Excellent outcome of allogeneic hematopoietic SCT with reduced-intensity conditioning for the treatment of chronic active EBV infection

Article metrics


Since we reported the first successful case of allogeneic hematopoietic SCT (allo-HSCT), we have performed allo-HSCT for 29 patients with chronic active EBV infection (CAEBV), using either myeloablative conditioning (MAC) allo-HSCT (MAST) or reduced-intensity conditioning (RIC) allo-HSCT (RIST). In this retrospective analysis we compared the outcomes after MAST and RIST to identify the optimal conditioning for patients with CAEBV. Of 29 patients, 11 underwent allo-HSCT with MAC, consisting of TBI (12 Gy), etoposide (900 mg/m2) and CY (120 mg/kg) or melphalan (210 mg/m2), and the remaining 18 patients received allo-HSCT after RIC, consisting of fludarabine (180 mg/m2) and melphalan (140 mg/m2) or CY (120 mg/kg), with/without antithymocyte globulin and low-dose irradiation. Donor sources were 8 related BM, 2 related peripheral blood, 5 CD34 selected cells from HLA-haploidentical donors, 8 unrelated BM and 8 unrelated cord blood. The 3-year-EFS rate was 54.5±15.0% for MAST group and 85.0±8.0% for RIST group, and the 3-year OS rate was 54.5±15.0% for MAST group and 95.0±4.9% for RIST group (P=0.016). Allo-HSCT after RIC seems to be a promising approach for the treatment of CAEBV.


EBV is a ubiquitous γ-herpes virus associated with a spectrum of benign and malignant human disease.1, 2 After primary infection, EBV can induce both replicative (productive/lytic) and latent (persistent) infections in lymphocytes. Latent EBV infection is linked to the development of a variety of lymphoproliferative diseases (LPDs), such as B-cell LPD and T-cell/natural killer cell (T/NK cell) LPD.3, 4, 5 EBV-associated B-cell LPD is a well-established phenomenon in patients with primary/secondary immunodeficiencies and in recipients of allogeneic transplants such as hematopoietic stem cell, heart, lung, liver, kidney and small intestine. These lymphoid proliferations, arising in the context of impaired T cell-mediated immunity, range from reactive polyclonal B-cell hyperplasias without cytogenetic abnormalities to monoclonal malignant lymphomas.6

In contrast to B-cell LPD, EBV-infected T/NK-cell LPD is usually observed in apparently immunocompetent persons. In the late 1980s, three different groups described T-cell lymphomas and T-cell LPD containing EBV DNA in patients with chronic active EBV infection (CAEBV).7, 8, 9 Similarly, in 1989, we unexpectedly found evidence of EBV-infected NK-cell LPD in five of seven studied patients with LPD of granular lymphocytes.10 Since that time, a spectrum of EBV-associated T/NK-cell LPDs have been increasingly recognized and more clearly defined: CAEBV, EBV-associated hemophagocytic lymphohistiocytosis, extranodal NK/T-cell lymphoma, nasal type, hypersensitivity to mosquito bites, hydroa vacciniform and aggressive NK-cell leukemia. In spite of the heterogeneity of the clinical, phenotypic and genotypic characteristics of these EBV+ T/NK-cell lymphoproliferations, many clinical and histological features are shared: (1) fever and hepatomegaly/splenomegaly; (2) extranodal lesions such as nose, skin and digestive tract with a propensity for extra-nodal dissemination; (3) a high incidence of hemophagocytic syndrome; (4) histopathological features of pleomorphism and angiocentricity/angioinvasion; (5) a mature phenotype of EBV-infected cells; (6) refractoriness to conventional treatments, and (7) a higher prevalence in Asia and Central America.

