowledge, there has been no large-scale study of cutaneous lesions of ATLL. In the present study, we studied macroscopic and histopathological features in patients with ATLL with skin manifestations. We also analysed the clonality of HTLV-I-infected cells. Materials and methods Patients This study was based on patients with HTLV-I with cutaneous lesions identified in the skin registry files of the Department of Pathology, Fukuoka University. All patients were positive for adult T-cell leukaemia virus-associated antigen and or amplified HTLV-I px gene. Southern blot analyses were performed in all patients, and monoclonal integration of HTLV-I proviral DNA into atypical lymphoid cells was detected in the cutaneous lesions of 46 patients. We divided the cutaneous lesions into three types (Fig. 1): erythema (including erythrodermic eruptions), papules and nodules (including subcutaneous induration). When two or more types of lesion were found, we recorded the most serious form using the following order of increasing severity: erythema, papules, nodules. Histopathological analysis In addition to confirming the diagnosis by histopathological examination, we further divided all 46 cases with monoclonal HTLV-I provirus into three histopathological types: perivascular infiltration, nodular infiltration and diffuse infiltration (Fig. 2). Cases with a perivascular pattern showed patchy perivascular infiltration (including three cases of lichenoid or band-like infiltration) of small to medium-sized atypical lymphoid cells with mild to moderate nuclear atypia in the upper a b c dermis. Epidermotropism was also observed in this group. Cases with nodular infiltration showed nodular infiltration of medium-sized to large atypical lymphoid cells with moderate to severe nuclear atypia. Cases with diffuse infiltration showed diffuse interstitial infiltration of medium-sized to large atypical lymphoid cells with moderate to severe nuclear atypia. The latter two types showed massive or extensive infiltration of atypical lymphoid cells throughout the entire dermis and or subcutaneous adipose tissue, and mitotic figures and convoluted nucleoli were occasionally seen. Epidermotropism was rarely seen in this group. Immunohistochemistry Paraffin-embedded skin specimens were stained immunohistochemically for CD2 (L26, B-cell marker; Dako, Glostrup, Denmark), CD45RO and CD3 (UCHL-1, T-cell marker; Dako) and or CD3 (BerH2; Dako). A portion of each skin specimen was maintained at ) C in a deep freezer. These tissues were examined using monoclonal antibodies to CD2, CD3, CD4, CD8, CD15, CD19, CD2 and CD3 (Coulter, Hialeah, FL, U.S.A.; Ortho, Raritan, NJ, U.S.A.; Becton Dickinson, Mountain View, CA, U.S.A.; Dako). DNA analysis The other part of each frozen tissue sample was used for DNA isolation and gene analysis. The methods used for this analysis have been reported previously. 1,11 We also examined T-cell receptor (TCR) genes Cb and Jc, the immunoglobulin heavy chain (JH) gene and proviral DNA of HTLV-I DNA (full-length probe, including gag, pol, env, px and LTR). Monoclonal integration of HTLV-I DNA was examined by digestion with EcoRI, using the method described previously by our group. 1 The isolated DNA was amplified using polymerase chain reaction (PCR). Specific primers were synthesized based on the published DNA sequences (primer px1: ATGCTGTTTCGCTTCTCAG; primer px2: TAAGGACCTT GAGGGCTTA; probe: TAGCGACGTCAGCGCCTTC) corresponding to the px region. Amplification was carried out in a DNA Thermal Cycler (Perkin-Elmer Cetus, Norwalk, CT, U.S.A.) with denaturation at 94 C (1 min), annealing at 5 C and extension at 72 C (2 min each). After 3 cycles of PCR amplification, one-tenth (1 ll) of the reaction mixture was analysed by Southern blot to confirm the size of the PCR product, in a manner similar to that reported in our previous study. 12 Statistical analysis and ethical considerations Fig 1. Macroscopic appearances of skin lesions in patients with adult T-cell leukaemia lymphoma. (a) Erythematous macules; (b) disseminated erythematous papules; (c) Solitary red nodule. Survival intervals were estimated from the date of diagnosis to the end point (date of the last follow-up or death). The actuarial overall survival was evaluated by the Kaplan Meier method and was analysed by the log-rank and Wilcoxon methods. P < Æ5 denoted the presence of a statistically significant difference. The study protocol was approved by the

