Acute Leukemias

Small number of HTLV-1-positive cells frequently remains during complete remission after allogeneic hematopoietic stem cell transplantation that are heterogeneous in origin among cases with adult T-cell leukemia/lymphoma

Article metrics


Allogeneic hematopoietic stem cell transplantation (allo-HSCT) can provide long-term remission for patients with adult T-cell leukemia/lymphoma (ATLL) caused by human retrovirus, human T-lymphocyte virus (HTLV-1). To understand how HTLV-1-positive cells including ATLL cells were suppressed by allo-HSCT, we examined HTLV-1 provirus load and residual ATLL cells in peripheral blood of transplant recipients using PCR-based tests. We found that the copy number of HTLV-1 genome, called provirus, became very small in number after allo-HSCT; however, in most cases, provirus did not disappear even among long-term survivors. Tumor-specific PCR tests demonstrated that most of HTLV-1-positive cells that remained long after transplantation were not primary ATLL cells but donor-derived HTLV-1-positive cells. We also found a case having very low amount of residual disease in peripheral blood even long after transplantation. There was only one recipient in whom we failed to show the presence of HTLV-1 genome and antibody against HTLV-1 even with an extensive search, which strongly suggested the elimination of HTLV-1 after allo-HSCT. These results demonstrated that after allo-HSCT the small amount of residual HTLV-1-positive cells were heterogeneous in origin and that long-term disease control for ATLL could be obtained without the complete elimination of HTLV-1.


Adult T-cell leukemia/lymphoma (ATLL) is a peripheral T-cell lymphoma caused by a retrovirus, human T-lymphocyte virus (HTLV-1), which randomly integrates into the genome of infected T cells.1, 2, 3 The HTLV-1 genome in T cells, called provirus, has been utilized for the diagnosis of the disease caused by or the carrier state of HTLV-1. For example, Southern blot analysis of HTLV-1, when it demonstrates a monoclonal proliferation of cells infected with HTLV-1, provides the strongest evidence for the diagnosis of ATLL.4 Southern blot analysis usually detects a monoclonal population composed of 3–5% of total cells, which is generally enough to diagnose ATLL. On the other hand, polymerase chain reaction (PCR)-based tests detect HTLV-1 genome with much higher sensitivity than Southern blot analysis, allowing us to monitor a small amount of HTLV-1 provirus load.5, 6

The clinical course of ATLL widely differs by clinical subtypes (acute, lymphoma, chronic and smoldering). The prognoses of acute and lymphoma types are very poor when treated with conventional or even high-dose chemotherapy;7, 8 however, with allogeneic hematopoietic stem cell transplantation (allo-HSCT), a long-term clinical remission (CR) is achievable as reported from several groups including ours.9, 10, 11 For example, among cases with acute ATLL, allo-HSCT reduced the volume of tumor cells in the peripheral blood to undetectable level when tested by morphological examination or Southern blot analysis, suggesting that the reduction of ATLL cells was less than 5% of WBC, as we reported previously.11

In this study, as an extension of our previous report, to understand how small the population of HTLV-1-positive cells would become after allo-HSCT and to test whether HTLV-1 could be eradicated, we investigated HTLV-1 provirus load and the minimum residual disease (MRD) in 22 cases of ATLL using PCR-based gene amplification. Since PCR for HTLV-1 provirus picked up not only ATLL cells, but also all cells infected with HTLV-1, including polyclonal non-ATLL cells, we introduced a specific PCR method to detect ATLL cells utilizing a unique integration site of HTLV-1 in each ATLL case.

We found that cells carrying HTLV-1 existed at the very low level in peripheral blood of long-term survivors after allo-HSCT. Most of them were donor-derived cells, but MRD was simultaneously present only in one case. We also experienced a single case in which anti-HTLV-1 antibodies became negative with no HLTV-1 genome amplified with PCR-based tests, suggesting the eradication of HTLV-1.