CAEBV is one of the representative disease among EBV-associated T/NK-cell LPDs, with recently proposed diagnostic guidelines as follows: (1) persistent or recurrent infectious mononucleosis-like symptoms such as fever, swelling of lymph nodes and hepatosplenomegaly; (2) unusual pattern of anti-EBV antibodies with raised anti-viral capsid antigen and anti-early antigen, and/or evidence of increased EBV genome number in affected tissues, including the peripheral blood; and (3) chronic illness that cannot be explained by other known disease processes at diagnosis.11

CAEBV occurs predominantly in children and young adults, but older adults with this disease are increasingly recognized. Interestingly, one-fourth of pediatric patients had a history of hypersensitivity to mosquito bites, and most of such cases are NK-cell type and rarely γδT-cell type.12, 13, 14

The prognosis of patients with CAEBV is very poor in both pediatric and adult patients. In the absence of effective therapy, almost all patients will die within 5–15 years from onset because of hepatic or cardiac failure, hemophagocytic syndrome, malignant lymphoma, opportunistic infections or intracranial/gastrointestinal bleeding.12, 13

To date, investigational therapies for CAEBV have comprised immunoregulatory drugs and antiviral agents, all resulting in disappointing outcomes. We sought to investigate, 10 years ago, a new therapeutic algorithm for CAEBV comprising sequential immunochemotherapy, combination chemotherapy and allogeneic hematopoietic SCT (allo-HSCT) in an attempt to reduce and/or eliminate EBV-infected T/NK cells. Because complete responses are rare with a correspondingly high relapse risk, we consider allo-HSCT for all patients when evidence of residual disease exists.

Since the first case report of successful allo-HSCT,15 we have performed allo-HSCT for 29 patients with CAEBV, using either myeloablative conditioning (MAC) allo-HSCT (MAST) or reduced-intensity conditioning (RIC) allo-HSCT (RIST). In this study, we have compared the outcomes after MAST and RIST to identify the optimal conditioning for patients with CAEBV.

Patients and methods


In all, 29 patients fulfilling the diagnostic criteria of CAEBV and treated with allo-HSCT between August 1997 and December 2008 were enrolled in this study. During the same period, 6 other patients with EBV-associated T/NK-cell LPDs (2 EBV-hemophagocytic lymphohistiocytosis, 2 lymphoma, 1 hypersensitivity to mosquito bites and 1 hydroa vacciniform) were also transplanted.

Analysis of EBV-infected cells

All test samples were collected with the informed consent of the patients or their parents. PBMCs were separated into CD3+, CD4+, CD8+, CD19+ and CD56+ cells using Dynabeads. Patients were defined as having T-cell-type infection when CD3+ cells were the major group of cells that harbored EBV. Patients were defined as having NK-cell-type infection when their CD56+ cells were the major group of cells infected with EBV. In case of T/NK-cell type, both CD3+ cells and CD56+ cells were the major group of cells that harbored EBV. The presence of EBV was detected using qualitative PCR analysis as reported.13, 16 The clonality of EBV-infected cell populations was determined using Southern blotting with an EBV terminal repeat probe as well as rearranged patterns of the Ig heavy chain gene and the T-cell receptor gene as reported.10

Peripheral blood viral load

EBV copy number in PBMCs, plasma or whole peripheral blood was measured using real-time quantitative PCR in consecutive samples obtained from each patient as previously described.13

Treatment strategy before allo-HSCT

All patients received immunochemotherapy consisting of prednisolone 1–2 mg/kg/day, etoposide 150 mg/m2/week and cyclosporine 3 mg/kg/day for 1 to 2 months to control disease symptoms as the first step of therapy. Second step consisted of additional chemotherapy agents, combined with cyclosporine, as follows: modified CHOP regimen (CY, pirarubicin hydrochloride, VCR and prednisolone), sequential high-dose ara-C (HDCA; ara-C 3 g/m2 every 12 h × 4 doses, L-asparaginase 6000 U/m2 × 1 dose), HDCA (ara-C 1.5 g/m2 every 12 h for 6 days) and VPL (VP-16 150 mg/m2 day 1, prednisolone 30 mg/m2 days 1–7, L-asparaginase 6000 U/m2 days 1–7). All patients received at least one of these regimens to reduce/eliminate EBV-infected T/NK cells. In the event of a <1 log reduction of EBV load, a further cycle was repeated or a different combination chemotherapy regimen was instituted before proceeding to allo-HSCT (Table 1).

