Lymphoproliferative disorders in renal transplant patients in Japan



Post-transplantation lymphoproliferative disorders (PT-LPD) are characterized by a clinically and morphologically heterogeneous group of lymphoid proliferation occurring after organ or bone marrow transplantation. The immunodeficient state provides a basis for lymphomagenesis probably through activation of oncogenic viruses. Twenty-four patients in whom PT-LPD developed after renal transplantation in Japan were analyzed. They received hemodialysis for 4 to 226 (median 13) months before transplantation. In situ hybridization was performed to detect Epstein-Barr virus (EBV). Polymerase chain reaction and Southern hybridization with primers in the tax and pol regions of human T-cell leukemia virus type I (HTLV-1) were performed on DNA extracted from paraffin-embedded specimens. Immunohistochemical analysis revealed that 12 cases were B-cell type, 10 cases (42%) T-cell type and 2 NK-cell type. Five of the T-cell cases were classified as adult T-cell lymphoma with proven HTLV-1 genome in the tumor and seropositivity for the virus. These cases were classified as adult T-cell lymphoma (ALT). More than 80% of B-cell, 30% of T-cell and both NK/T-cell lymphomas were EBV-positive. Co-infection of EBV and HTLV-1 was found in 2 cases with ATL. These findings showed that ATL is common among Japanese renal transplant patients, which might be due to transmission of HTLV-1 via blood transfusion during hemodialysis. © 2001 Wiley-Liss, Inc.

Immunodeficient individuals show an increased incidence of cancer, and malignant lymphoma is one of the most common malignancies developing under such conditions. Penn et al.1 first reported the increased incidence of malignant lymphoma in recipients of solid-organ transplants treated with immunosuppressive agents. Malignant lymphomas are also frequently observed in patients with acquired immune deficiency syndrome (AIDS).2 Irrespective of the cause, the immunodeficient state is considered to provide a basis for development of malignant lymphomas probably through activation of Epstein-Barr virus (EBV). Latent infection genes of EBV, especially latent membrane protein 1 (LMP-1) and EB nuclear antigen 2 (EBNA-2), show transforming activity in infected cells. LMP-1 in EBV-infected cells plays a central role in B-lymphomagenesis by mimicking members of the tumor necrosis factor (TNF) receptor family, thereby transmitting growth signals from the cell membrane to the nucleus through cytoplasmic TNF-associated factors.3 These molecules, however, serve as targets for host cytotoxic T-lymphocytes (CTL).4–6 In the immunodeficient state, lymphoid cells expressing these molecules can escape from host CTL, which may result in development of malignant lymphoma.7 This might explain the increased risk of malignant lymphoma in immunodeficient individuals.

Post-transplantation lymphoproliferative disorders (PT-LPD) are a clinically and morphologically heterogeneous group of lymphoid proliferative disorders occurring after organ or bone marrow transplantation.8 PT-LPD are EBV related, have been B-cell predominant and have aggressive clinical courses. PT-LPD have been categorized into 4 groups:9 reactive plasmacytic hyperplasia, polymorphic PT-LPD, monomorphic PT-LPD and others including plasmacytoma-like PT-LPD. Clinical outcomes of plasmacytic hyperplasia and polymorphic PT-LPD are significantly better than that of monomorphic PT-LPD.

Previous immunophenotypic studies in Western countries revealed that the majority of PT-LPD are B-cell derived.10, 11 This was in accordance with the development of non-Hodgkin's lymphoma (NHL) in immunocompetent individuals in the same areas. Adult T-cell leukemia/lymphoma (ATL), which is rare in Western countries, is common in Japan.12 ATL is especially common in southwestern Japan, where it is endemic, an area in which more than 75% of adult NHL are of T-cell type. A causal relationship has been well established between ATL manifestations and infection by human T-cell leukemia virus type I (HTLV-1), discovered in cultivated tumor cells of patients with cutaneous T-cell malignancy13 or detected in T-cell lines derived from leukemic cells and from lymphocytes from ATL.14 Almost all renal transplant patients in Japan receive hemodialysis in which blood transfusion, one of the main routes of HTLV-1 transmission, is commonly employed. Therefore, PT-LPD in renal transplants in Japan might include ATL.

In our study, the PT-LPD in renal transplants in Japan were analyzed with special emphasis on immunophenotype and relationship with EBV and HTLV-1.