3 78 Cutaneous lesions in ATLL, T. Yamaguchi et al. a b c Fig 2. Histopathological findings (low magnification). (a) Perivascular and patchy infiltrate of small to medium-sized atypical lymphoid cells with mild to moderate nuclear atypia in the upper dermis. (b) Diffuse interstitial infiltrate of medium-sized to large atypical lymphoid cells throughout the entire dermis. (c) Nodular infiltrate of mediumsized to large atypical lymphoid cells throughout the entire dermis. Haematoxylin and eosin; original magnification 1. Human Ethics Review Committee of Fukuoka University School of Medicine. Results Clinical features The patients comprised 44 men and 36 women (male female ratio 1Æ2 : 1), age range years (mean 61Æ8). Bone marrow involvement was detected in 19 of the patients. Twenty-eight cases were histopathologically confirmed to have lymph node lesions of lymphoma. Pulmonary involvement was detected in eight patients. At the time of this report, 19 patients were still alive at 6 68 months after diagnosis with a median follow-up time of 36 months following diagnosis. The mean survival time (MST) of all patients was 24 months. Patients received treatment with topical corticosteroids, psoralen plus ultraviolet A therapy, electron beam therapy or systemic combination chemotherapy with prednisolone, vincristine, doxorubicin, cyclophosphamide, methotrexate, etoposide, interferon and other agents. Histopathology and immunohistology Individual cutaneous lesions showed various degrees of lymphoma cell infiltration from the epidermis (as Pautrier s microabscesses or exocytosis) to the deep dermis and even to the subcutaneous fatty tissue. These atypical lymphoid cells were pleomorphic in shape and size (small, medium and large). Histopathologically, cases with erythema showed perivascular or diffuse infiltration in the upper dermis of small to medium-sized atypical lymphoid cells with mild to moderate nuclear atypia. Mitotic figures were rarely seen. Cases with papules and nodules showed nodular or diffuse infiltration from the dermis to the subcutaneous adipose tissue of medium-sized to large atypical lymphoid cells with moderate to severe nuclear atypia. Mitotic figures were occasionally seen in these lesions (Figs 2 and 3). These atypical lymphoid cells were positive for CD2, CD3, CD4, CD5 and CD25, but negative for CD7, CD8, CD19, CD2, CD3, CD68 and T-cell intracellular antigen-1. Human T-cell leukaemia virus type I analysis In all cases, clonal integration of HTLV-I proviral DNA in the skin lesion was examined by Southern blot analysis. Forty-six patients showed clonal integration of HTLV-I proviral DNA. We then analysed the survival rate according to the presence or absence of HTLV-I proviral DNA integration in skin lesions. The 46 patients with positive HTLV-I proviral DNA integration in skin lesions showed more rapid progression with a shorter clinical history than the remaining 34 cases with no clonal integration of HTLV-I proviral DNA. Furthermore, the MST of patients with HTLV-I proviral DNA integration was 14 months, while that of patients with negative proviral DNA integration was 72 months (Fig. 4a). Tissues of patients negative for proviral DNA integration were amplified by PCR, and HTLV-I infection was confirmed. One reason for the lack of monoclonal proviral DNA bands was the small population of HTLV-I-infected cells. a b c Fig 3. Histopathological findings (high magnification). (a) Small to medium-sized atypical lymphoid cells with mild to moderate nuclear atypia showing perivascular infiltration. (b) Medium-sized to large atypical lymphoid cells with moderate nuclear atypia showing diffuse interstitial infiltration. (c) Medium-sized to large atypical lymphoid cells with severe nuclear atypia showing nodular infiltration. Haematoxylin and eosin; original magnification 1.