Table 1 Treatment strategy for chronic active EBV infection

Definition of disease status before allo-HSCT

Disease status before allo-HSCT was assessed based on clinical features and EBV load and consequently classified as either active or non-active. Active disease was defined by the existence of symptoms and signs such as fever, persistent hepatitis, significant lymphoadenopathy, hepatosplenomegaly, pancytopenia and/or progressive skin lesions alongside an elevated EBV load in the peripheral blood. When disease condition persisted in active status during chemotherapy, HSCT was planned as soon as possible before developing to a fulminant clinical course.

Conditioning regimen, GVHD prophylaxis and TRM

Conditioning regimens were assigned as myeloablative or reduced intensity as follows. MAC consisted of TBI (12 Gy in six fractions), etoposide (900 mg/m2 × 1 dose) and CY (120 mg/kg in 2 doses) or melphalan (210 mg/m2 in 2 to 3 doses), and RIC included fludarabine (180 mg/m2 in 5 to 6 doses) and melphalan (140 mg/m2 in 2 doses) or CY (120 mg/kg in 2 doses), with/without antithymocyte globulin and low-dose irradiation.

The application of GVHD prophylaxis was variable because of the different stem cell sources and degree of HLA disparity. Cyclosporine (3 mg/kg daily by continuous infusion) was used for HLA-matched related BMT, and tacrolimus (0.02 mg/kg daily by continuous infusion) was used for CD34-positive cell transplant (CD34). In case of HLA-matched unrelated BMT and HLA-mismatched unrelated cord blood transplant (UCBT), tacrolimus and a short course of MTX (7.5 mg/m2 on days 1, 3 and 6) were used. In other situations, GVHD prophylaxis was intensified by adding anti-T lymphocyte globulin or antithymocyte globulin (ATG, horse). TRM was defined as any death that occurred while the patient was in remission.

Statistical analysis

Statistical analysis was performed using the SPSS program (SPSS ver.14; SPSS Inc., Chicago, IL, USA), and survival after allo-HSCT was estimated using the Kaplan–Meier method.


In all, 13 patients were defined as having T-cell type, 13 patients had NK-cell type and 3 were classified as T/NK-cell type. Of the 29 patients, 11 received MAC and 18 patients received RIC (Table 2).

Table 2 Patient characteristics

Clinical features and outcomes of MAST group and RIST group

The clinical features and the post transplant clinical courses of the 11 patients in the MAST group and 18 patients in the RIST group are summarized in Tables 3 and 4, respectively. The median age at onset was 8 years in the MAST group compared with 14 years in the RIST group. This difference is mainly accounted for by the wide distribution of patients' age in the RIST group. Patients with active disease status before the preconditioning, thought to be a risk factor for poor outcome, comprised 36% in the MAST group and 34% in the RIST group.

Table 3 Clinical characteristics of 11 patients with CAEBV treated with MAST
Table 4 Clinical characteristics of 18 patients with CAEBV treated with RIST

Peripheral blood EBV load (whole blood) in tested patients ranged from 103 to 107 copies/mL (<2 × 102 copies/mL in healthy volunteers).

Time from diagnosis to HSCT ranged from 4 months to 11 years (median 3 years) in the MAST group and from 5 months to 19 years (median 14.5 months) in the RIST group. Stem cell sources comprised 3 CD34, 2 related BM and 6 unrelated BM in the MAST group, and 4 related BM, 8 unrelated cord blood, 2 CD34, 2 related peripheral blood and 2 unrelated BM in the RIST group.

Acute GVHD (aGVHD, grades II to IV) was observed in 6/9 and chronic GVHD (cGVHD) was observed in 2 patients; one extensive type and one limited type in the MAST group. Similarly, in the RIST group, aGVHD (II to IV) was observed in 11/16, and 3 patients (3/17) developed cGVHD; one extensive type and two limited type.