Patient characteristics

Twenty-four patients in whom lymphoproliferative disorders developed after renal transplantation were selected for the present study: 6 patients through a review of the records of renal transplantation during the period from 1970 to 1995 at 3 hospitals in Japan (Osaka University Hospital, Hyogo Prefectural Nishinomiya Hospital and Kinki University Hospital); 10 through the review of Japanese journals; 5 through the “Annual of the Pathological Autopsy Cases in Japan (1977–1993)”; and 3 through consultation case files of our department. Histological specimens obtained by biopsy or autopsy were fixed in 10% formalin and routinely processed for paraffin-embedding. All of the histological sections were reviewed by 2 of the authors (YH and KA). PT-LPD were classified based on the description of Harris et al.9 Reactive plasmacytic hyperplasia is characterized by mixed proliferation of lymphocytes, plasma cells and rare immunoblasts with no or only minimal cytological atypia. Polymorphic PT-LPD consist of the full range of B-cell maturation from immunoblasts to plasma cells, with small- and medium-sized lymphocytes and numerous cells resembling centrocytes/cleaved follicular center cells. Plasmacytoma-like PT-LPD is composed of a monomorphic population of cytologically atypical plasma cells. Monomorphic PT-LPD, showing sufficient architectural and cytological atypia to be diagnosed as high-grade lymphoma on morphological grounds, were classified according to the revised European American Lymphoma (REAL) classification.15 The term PT-LPD, proposed in Western countries, basically encompasses B-cell proliferative diseases. In the present series, T-cell and NK-cell lymphomas were found and were also classified according to the REAL classification. Clinical findings were available in all cases.

Immunohistochemical staining

The immunoperoxidase procedure (ABC method) was carried out as previously described16 in all cases. The monoclonal antibodies (MAbs) used in our study and their suppliers, reactivities and dilutions are shown in Table I. Before incubation, sections were pretreated with 1% trypsin in Tris buffer (pH 7.8) for CD3ϵ, Ber-H2 and LMP-1 by microwaving for 5 min in citrate buffer (pH 6.0) for CD8, CD16 and CD56 and by microwaving for 5 min in 1 mM EDTA (pH 8.0) for CD4.

Table I. Antibody panel
L-26 (CD20)1:200B-lymphocytesKyowa Medex, Tokyo, Japan
CD3ϵ1:100T-lymphocytesDakopatts, Glostrup, Denmark
UCHL-1 (CD45RO)1:100T-lymphocytesDakopatts
N-CAM (CD56)1:40NK/T-lymphocytesZymed Laboratories, San Francisco, CA
2H7 (CD16)1:200NK/T-lymphocytesNovocastra, Newcastle, UK
TIA-11:500Cytotoxic granulesCoulter, Hialeah, FL
1F6 (CD4)1:200Helper/inducer T-lymphocytesNovocastra
C8/144B (CD8)1:200Suppressor/cytotoxic T-lymphocytesDakopatts
Ber-H2 (CD30)1:10Reed-Sternberg cellsDakopatts
βF-11:50TCR β receptor chainEndogen, Woburn, MA
CSI-4 (LMP-1)1:20Latent membrane proteinDakopatts
PE-2 (EBNA 2)1:10Epstein-Barr virus nuclear antigen 2Dakopatts

In situ hybridization (ISH)

EBV RNA-ISH was performed as previously described.17 As a positive control, the Raji cell line was used. As negative controls, the hybridizing mixture was employed with (a) sense probe and (b) antisense probe after RNase (Sigma) treatment.

DNA extraction for analysis of immunoglobulin heavy chain gene

DNA was extracted from formalin-fixed, paraffin-embedded tissues. Briefly, 25 μm sections were cut and suspended in 40% Chelex-100 resin solution (Bio Rad Laboratories, Hercules, CA) boiled for 10 min and centrifuged for 5 min at 18,000g. The supernatant was used as the DNA template for PCR. Preservation of PCR-amplified products was examined using β2-microglobulin primers (KM29, 5′-GGTTGGCCAATCTACTCCCAGG-3′ and KM38, 5′-TGGTCTCCTTAAACCTGTCTTG-3′). The size of the amplified product was 262 bp.