4 Survival (%) Cutaneous lesions in ATLL, T. Yamaguchi et al. 79 Fig 4. Survival curves of patients with adult T-cell leukaemia lymphoma with cutaneous manifestations: (a) with or without monoclonal proviral DNA integration (P < Æ1); (b) according to macroscopic findings (P ¼ Æ6); (c) according to infiltration pattern of atypical lymphoid cells (P ¼ Æ1); and (d) according to size of atypical lymphoid cells (P ¼ Æ8). (a) 1 Survival (%) Survival (%) 4 2 (c) proviral DNA (-) (b) 1 4 erythema papule proviral DNA(+) 2 nodule (d) 1 Survival (%) perivascular 4 medium small 2 large nodular diffuse Human T-cell leukaemia virus type I provirus status and features of cutaneous lesions In 46 cases positive for proviral DNA integration, 1 were classified as acute (MST 6 months), nine as chronic (MST 2 months), nine as lymphoma (MST 11 months) and 18 as smouldering (MST 35 months). The cutaneous eruptions in these 46 cases were also classified by macroscopic findings: 21 showed solitary or multiple red nodules (including three cases of subcutaneous multiple indurations), eight showed multiple red papules, and 17 showed erythema (including three cases of erythrodermic eruptions). Furthermore, in the same group of 46 cases, 2 showed nodular infiltration, 17 showed diffuse infiltration, and nine showed perivascular infiltration (including two cases with a lichenoid or band-like infiltration pattern). The macroscopic findings approximately corresponded to histological findings. We then studied the correlation between macroscopic and histopathological findings in cases positive for proviral DNA integration. In cases with nodules, 18 showed nodular infiltration, three showed diffuse infiltration, and none showed perivascular infiltration. In cases with papules, two showed nodular infiltration, six showed diffuse infiltration, and none showed perivascular infiltration. In cases with erythema, eight showed diffuse infiltration, nine showed perivascular infiltration and none showed nodular infiltration. Patients with nodules and papules showed poor prognosis: the MST was 9 months for individuals with nodules, 11 months for those with papules, and 32 months for those with erythema. Thus, in our series, patients with nodules and papules showed a poorer prognosis than those with erythema. We also analysed prognosis according to histopathological type. Patients with nodular infiltration of atypical lymphoid cells had the poorest prognosis (MST 9 months); the MST was 2 months for patients with diffuse infiltration and 24 months for those with perivascular infiltration. Finally, we analysed prognosis according to the size of atypical lymphoid cells. Large atypical lymphoid cells had nuclei at least twice the size of those in small atypical lymphoid cells, and the nuclei were usually larger than tissue macrophage nuclei. Large atypical lymphoid cells were found in lesions displaying a nodular and diffuse infiltration pattern. Medium-sized atypical lymphoid cells were found in varying numbers in lesions of all infiltration patterns. Small atypical lymphoid cells were found in lesions displaying perivascular and diffuse infiltration patterns. When lymphoma cells of two or more sizes were found in one histological specimen, we recorded the ones that occurred most frequently. The presence of large, medium-sized and small atypical lymphoma cells was associated with progression from worst to best prognosis (Fig. 4b d). Discussion We have reported the clinical and histopathological features of cutaneous lesions of ATLL. Cases positive for proviral DNA integration showed poorer prognosis than those negative for proviral DNA integration. The former group showed mostly nodular or diffuse infiltration of medium-sized to large atypical lymphoid cells with severe nuclear atypia. Among patients positive for proviral DNA integration, those with nodules or papules had an extremely poor prognosis compared with patients with erythema. Furthermore, patients with nodular infiltration of lymphoma cells exhibited an extremely poor prognosis compared with those with diffuse and perivascular infiltration. Clonal evolution is seen in the multistep carcinogenesis of most human malignancies. A clonal change of ATLL at the time of crisis, corresponding to one stage in multistep