The incidence of TRM in the MAST group was 5/11 (45%), accounted for by 2 deaths with veno-occlusive disease, 1 infection-associated hemophagocytic syndrome, 1 renal failure (BK virus) and 1 GVHD. Six patients were in continuous CR from 69 months to 134 months, corresponding to a 3-year EFS and OS of 54.5±15.0% for both outcomes. In contrast, the incidence of TRM in the RIST group was 1/18 (5.6%), because of a subarachnoid hemorrhage. Two patients (RIC-6 and RIC-9) experienced graft failure and mixed chimerism, respectively, who both subsequently received a successful second RIST (Table 5). A third RIC allo-HSCT was performed for a patient (RIC-16) who had failed engraftment twice at a previous hospital where RIC allo-HSCT was conducted without chemotherapy before the transplant preconditioning. The 3-year EFS for the RIST group was 85.0±8.0% and the 3-year OS was 95.0±4.9% (Figure 1).

Table 5 Feasibility of second or third transplant with RIC
Figure 1

EFS and OS for 11 patients with MAC allo-HSCT and 18 patients with RIC allo-HSCT. MAC=myeloablative conditioning; RIC=reduced-intensity conditioning.


CAEBV is no longer an enigmatic disease, but rather can be defined as one of the representative EBV-associated T/NK-cell LPDs according to our current diagnostic criteria and the published experience from Japan.11 Although CAEBV is a high-mortality, high-morbidity disease with life-threatening complications, a standard treatment approach has not been established. However, sequential immunochemotherapy, combination chemotherapy followed by HSCT have been reported to result in continuous CR in patients with CAEBV. Reduction and elimination of EBV-infected T/NK cells is likely to be essential to achieve long-term remission of EBV-associated T/NK-cell LPD, and it is clearly important to distinguish this entity from the well-described EBV-associated B-cell LPD.16, 17

Our therapeutic strategy consists of immunochemotherapy, combination chemotherapy and HSCT. Step 1 is crucial to render the disease inactive, which is achieved by targeting macrophages and suppressing activated T/NK cells and the associated hypercytokinemia. This therapeutic approach has recently become a gold standard for the treatment of hemophagocytic syndrome/hemophagocytic lymphohistiocytosis.18 Step 2 comprises combination chemotherapy regimens such as CHOP, sequential HDCA (the Capizzi regimen), HDCA and VPL, which is intended to eradicate EBV-infected T/NK cells. Although our experience is that HDCA is most effective in reducing EBV DNA load, the efficacy of chemotherapy, as judged by clinical features and degree of liver dysfunction, was similar among the different combination chemotherapies. By using this approach we achieved a non-active disease status before allograft conditioning in 64% of the MAST group and 66% of the RIST group. In a small number of patients whose EBV load became undetectable after step 2, a durable CR without allogeneic transplantation was achieved. Thus far, two patients with hypersensitivity to mosquito bites (1 NK type and 1 γδT type) who received HDCA have remained in CR for >6 years.19 In addition, we have succeeded to treat one more patient who developed EBV-infected T-cell LPD after cardiac transplant (sustained CR for >2 years). Therefore, more effective treatment strategy, including new combination chemotherapy as well as other approaches, remains to be established. The use of EBV-specific cytotoxic T cells might be considered in step 2.6

In this study the median duration from disease onset to HSCT was shorter in RIST group (14.5 months) than in MAST group (3 years). This may potentially confer a survival advantage to the RIST group, as a longer interval between disease onset and HSCT is associated with more frequent life-threatening complications and comorbidities.

As step 3, HSCT is required to eradicate any residual infected cells and to reconstitute normal immunity. There have been several encouraging reports of MAST. But the number of patients in each report was small, and reducing the incidence of life-threatening complications of MAST seems to be the most important issue to be solved. Over the past decade, a range of RIC protocols have been designed with the aim of reducing toxicity while exploiting the graft-versus-tumor effect.20 Our first patient with CAEBV who underwent RIST instead of MAST was a 31-year-old woman. After receiving chemotherapy consisting of CHOP, sequential HDCA and HDCA, a reduction in the peripheral blood EBV DNA load reached the lower threshold of detection. We then performed related BMT using RIC consisting of fludarabine and melphalan from an HLA-identical sibling donor. Her post transplant clinical course was uneventful, and she menstruated again 3 months after HSCT.21 This case encouraged us to continue RIST with fludarabine and melphalan for the treatment of CAEBV.