For rearrangement analysis of immunoglobulin heavy chain gene, a 2-step semi-nested PCR was performed in 6 cases (patient nos. 2, 4, 6, 7, 11 and 12), in which preservation of DNA was confirmed. The primer sequences were 5′-TGG[A/G]TCCG[A/C]CAG[G/C]C-[C/T][C/T]C[A/C/G/T]GG-3′ (termed Fr2A) for the second framework portion of the VH region and 5′-TGAGGAGACGGTGACC-3′ (termed LJH) or 5′-GTGACCAGGGT[A/C/G/T]CTTTGGCCCCCAG-3 (termed VLJH) for the JH region.18 Each PCR experiment included a sample without a DNA template as a negative control, a sample with DNA extracted from the tonsil of a chronic tonsilitis patient as a reactive control and a sample with DNA extracted from a lymphoblastoid cell line as a positive control. The first PCR consisted of 35 cycles with primers Fr2A and LJH and 1 μl of template DNA. Each PCR cycle consisted of denaturation for 15 min at 95°C, annealing for 30 min at 62°C and extension for 1 min at 72°C. The second PCR consisted of 30 cycles with Fr2A and VLJH, using 1 μl of a 1:100 dilution of the first PCR product as a template. Each PCR cycle consisted of denaturation for 15 min at 95°C, annealing for 30 min at 65°C and extension for 1 min at 72°C. The second PCR products were electrophoresed on 2.5% agarose gels and stained with ethidium bromide to visualize the DNA under ultraviolet light.

DNA extraction for detection of HTLV-1 genome

DNA was extracted from paraffin-embedded sections using chelating resin (Sigma, St. Louis, MO). In each case, preservation of the DNA extracted from the lesions was confirmed by amplification with PCR primers specific for a 123-bp segment from exon 7-8 of the β-globin gene (exon 7, 5′-CTTCTGACACAACTGTGTTCACTAGC-3′; exon 8, 5′-TCACCACCAACTTCATCCACGTTCAC-3′).19 For amplification of β-globin, 35 PCR cycles of 94°C/60°C/72°C were performed.20 Twenty-three of 24 cases showed amplified DNA of the expected size by PCR for β-globin.

Amplification of the HTLV-1 proviral genome was performed by PCR. For amplification of HTLV-1 proviral DNA, 40 PCR cycles of 94°C/58°C/72°C were performed with primers designed to amplify a 159-bp segment in the tax region common to the HTLV-1 and HTLV-2 proviral genome (5′-CGGATACCCAGTCTACGTGT-3′, 5′-GAGC-CGATAACGCGTCCATCG-3′)21 and a 119-bp segment in the pol region of the HTLV-1 proviral genome (5′-CTT-CACAGTCTCTACTTGTGC-3′, 5′-CGGCAGTTCTGTGACAGGG-3′).22 The amplified products were separated by electrophoresis in 2% agarose gels containing 2 μg/ml ethidium bromide in Tris-borate EDTA buffer. After electrophoresis, the gels were examined by using an ultraviolet transilluminator. The HTLV-1 proviral DNA PCR products were separated by electrophoresis and transferred on Hybond N+ membranes (Amersham, Buckinghamshire, UK). Oligonucleotide probes that hybridize to each of the intervening sequences between the 2 primers of HTLV-1 proviral genome (tax region: 5′-ACGCCCTACTGGCCACCTGTCC-AGAGCATCAGATCACCTG-3′; pol region: 5′-CCGCAGCTGCACTA-ATGATTGAACTTGAGAAGGAT-3′) were labeled with fluorescein-deoxyuracil triphosphate using a 3′-oligolabeling and detection system (Amersham). Subsequent hybridization and development were performed with the enhanced chemiluminescence detection system, according to the procedures provided by the manufacturer.