5 Cutaneous lesions in ATLL, T. Yamaguchi et al. carcinogenesis, is unusual. Two mechanisms could explain the development of a clonal change in ATLL. One is an oligoclonal expansion of HTLV-I-infected premalignant cells even in asymptomatic carriers, detected by inverse PCR or by linker ligation PCR. 13,14 These oligoclones in HTLV-I carriers may be the origin of a new ATLL clone during crisis. The other mechanism is the immunosuppressive state documented in patients even in the early stage of ATLL, or in HTLV-I carriers. 15,16 Three studies have documented the development of ATLL during immunosuppressive therapy in renal transplant recipients On the other hand, in patients with cutaneous T-cell lymphoma (CTCL), especially those with early stage CTCL, polyclonal T-cell populations were detected in tissue DNA. For example, two different identical TCR gene rearrangement bands were identified in one patient, thus describing two groups of patients with early stage mycosis fungoides (MF): those with a monoclonal T-cell neoplasm and those with polyclonal lymphoproliferative MF. 2 These findings also suggested that CTCL evolves from a reactive process into a clonal T-cell neoplasm. The pathogenesis of cutaneous lesions of ATLL is entirely different from that of CTCL. In ATLL, previous studies indicated the existence of skin lesions in cases with a polyclonal or oligoclonal state prior to the monoclonal state In the present study, in some cases positive for proviral DNA, the clinical appearance changed from erythema to nodules during the course of the disease, possibly reflecting a clonal evolution from polyclonal to monoclonal in cutaneous lesions of ATLL. Further studies are necessary to confirm this hypothesis such as analysis of the integration pattern of HTLV-I proviral DNA and the use of inverse PCR. ATLL commonly involves the skin as well as peripheral blood and lymph nodes, although some reports have described patients who presented with only cutaneous lesions without either leukaemic change or visceral invasion for many years. 6 9 Johno et al. 7 proposed a new category, catll, representing skin lesions persisting throughout the entire disease course, not easily developing into leukaemia or nodal lymphoma, and indicated that the prognosis of the tumoral type of catll (MST 26 months) was worse than that of the erythematopapular type (MST months). 7 In the present study, 46 cases positive for proviral DNA were classified as acute leukaemia (n ¼ 1), chronic leukaemia (n ¼ 9), lymphoma (n ¼ 9) and smouldering (n ¼ 18), and the MST of each clinical subtype was 6, 2, 11 and 35 months, respectively. In our series, nine cases of the smouldering type were compatible with catll. This group included four cases with nodules, one case with papules and four cases with erythema, and the MST of all patients with catll was 3 months. catll showed a poorer prognosis than the smouldering type, but an extremely favourable prognosis compared with the other clinical subtypes. We also propose two clinical subtypes of catll: the erythematopapular and the tumoral types. The MST of each type was months and 19 months, respectively. Based on these findings, the presence of nodules in patients with catll is associated with a poor prognosis. catll should be defined as a special subtype of ATLL in which monoclonal malignant cells proliferate in the skin lesions only. Careful follow-up is necessary because the prognosis is not always good. In the present study, our results suggest that the type of eruption, infiltration pattern and size of lymphoma cells correlate with prognosis in cutaneous lesions of ATLL including catll. It is often difficult to differentiate between catll and CTCL by histopathology and phenotype. However, expression of CD25 is significantly higher in lymphoma cells in catll compared with CTCL. In addition, serum interleukin-2 receptor levels are higher in catll than in CTCL. 21 In contrast, in cases negative for proviral DNA integration, 29 of 34 exhibited a perivascular infiltration pattern with a relatively small number of atypical lymphoid cells and mild nuclear atypia. The remaining five cases were ATLL carriers with multiple plaques or tumours: three showed diffuse infiltration of medium-sized atypical lymphoid cells with moderate nuclear atypia, and two showed nodular infiltration of medium-sized to large atypical lymphoid cells with moderate to severe nuclear atypia. In these cases, immunohistochemistry showed that lymphoma cells were positive for CD2, CD3, CD4, CD5 and CD45RO, and negative for CD8, CD25 and CD3. TCR genes were clonally rearranged in all five cases. Based on these findings, we regard these cases as CTCL in ATLL carrier patients. A definite diagnosis of catll can be made following identification of integrated HTLV-I provirus in the DNA of tumour cells. The reported incidence of antibody-positive MF is high in ATLL-endemic areas Cases of antibodypositive MF show a more indolent clinical course than cases of ATLL with cutaneous lesions. 