There was no significant difference in aGVHD (P=0.915) and cGVHD (P=0.835) between MAST group (aGVHD; 6/9, cGVHD; 2/9) and RIST group (aGVHD; 11/16, cGVHD; 3/17). Although the proportion of patients with active and non-active disease status before the preconditioning was similar in the two groups, a higher incidence of TRM was observed in MAST group (45%) and a lower incidence in RIST group (5.6%), which is consistent with two recent reports.22, 23 Gotoh et al.22 reported 15 patients with CAEBV who received HSCT (5 MAST and 10 RIST). In this study, a high incidence of TRM in the MAST group (3/5) and a lower incidence in the RIST group (1/10) was described. In contrast to our data, however, a relatively high relapse rate was observed in their RIST group compared with our cohort (30 vs 11%). Although data of preceding treatments before the preconditioning were not stated in their paper, it may be that our strategy, including immunochemotherapy and combination chemotherapy before the preconditioning, may contribute to the superior outcomes. Sato et al. 23 collected 42 cases of CAEBV who underwent allo-HSCT (31 MAST and 11 RIST) between March 1997 and October 2003 in Japan, and their EFS was 56%. These two reports suggest that a longer interval between disease onset and HSCT is associated with a higher mortality rate, consistent with the natural history of CAEBV patients who tend to have more life-threatening complications and comorbidities as their clinical course progresses.12

Recipients of allo-HSCT sometimes show an increased EBV load, which may progress to EBV-associated LPD. These complications have been linked to (1) ex vivo or in vivo T-cell depletion, (2) unrelated donor, (3) HLA-mismatched donor, (4) antithymocyte/lymphocyte globulin, (5) RIC and (6) UCBT.24 In this study, two-thirds of the RIST group had these high risk factors. Recently, UCBT has become a valuable alternative for patients who require HSCT but lack a suitable donor, and 8 UCBTs were included in our RIST group. Brunstein et al.25 recently reported a high incidence of EBV-associated B-cell LPD after UCBT when RIC and ATG were combined (ATG+ 21% vs ATG− 2%). Although six of the eight UCBTs in the present study were treated with RIC and ATG/anti-T lymphocyte globulin, none developed B-cell LPD. Possible explanations for the difference include: (1) an ATG/anti-T lymphocyte globulin dose equivalent to 1/2 to 1/4 of standard doses and (2) careful monitoring of EBV load and EBV Ab titers during and after UCBT. In the case of UCBT, the donor cells are assumed to be EBV negative. Therefore, EBV infection after UCBT constitutes a reactivation of the endogenous EBV strain or a new primary infection with an exogenous EBV strain. We have assessed 57 recipients of 78 UCBTs undertaken before 2008 to determine EBV serology. Among the 57 patients, 46 were EBV seropositive before UCBT. After UCBT, 19 of the 46 seropositive recipients became EBV seronegative. After conversion to EBV seronegativity, RIC-7 experienced re-infection of EBV 10 months later. Subsequently, RIC-7 developed EBV-infected NK-LPD of donor origin. Thus, EBV infection after UCBT can be considered of two types: endogenous transmission and exogenous re-infection. Serial monitoring of the EBV load and EBV serologic evaluation are required after UCBT for the early prediction and treatment of active EBV infection.

As shown in Table 5, RIC-6 relapsed from mixed chimerism and RIC-9 showed autorecovery, resulting in relapse. We performed a second transplant with RIC 4 months and 6 months later, respectively, and succeeded in achieving CR. RIC-16 had received two previous RIST without chemotherapy before the preconditioning, resulting in autologous reconstitution. She was then referred to our institute for consideration of a third transplant. After administration of step 1 and 2 protocol, a third UCBT with RIST was undertaken, resulting in a sustained CR.