Clinical outcome

Clinical outcome was evaluated according to the guidelines of the International Workshop to standardize response criteria for non-Hodgkin's lymphoma.23

Statistical methods

Actuarial survival curves were calculated by using the Kaplan and Meier method,24 and the differences were examined by using the log-rank test to detect significant prognostic factors.25 Factors examined were sex, age (<40, >40 years), type of immunosuppressive agent (cyclosporin A vs. others), primary site of lesion (nodal vs. extranodal), stage of disease (stage I, II vs. III, IV), histology (diffuse large B-cell lymphoma, peripheral T-cell lymphoma, ATL vs. others), immunophenotype of proliferating cells (T vs. B and NK cell), EBER-1 positivity, gastrointestinal involvement and treatment (adjuvant therapy or not). Multivariate analysis was performed by using Cox's proportional-hazards model.26


Clinical findings

Age of patients at the time of transplantation ranged from 17 to 55 (median 33) years and that at diagnosis of PT-LPD ranged from 23 to 56 (median 40) years. The interval between renal transplantation and tumor development ranged from 1 to 264 (median 48) months. Male to female ratio was 5:1. Information regarding hemodialysis history was available in 20 patients, and all received hemodialysis 4 to 226 (median 13) months before renal transplantation. As immunosuppressive agents, cyclosporin A (CyA) was used in 14 patients, FK506 was used in 3 and azathiopurine (AZ) and/or prednisolone in 7. The primary site of PT-LPD was extranodal in 16 cases and nodal in 5. Based on the records of physical examinations, surgical notes and pathological examination of the specimens, the Ann Arbor staging scheme was applied in all cases: 9 patients had stage I, 2 stage II, 5 stage III and 8 stage IV disease.

Histological features and immunophenotypes

The distributions of the histological types and immunophenotypes are summarized in Tables II–IV. Immunohistochemistry revealed that 12 cases were B-cell type, 10 T-cell type and 2 NK/T-cell type (Fig. 1).

Table II. Summary of lymphoproliferative disorders with B-cell phenotype
No.Age/sexHistologyPrimary siteStageTreatment for lymphomaResponseFollow-up (months)
  1. Poly, diffuse polymorphic; PL, plasmacytoma; DLBL, diffuse large B-cell lymphoma; IFN, interferon; CR, complate remission; PD, progressive disease; DID, death due to intercurrent disease; A, alive; DT, death due to tumor.

142/MPolyPara-aortic LNIIINo therapyNo therapy1 DT
245/MPolyGingivaIReduction of immunosuppressive agentsCR37 A
340/MPolyIleumIIINo therapyNo therapy0 DID
427/MPolyUndeterminedIVNo therapyNo therapy2 DT
523/MPL-likeStomachIINo therapyNo therapy0 DID
640/MPL-likeStomachISurgery + IFNαCR42 A
744/FDLBLIleumISurgery + anti-viral agentsCR84 A
840/MDLBLSpleenIVChemotherapyPD2 DT
931/FDLBLBrainIRadiationPD2 DID
1041/MDLBLLungIChemotherapyPD1 DT
1155/MDLBLStomachIVSurgeryPD1 DT
1240/MDLBLInguinal LNIVAnti-viral agentsCR12 A
Table III. Summary of lymphoproliferative diseases with T- and NK-cell phenotype
 No.Age/sexHistologyPrimary siteStageTreatment for lymphomaResponseFollow-up (months)
  1. PT, peripheral T-cell lymphoma, not specified; ATL, adult T-cell leukemia/lymphoma; CR, complete remission; PD, progressive disease; DT, death due to tumor; A, alive.

T-cell1332/MPTLiverIIIChemotherapyPD1 DT
1440/MPTSmall intestineIVChemotherapyPD1 DT
1537/MPTCervical LNIIINo therapyNo therapy1 DT
1644/MPTMuscleIVUnknownPD1 DT
1734/MPTLiverIIReduction of immunosuppressive agentsPD1 DT
1832/MATLUndeterminedIVChemotherapyPD28 DT
1932/MATLInguinal LNIIIUnknownPD27 DT
2043/MATLSkinIChemo- and radiotherapyPD5 DT
2156/MATLSpleenIVChemotherapyPD1 DT
2247/FATLUndeterminedIReduction of immunosuppressive agentsUnknown6 A
NK/T-cell2327/MNasalNasal cavityIChemotherapyCR45 A
2435/MNodalCervical LNIIIChemotherapy and reduction of immunosuppressive agentsCR12 A
Table IV. Immunohistochemical findings of T- and NK-cell phenotype
 No.Age/sexHistologyCD20CD45RoCD3ϵCD4CD8β-F1CD56CD16TIA-1LMP-1EBNA 2
  1. ND, not done; PT, peripheral T-cell lymphoma; ATL, adult T-cell leukemia/lymphoma.