Patients and methods

Clinical features of patients with ATLL

The diagnosis and classification of ATLL was based on the criteria proposed by the Lymphoma Study Group of Japan.12 Twenty-two patients with the diagnosis of acute or lymphoma type ATLL who received allo-HSCT in three hospitals in Nagasaki, an endemic area of HTLV-1 in Japan, between September 1997 and May 2004 were included in this study.

Table 1 summarizes the clinical characteristics of these patients. Median age of the patients was 48 years. In 21 of all 22 cases, donor-derived hematopoiesis was obtained (Table 2). Only one patient (case 21) did not achieve CR after allo-HSCT and seven patients experienced a relapse of ATLL. At the time of analysis, 11 patients were alive and nine of these patients remained in CR.

Table 1 Characteristics of patients and transplantation
Table 2 Results of transplantation

Quantitative measurement of HTLV-I provirus load in peripheral blood

Peripheral blood samples were collected from the patients after they gave a written informed consent. Genomic DNA was extracted from mononuclear cells (MNC) of peripheral blood using the QIAGEN DNA Midi Kit (QIAGEN, Hiden, Germany) and from paraffin-embedded sample using DEXPAT (TAKARA BIO INC, Shiga, Japan). Quantitative measurement of HTLV-1 provirus was performed with real-time quantitative PCR (RQ-PCR) using the LightCycler System and DNA Master Syber Green I (Roche diagnostics, Mannheim, Germany) as reported previously.13 In brief, 30 ng of genomic DNA was used as a template and the copy number of HTLV-1 provirus was assessed by the ratio of the amount of tax region of HTLV-1 and that of beta globin gene (tax copies/MNC=2 × copy number of tax/copy number of beta-globin gene). The mean value of two experiments was shown as the copy number of HTLV-1 provirus load. Figure 1 shows the correlation between the ratios of the positive control plasmid containing tax region in the irrelevant plasmids and the results of RQ-PCR tests in a log-scale graph. A statistically significant correlation was found (r=0.89, P<0.001). This system could quantify one copy of the tax gene in 5000 cells.

Figure 1

Correlation of the ratio of tax copy number between control plasmid and the quantification using RQ-PCR. Control plasmids containing the tax region of HTLV-1 were serially diluted with plasmids containing irrelevant sequence (beta-globin) and the ratio of target plasmid was quantified using the RQ-PCR method.

Detection of primary ATLL cells with inverse PCR

To detect the residual ATLL cells, we performed an inverse PCR as reported by Takemoto et al.14 that amplified the integration site of HTLV-1 in the genome of tumor cells whose sequence was then utilized to establish case-specific PCR primers that amplified a part of HTLV-1 (LTR) and the franking region. Each PCR in this study could at least detect one primary ATLL cell among 10 000 normal cells. PCR condition and the DNA sequence of the primer sets in nine cases tested are available upon request.

Colony formation and the expansion of HTLV-I-infected cells to test the origin of those cells

Previously, we established a method to clonally amplify HTLV-I-infected cells.15 In brief, MNC in the peripheral blood were cultured in semisolid media containing 0.93% methylcellulose dissolved in Iscove’s modified Dulbecco’s medium (IMDM) supplemented with 20% fetal calf serum (FCS) and 200 ng/ml of recombinant human interleukin (rhIL)-2 (TECHNE Corp., Minneapolis, MN, USA). After three weeks of culture, each colony grown in the semisolid media was picked up individually and transferred to liquid culture (IMDM with 20% FCS and 20 ng/ml of rhIL-2) for clonal expansion. All cell culture was performed at 37°C with 5% CO2. The origin of cells (donor or recipient) was assessed by means of sex mismatch (using Y chromosome specific SRY gene detection) or the difference of the number in short tandem repeat (STR method).