In conclusion, 29 patients with CAEBV were transplanted using a variety of stem cell sources. In all, 6 of the 11 patients with MAST and 15 of the 18 patients with RIST were successfully treated and continue in remission. In addition, 2 patients of the 18 RIST group failed to engraft, but were successfully treated with a second RIST. One of the 18 RIST patients received a successful third RIST after failing RIC allo-HSCT twice at a previous hospital. OS was 54.5% (6/11) in the MAST group and 95% (17/18) in the RIST group. The excellent outcomes in the RIST group may be explained by (1) early intervention after the disease onset, (2) low incidence of TRM, (3) combination chemotherapy before the transplant preconditioning to reduce/eliminate EBV-infected T/NK cells and (4) regular monitoring of EBV load and serologic evaluation after HSCT to predict and treat active EBV infection.


  1. 1

    Kawa K . Epstein-Barr virus-associated diseases in humans. Int J Hematol 2000; 71: 108–117.

  2. 2

    Cohen JI . Epstein-Barr virus infection. N Engl J Med 2000; 343: 481–492.

  3. 3

    Kawa K . Diagnosis and treatment of Epstein-Barr virus-associated natural killer cell lymphoproliferative disease. Int J Hematol 2003; 78: 24–31.

  4. 4

    Kimura H, Hoshino Y, Hara S, Sugaya N, Kawada J, Shibata Y et al. Differences between T cell-type and natural killer cell-type chronic active Epstein-Barr virus infection. J Infect Dis 2005; 191: 531–539.

  5. 5

    Oshima K, Kimura H, Yoshino T, Kim CW, Ko YH, Lee SS et al. Proposed categorization of pathological states of EBV-associated T/natural killer-cell lymphoproliferative disorder (LPD) in children and young adults: overlap with chronic active EBV infection and infantile fulminant EBV T-LPD. Pathol Int 2008; 58: 209–217.

  6. 6

    Heslop H . How I treat EBV lymphoproliferation. Blood 2009; 114: 4002–4008.

  7. 7

    Jones JF, Shurin S, Abramowsky C, Tubbs R, Sciotto C, Wahl R et al. T-cell lymphomas containing Epstein-Barr viral DNA in patients with chronic Epstein Barr virus infections. N Engl J Med 1988; 318: 733–741.

  8. 8

    Kikuta H, Taguchi Y, Tomizawa K, Kojima K, Kawamura N, Ishizaka A et al. Epstein-Barr virus genome-positive T lymphocytes in a boy with chronic active EBV infection associated with Kawasaki-like disease. Nature 1988; 333: 455–457.

  9. 9

    Ishihara S, Tawa A, Yumura-Yagi K, Murata M, Hara J, Yabuuchi H et al. Clonal T-cell lymphoproliferation containing Epstein-Barr (EB) virus DNA in a patient with chronic active EB virus infection. Jpn J Cancer Res 1989; 80: 99–101.

  10. 10

    Kawa-Ha K, Ishihara S, Ninomiya T, Yumura-Yagi K, Hara J, Murayama F et al. CD3-negative lymphoproliferative disease of granular lymphocytes containing Epstein-Barr viral DNA. J Clin Invest 1989; 84: 52–55.

  11. 11

    Okano M, Kawa K, Kimura H, Yachie A, Wakiguchi H, Maeda A et al. Proposed guidelines for diagnosing chronic active Epstein-Barr virus infection. Am J Hematol 2005; 80: 64–69.

  12. 12

    Ishihara S, Okada S, Wakiguchi H, Kurashige T, Morishima T, Kawa-Ha K . Chronic active Epstein-Barr virus infection in children in Japan. Acta Paediatr 1995; 84: 1271–1275.