Figure 1.

(a) Peripheral T-cell lymphoma, unspecified, showing a proliferation of large convoluted cells (patient no. 15). (b) Adult T-cell lymphoma with irregular, often convoluted nuclei (patient no. 20). (c) NK/T-cell lymphoma with a polymorphous pattern of proliferation consisting of medium to large cells with irregular nuclei and macrophages (patient no. 23). (d) Tumor cells exhibited a surface staining for CD56 (patient no. 23). (H&E, scale bar = 33 μm)

Clinical outcome

Follow-up period for survivors calculated from the date of initial diagnosis of PT-LPD ranged from 6 to 84 (median 39.5) months. The 1- and 5-year overall survival rates in all patients were 40.0% and 26.7%, respectively, and 5-year event-free survival rate in patients receiving adjuvant therapy was 71.4% (Fig. 2). Half of the B-cell patients and 67% of T-cell patients died within 6 months. Univariate analysis showed that sex (p < 0.05), histology (p < 0.01), clinical stage (p < 0.05) and type of immunosuppressive agents used (p < 0.05) were significant prognostic factors, i.e., female patients, early stage of disease, polymorphous morphology, plasmacytoma or NK/T-cell type morphology and use of CyA showed a more favorable prognosis.

Figure 2.

Five-year overall survival rate in all patients was 26.7%. Event-free five-year survival rate in patients receiving adjuvant therapy was 71.4%. The patients medicated with CyA had a more favorable prognosis than those medicated with other agents.

Multivariate analysis revealed that the use of CyA for immunosuppression at the renal transplantation was a significant independent factor for overall survival (Fig. 2).

Immunoglobulin heavy chain gene analysis

A single band of the expected size (240–260 bp) was detected in all 6 cases examined, indicating the monoclonal nature of the lesions. Histologic types of these cases included polymorphic (2 patients) and plasmacytoma-like (1 patient) PT-LPD and diffuse large B-cell lymphoma (3 patients).

EBV analysis

ISH revealed positive signals in the nucleus of the majority of large cells in 10 of 12 (83%) B-cell cases, and these cases were shown to express LMP-1 and EBNA 2 on immunohistochemical analysis. Three (patient nos. 15, 19, 20) of the 10 cases with T-cell phenotype were EBV- positive (Fig. 3): 2 cases expressed LMP-1 and 3 expressed EBNA-2. Both NK/T-cell lymphomas were also EBV-positive.

Figure 3.

(a) Peripheral T-cell lymphoma cells with cytoplasmic staining for LMP-1 protein (New Fucsine, red) exhibited positive intranuclear signals for EBER-1 (NBT stain, blue-purple) (Scale bar = 33 μm). (b) EBNA-2 protein expression in the proliferating cells of the same case (patient no. 15). (Methylgreen counterstain, scale bar = 50 μm)

HTLV-1 analysis

PCR and Southern blot hybridization analyses for HTLV-1 proviral genome were performed in all T-cell cases, and 5 (patient nos. 18–22) showed positive results (Fig. 4). Four (patient nos. 18–21) of these 5 patients were sero-positive for anti-HTLV-1 antibody at the time of tumor development. The remaining patient (no. 22), who was sero-negative for anti-HTLV-1 antibody at the time of renal transplantation, showed seroconversion 9 months after tumor development. These 5 cases were judged as having ATL. Of these 5 cases, 2 (nos. 19, 20) showed co-infection with HTLV-1 and EBV. Three of the ATL patients were from Kochi (no. 19), Kagoshima (no. 21) and Okinawa prefecture (no. 22), all of which are situated in the ATL-endemic area. The birthplace of another patient (no. 20) was Kagoshima. Two patients (nos. 18, 20) had a history of blood transfusion. Three patients (nos. 19, 20, 22) had a history of hemodialysis with durations of 148, 23 and 33 months, respectively, but no information regarding hemodialysis history was available for the other 2 patients (nos. 18, 21). One patient (no. 18), who had no relation to the HTLV-1 endemic area, received a blood transfusion from an HTLV-1-positive sibling 13 years prior to renal transplantation. HTLV-1 proviral genome was not found in any of 12 B-cell LPD or NK/T-cell lymphoma cases. Serum anti-HTLV-1 antibody was examined in 2 of the B-cell LPD and in 1 of the NK/T-cell lymphoma cases and showed negative results.