Quantitative measurement of HTLV-I provirus after allo-HSCT

A total of 86 samples in 22 patients were collected; samples per patient were from 1 to 10 (median 3.5 samples) with median sampling time of 6 months from transplant (0.5 month to 8.3 years). The copy numbers of HTLV-1 provirus in each case are shown in Figure 2a and b. Most of the samples contained a low amount of HTLV-1 provirus, except for two conditions: (1) transplantation from a carrier donor and (2) right before (about 2 weeks) or after the clinical relapse of ATLL. In 22 samples transplanted from carrier donors, the provirus load was always 500 copies/105 cells or more despite the clinical disease status at sampling. The average copy number of HTLV-1 was significantly higher in patients transplanted from a carrier donor than from a noncarrier donor (mean value, 15 000 and 760 copy/105 cells, respectively, P<0.0001).

Figure 2

Quantification of HTLV-1 provirus load in the peripheral blood of recipients. Case number is on the left side of the figures. Case number with plus mark represents transplantation from a carrier donor. Copy number of provirus is shown as a gray or white box: three gray boxes represent virus load 10−2; two gray boxes, 10−210−3; one gray box, 10−310−5; white box, below detection level. Time after transplantation is described as month (m) or year (y). Cross mark represents death of the case and arrow indicates the time of relapse of ATLL. Cases treated with myeloablative conditioning are shown in (a) and those received RIST are in (b).

Within 6 months from transplantation, the provirus load became undetectable at least once in eight out of 15 cases (case numbers 2, 3, 4, 10, 12, 16, 17 and 18). However, in all seven cases tested later, the copy number of HTLV-1 provirus became detectable again. At the time of the last follow-up, provirus load was below the detection level in only two cases (case numbers 16 and 17). The provirus load during the early period following transplantation was not related to the type of conditioning regimen, disease status before the transplantation or the duration of survival. There was no statistically significant association between provirus loads and the development of severe acute GVHD (data not shown). No specific pattern in the kinetics of virus load was noticed among long-term survivors or among patients that experienced relapse.

Analysis of MRD in the peripheral blood

As a low level of HTLV-1 provirus load was detected in the peripheral blood of most patients, we tested whether primary ATLL cells remained as MRD using specific PCR for primary ATLL cells, which amplified a unique franking genomic region of the HTLV-1 integration site in each case. In nine cases (cases 1, 5, 9, 10, 15, 18, 19, 21 and 22), 34 samples were analyzed with this method (Table 3 and Figure 3). Although the sensitivity of the inverse PCR varied from case to case, the amount of MRD that could be detected by this method was always below the provirus load quantified by RQ-PCR in every sample (data not shown).

Table 3 DNA sequence of the franking region of HTLV-1 integration site
Figure 3

MRD of ATLL after transplantation. MRD of ATLL was assessed using case-specific inverse PCR method. Results of the inverse PCR are shown under the boxes that represent the copy number of provirus. Marks in this figure are the same as in Figure 2.

Eighteen out of 19 samples collected after this period were negative for MRD regardless of the presence of HTLV-1 provirus. An exception was the sample taken at the time of relapse that took place 8.3 years after transplantation in case 1. CR was continuously maintained in this case and the peripheral blood samples at 6 and 7 years from transplantation were negative in the MRD test. A subcutaneous tumor, which developed at relapse, consisting mostly of CD4-positive cells, had the same integration site of HTLV-1 as primary ATLL cells, demonstrating that the primary ATLL cells had persisted for more than 8 years as MRD.

Analysis of the origin of cells carrying HTLV-1 provirus

Although most of the cells carrying provirus were HTLV-1-infected cells and were not derived from ATLL clones, these findings raised the question of whether these infected cells derived from recipients or donors. To answer this question, we cultured peripheral blood MNC in semisolid media in the presence of rhIL-2 to clonally expand cells infected with HTLV-1. Among 10 cases that maintained CR more than a year, samples were obtained from eight cases. In five out of eight cases, we could establish 30 cell lines (Table 4). Each cell line contained HTLV-1 provirus (data not shown).