  13. 13

    Kimura H, Hoshino Y, Kanegane H, Tsuge I, Okamura T, Kawa K et al. Clinical and virologic characteristics of chronic active Epstein-Barr virus infection. Blood 2001; 98: 280–286.

  14. 14

    Kawa K, Okamura T, Yagi K, Takeuchi M, Nakayama M, Inoue M . Mosquito allergy and Epstein-Barr virus-associated T/natural killer-cell lymphoproliferative disease. Blood 2001; 98: 3173–3174.

  15. 15

    Okamura T, Hatsukawa Y, Arai H, Inoue M, Kawa K . Blood stem-cell transplantation for chronic active Epstein-Barr virus with lymphoproliferation. Lancet 2000; 356: 223–224.

  16. 16

    Okamura T, Kishimoto T, Inoue M, Honda M, Yamashita N, Wakiguchi H et al. Unrelated bone marrow transplantation for Epstein-Barr virus-associated T/NK-cell lymphoproliferative disease. Bone Marrow Transplant 2003; 31: 105–111.

  17. 17

    Kawa K, Okamura T, Yasui M, Sato E, Inoue M . Allogeneic hematopoietic stem cell transplantation for Epstein-Barr virus-associated T/NK-cell lymphoproliferative disease. Crit Rev Oncol Hematol 2002; 44: 251–257.

  18. 18

    Henter JI, Horne A, Arico M, Egeler R, Filipovich A, Imashuku S et al. HLH-2004: diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007; 48: 124–131.

  19. 19

    Koyama M, Takeshita Y, Sakata A, Sawada A, Yasui M, Okamura T et al. Cytotoxic chemotherapy successfully induced durable complete remission in 2 patients with mosquito allergy resulting from Epstein-Barr virus-associated T-/natural killer cell lymphoproliferative disease. Int J Hematol 2005; 82: 437–440.

  20. 20

    Shimoni A, Hardan I, Shem-Tov N, Rand A, Herscovici C, Yerushalmi R et al. Comparison between two fludarabine-based reduced-intensity conditioning regimens before allogeneic hematopoietic stem-cell transplantation: fludarabine/melphalan is associated with higher incidence of acute graft-versus-host disease and non-relapse mortality and lower incidence of relapse than fludarabine/busulfan. Leukemia 2007; 21: 2109–2116.

  21. 21

    Sakata N, Sato E, Sawada A, Yasui M, Inoue M, Kawa K . Chronic active Epstein-Barr virus infection treated with reduced intensity stem cell transplantation. Rinsho Ketsueki 2004; 45: 393–396.

  22. 22

    Gotoh K, Ito Y, Shibata-Watanabe Y, Kawada J, Takahashi Y, Yagasaki H et al. Clinical and virological characteristics of 15 patients with chronic active Epstein-Barr virus infection treated with hematopoietic stem cell transplantation. Clin Infect Dis 2008; 46: 1525–1534.

  23. 23

    Sato E, Ohga S, Kuroda H, Yoshiba F, Nishimura M, Nagasawa M et al. Allogeneic hematopoietic stem cell transplantation for Epstein-Barr virus-associated T/natural killer-cell lymphoproliferative disease in Japan. Am J Hematol 2008; 83: 721–727.

  24. 24

    Kawa K, Sawada A, Koyama M, Inoue M . Epstein-Barr virus infection after unrelated cord blood transplantation: reactivation or reinfection? Int J Hematol 2007; 85: 267–269.

  25. 25

    Brunstein CG, Weisdorf DJ, Defor T, Barker J, Tolar J, van Burik J et al. Marked increased risk of Epstein-Barr virus-related complications with the addition of antithymocyte globulin to a nonmyeloablative conditioning prior to unrelated umbilical cord blood transplantation. Blood 2006; 108: 2874–2880.

Download references


We are grateful to Dr Christopher P Fox for his advice and critical reading of the paper. We thank all patients and staffs included in this study for their participation.

Author information

Correspondence to K Kawa.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article


  • EBV
  • T/NK-cell LPD
  • allogeneic HSCT
  • MAST
  • RIST

Further reading