Figure 4.

An ethidium bromide-stained agarose gel of the PCR amplification products of the pol region of the HTLV-1 genome. Five cases (patient nos. 18–22) exhibited the 119 bp band. P, positive control (HTLV-1 cell line MT-2); N, negative control (Raji cell line).

Typing of HLA-A, -B

Records on HLA type of the recipients were available for 19 patients (Table V): 4 of 9 patients with B-cell phenotype and 8 of 8 patients with T-cell type had the HLA-A-2 or -11 allele. The difference in frequency of HLA-A-2 or A-11 between cases with B- and T-cell lymphoma was shown to be significant by χ2 test (p < 0.05).

Table V. HLA-A and -B alleles in renal transplant patients developing PT-LPD
No.Age/sexImmunophenotypeHistologyEBER-1 ISHHLA-allele
  1. NA, not available.

340/MBDPo+W24, —w52, w59
427/MBDPo+24, 3151, 61
523/MBPL+9, 11w22, w22
640/FBPLw31, w11w61, w39
744/FBDL+24, 2652, 62
840/MBDL+24, 3151, w54
931/FBDL+w23 + w24, 11w54 + w55, w35
1155/MBDL+2, 2413, 52
1240/MBDL+26, 3135, 51
1440/MTPT2, 955, 52
1537/MTPT+2, 2459, 13
1644/MTPT11, 2639, 39
1734/MTPT2, 1151, 60 or 48
1832/MTATL2, 2452, 63
1932/MTATL+24, 1151, w67
2043/MTATL+2, 2462, 55
2247/MTATL11, 254, 35
2327/MNK/TNasal+25, w1944, 40
2435/MNK/TNodal+24, 2451, 46


In Japan, renal cancer, thyroid cancer and NHL are the most common forms of cancer among renal transplant patients, in contrast to Western countries where cancer of the skin and lips are the most common malignancies.27 The majority of renal transplant patients in Japan receive hemodialysis before renal transplantation. This is in contrast to those in Western countries where dialysis is less frequently employed before transplantation.28 Employment of hemodialysis before renal transplantation in Japan might affect the nature of malignant lymphomas that subsequently develop in these patients.

The present frequency of T-cell lymphoproliferative disorders in the Japanese series was 42% (10/24), which was significantly higher than those reported in Mexico (0%; 0/12), the United States (2%; 1/61) and France (13%; 3/24) (p < 0.05) (Fisher's exact test).11, 29, 30 A review of the international literature revealed 36 cases of T-cell PT-LPD,11 including only 1 case of ATL.30 In the present series, 5 of 10 (50%) cases of T-cell PT-LPD were ATL, i.e., sero-positive for HTLV-1 and HTLV-1 genome in the tumor tissues. CyA, which is commonly used as an immunosuppressive agent in renal transplantation, might cause new infection or activation of latent infection of HTLV-1. However, Natazuka et al.31 reported that CyA suppressed the proliferation of HTLV-1-infected T cells in vitro by inhibition of IL-2 mRNA transcription. Nonetheless, it is important to consider the possibility of HTLV-1 infection in renal transplant patients in Japan as well as in immunocompetent individuals.

HTLV-1 is transmitted mainly via 3 routes: milk-borne transmission from mother to child, sexual transmission from husband to wife and transmission through blood transfusion.32 Previous studies showed a mean incidence of 55% in the United States33 and about 90% in Japanese patients34 undergoing hemodialysis received blood transfusion annually in the 1980s. Sero-positive rates for HTLV-1 among hemodialysis patients in both ATL-endemic (19.7%) and non-endemic areas (3.8%) in Japan were significantly higher than those among healthy individuals, i.e., 3.6% and 0.5%, respectively.35, 36 Hemodialysis before transplantation might provide an opportunity for new infection or activation of viruses. Indeed 2 (nos. 19, 21) of 5 ATL patients in the present series received renal transplantation in ATL-endemic areas. Patient no. 19 had a more than 10-year history of hemodialysis, but clinical records on this point were not available for patient no. 21. The birthplace of 1 patient (no. 20) was in an ATL-endemic area, where he had received hemodialysis for 1 year. Another patient (no. 18) received a blood transfusion 13 years before transplantation from a sibling who was sero-positive for HTLV-1. A screening program of anti-HTLV-1 antibody for blood donors in Japan was introduced in 1986. Thereafter, transfusion-transmitted infection of HTLV-1 has been successfully prevented.32

Seroconversion of the recipients after receiving HTLV-1-containing blood usually occurs within 50 days after transfusion.32 In the immunosuppressive state, HTLV-1 genome-positive ATL without seroconversion might occur37 or seroconversion might be delayed. One of the present patients with HTLV-1 genome in the tumor (no. 22) showed seroconversion at 10 months after renal transplantation.