23,24 In the present study, cases of CTCL among ATLL carriers showed a greatly more favourable prognosis, with a survival time of 3 15 years after the onset of disease, and an MST of 9Æ2 years. Furthermore, amplification of the HTLV-I px region was detected in all 34 patients negative for proviral DNA integration. Therefore, monoclonality could not be detected in this group. In conclusion, as ATLL is accompanied by a diversity of skin manifestations, macroscopic and histopathological findings are useful tools to predict prognosis. Our results highlight the importance of macroscopic and histopathological appearance in predicting the prognosis of ATLL with cutaneous lesions. References 1 Takatsuki K. Overview of adult T-cell leukemia lymphoma. Jpn Soc Res 1985; 25: Yoshida M, Seiki M, Yamaguchi K, Takatsuki K. 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 USA 1984; 81: Takatsuki K, Yamaguchi K, Watanabe T. Adult T-cell leukemia and HTLV-1 related diseases. Gann Mono Can Res 1992; 39: Shimoyama M, for members of the Lymphoma Study Group ( ). Diagnostic criteria and classification of clinical subtypes of adult T-cell leukaemia-lymphoma. Br J Haematol 1991; 79:

6 Cutaneous lesions in ATLL, T. Yamaguchi et al Chan HL, Su IJ, Kuo T et al. Cutaneous manifestations of adult T-cell leukemia lymphoma. Report of three different forms. JAm Acad Dermatol 1985; 13: Dosaka N, Tanaka T, Miyachi Y et al. Examination of HTLV-1 integration in the skin lesions of various types of adult T-cell leukemia (ATL): independence of cutaneous-type ATL confirmed by Southern blot analysis. J Invest Dermatol 1991; 96: Johno M, Ohishi M, Kojo Y et al. Cutaneous manifestations of adult T cell leukemia lymphoma. In: Advances in Adult T-Cell Leukemia and HTLV-1 Research (Takatsuki K, Hinuma Y, Yoshida M, eds). Tokyo: Japan Scientific Society Press, 1992; Takahashi K, Tanaka T, Fujita M et al. Cutaneous-type of adult T-cell leukemia lymphoma. Arch Dermatol 1998; 124: Yagi H, Takigawa M, Hashizume H. Cutaneous type of adult T cell leukemia lymphoma: a new entity among cutaneous lymphomas. J Dermatol 23; 3: Ohshima K, Kikuchi M, Yoshida T et al. Lymph nodes in incipient adult T-cell leukaemia lymphoma with Hodgkin s disease-like histologic features. Cancer 1991; 67: Yoshida T, Kikuchi M, Ohshima K et al. ATLA-positive adult T-cell leukemia lymphoma with no monoclonal proviral DNA. A clinicopathological and immunological study. Cancer 1989; 64: Ohshima K, Suzumiya J, Kato A et al. Clonal HTLV-1 infected CD4+ lymphocytes and non-clonal non HTLV-1-infected giant cells in incipient ATLL with Hodgkin-like histologic features. Int J Cancer 1997; 72: Takemoto S, Matsuoka M, Yamaguchi K, Takatsuki K. A novel diagnostic method of adult T-cell leukemia: monoclonal integration of human T-cell lymphotropic virus type I provirus DNA detected by inverse polymerase chain reaction. Blood 1994; 84: Wattel E, Vartanian JP, Pannetier C, Wain-Hobson S. Clonal expansion of human T-cell leukemia virus type I-infected cells in asymptomatic and symptomatic carriers without malignancy. J Virol 1995; 69: Yamaguchi K, Nishimura H, Kohrogi H et al. A proposal for smoldering adult T-cell leukemia: a clinicopathologic study of 5 cases. Blood 1983; 62: Tachibana N, Okayama A, Ishizaki J et al. Suppression of tuberculin skin reaction in healthy HTLV-I carriers from Japan. Int J Cancer 1988; 42: Zanke BW, Rush DN, Jeffery JR, Israels LG. HTLV-I T cell lymphoma in a cyclosporine-treated renal transplant patient. Transplantation 1989; 48: Tsurumi H, Tani K, Tsuruta T et al. Adult T-cell leukemia developing during immunosuppressive treatment in a renal transplant recipient. Am J Hematol 1992; 41: Williams NP, Buchner LM, Shah DJ, Williams W. Adult T-cell leukemia lymphoma in a renal transplant recipient: an opportunistic occurrence. Am J Nephrol 1994; 14: Whittaker SJ, Smiths NP, Russell-Jones R, Luzzatto L. Analysis of b, c, and d T-cell receptor genes in mycosis fungoides and Sézary syndrome. Cancer 1991; 68: Oishi M, Johno M, Ono T et al. Differences in IL-2 receptor levels between mycosis fungoides and cutaneous type adult T-cell leukemia lymphoma in the early stages of the disease. J Invest Dermatol 1994; 12: D Incan M, Antoniotti O, Gasmi M et al. HTLV-1-associated lymphoma presenting as mycosis fungoides in an HTLV-1 non-endemic area: an in vivo-molecular study. Br J Dermatol 1995; 132: Miyagi T, Miyasato H, Higa T et al. ATL (adult T cell leukemia) clinically manifesting MF (mycosis fungoides). Nishinihon J Dermatol 1991; 53:53 4 (in Japanese). 24 Amagasaki T, Momita S, Suzuyama J et al. Detection of adult T-cell leukemia-associated antigen in T-cell malignancies in the Nagasaki district of Japan. Cancer 1984; 15: Tobinai K, Nagai M, Setoya T et al. Anti-ATLA (antibody to adult T-cell leukemia virus-associated antigen), highly positive in OKT4- positive mature T-cell malignancies. Jpn J Clin Oncol 1983; 13 (Suppl. 2):