EBV-associated malignancies are categorized according to the latent gene expression into Latency (Lat) I including Burkitt's lymphoma, Lat II including Hodgkin's disease and nasopharyngeal carcinoma and Lat III including PT-LPD.38 Our B- and T-cell lymphoma cases expressed both LMP-1 and EBNA 2 and were thus categorized as Lat III. Association between EBV and T-cell lymphomas was also reported.39 Co-infection of EBV and HTLV-1 was observed in 2 of the 5 present cases with ATL. A previous study showed that EBV was present within tumor cells in 17% of cases with ATL in Japan.40

Six cases with PT-LPD of the NK/T-cell type have been reported to show an aggressive clinical course, advanced clinical stage and frequent leukemic changes.41–43 The present series included one each of nasal and nodal NK/T-cell lymphoma. Both cases were EBV-associated but showed favorable prognosis.

The cells expressing viral antigens are eliminated primarily by CTL in an MHC class I-restricted manner.39 Among EBV latent infection genes, EBNA-2, 3, 4 and 6 and LMPs can induce efficient CTL responses in a manner restricted to HLA-A-2, -11, B-7, -8, -27 and –44.4, 5, 35 In the present cases, there was a significant difference in the frequency of HLA class I subtype between T- and B-cell cases. All 8 patients with T-cell type had HLA-A-2 or -11, while 44% of B-cell type had these subtypes. The combined frequency of individuals with A-2 and A-11 alleles in the population is rather similar in each district of Japan except in Okinawa where the frequency (40.6%) is slightly (5%) higher than in other districts.44, 45 One of the 8 patients with T-cell type was born in Okinawa. These observations suggested that in the immunodeficient state, HLA class I molecules play a role in the development of B-cell lymphoma.

In conclusion, T-cell lymphoma of ATL type is common among Japanese renal transplant patients, which might be caused by transmission of HTLV-1 via blood transfusion during hemodialysis.


The authors thank Mr. Y. Kabutomori, Ms. C. Tanaka and Ms. T. Watanabe for technical assistance with the immunohistochemical procedures, Drs. S. Sonoda and S. Yashiki (Kagoshima University) for valuable information regarding HLA typing and the following pathologists and physicians for providing patient information: Drs. M. Yamaguchi (Sendai Shakai Hoken Hospital), A. Mukai (Tsukuba University), K. Hamaguchi (Sakura National Hospital), T. Wakabayashi, A. Masunaga (Tokyo University), H. Matsushita, Y. Endo (Toranomon Hospital), M. Naitoh, T. Hasegawa (Niigata University), K. Ohnishi, K. Miura (Hamamatsu Medical School), S. Nakamura (Aichi Cancer Hospital), M. Kuroda (Fujita Health University), S. Murakami (Social Insurance Chukyo Hospital), N. Mori (Nagoya University), A. Okuyama (Osaka University), K. Kageyama (Osaka Medical College), M. Imanishi, T. Uesugi (Kinki University), K. Kitagawa, H. Yasoshima, A. Kubota, K. Uematsu (Hyogo College of Medicine), Y. Ichikawa, T. Kohro (Hyogo Prefectural Nishinomiya Hospital), H. Ushio (Ushio Clinic), T. Matsuno (Okayama University), E. Tahara, H. Yokozaki, S. Matsubayashi (Hiroshima University), T. Okabayashi, T. Horimi, Y. Azuma, K. Iwata (Kochi Prefectural Hospital), T. Kasai, H. Hashimoto (University of Occupational and Environmental Health), C. Yasunaga, K. Yamamoto (Saiseikai Yahata Hospital), T. Hayashi (Miyazaki Prefectural Hospital), H. Yoshida (Kagoshima University) and S. Kojya (Ryukyu University).