Table 4 Origin of colony-forming cells in recipients

In case 20, in which the graft was rejected after transplantation, all eight cell lines were derived from the recipient cells. Among other four cases, 22 out of 23 cell lines were found to originate from the donor cells including one cell line of case 1 that received transplantation from a noncarrier donor. In case 5, despite long-term CR (4.5 years) and complete donor chimerism in the peripheral blood, there was one cell line (one of seven cell lines) that derived from a recipient. By using the established cell line of recipient origin, we determined the franking genomic sequence of HTLV-1 integration site and set up the inverse-PCR. It was applied retrospectively to the genomic DNA extracted from a paraffin-embedded lymph node, which was a biopsy sample for the initial diagnosis in case 5. The lymph node sample had the same integration site of HTLV-1 as the cell line established 4.5 years after transplantation. Although two peripheral blood samples taken 4.5 years after transplantation were negative for this inverse-PCR, the colony-formation method could detect MRD in the same sample in case 5.

Negative results in the tests for HTLV-1 infection in case 16

In cases 16 and 17, at the time of the last follow-up, HTLV-1 provirus load was below the sensitivity of PCR (1 provirus/105 cells). However, the test for antibody against HTLV-1, which is widely used to demonstrate the infection with HTLV-1, was found to be negative only in case 16 (Table 5). Three different methods (Western blotting, particle agglutination and fluorescent antibody test) failed to demonstrate antibodies against HTLV-1 in this case. PCR tests for other parts apart from tax of HTLV-1, gag and env regions, were also negative. All extensive searches for HTLV-1 infection became negative 2.3 years after transplantation and remained negative 8 months later, 3.1 years from transplantation when this manuscript was written.

Table 5 Serial tests for anti HTLV-1 antibody and provirus in case 16


In the present study, we measured HTLV-1 provirus load, detected MRD and determined the origin of HTLV-1 positive cells in the peripheral blood in 22 cases with ATLL treated with allo-HSCT. The HTLV-1 provirus load was reduced at least once to low levels (less than 1000 copies/105 cells) in most cases even among those who were transplanted in the status other than CR or those who received a reduced-intensity conditioning. These results showed a strong anti-ATLL effect of allo-HSCT in the short period after transplantation. The average dose of HTLV-1 provirus was significantly higher among cases transplanted from HTLV-1 carrier donors, suggesting the carryover of the virus positive cells from the donors. However, the level of provirus load after transplant did not always correlate to the final clinical outcome. Surprisingly, among most of the patients who survived more than 2 years, HTLV-1 provirus was detectable, although at a lower level, by PCR in their peripheral blood. Contrary to our results, Hishizawa et al.16 using a quantitative PCR method similar to ours, reported the kinetics of HTLV-1 provirus load after allo-HSCT in five cases with ATLL, and they showed that HTLV-1 provirus load was undetectable in two cases in continuous CR. Major differences between their report and ours are the length of the follow-up period (1–15 and 1–84 months) and the number of patients (five and 22 cases). The longer observation periods and larger case number in our study might have facilitated the notice of the reappearance of HTLV-1-positive cells after allo-HSCT.

In contrast with the frequent positive results of provirus load, MRD of primary ATLL was rarely detectable after transplantation. In particular, after 6 months from transplantation, all samples of five cases tested during remission were negative for the MRD test despite the detectable level of provirus load, clearly demonstrating the presence of HTLV-1-positive cells other than ATLL in the peripheral blood of these patients.

HTLV-1-positive cells present in the recipients after allo-HSCT could be theoretically categorized into four groups: (1) MRD of primary ATLL cells, (2) non-ATLL cells of a recipient carrying HTLV-1 (e.g. T lymphocytes at the carrier state), (3) donor-derived cells infected with HTLV-1 in the host after transplant and (4) infused donor cells in the case of transplantation from a carrier of HTLV-1. Based on the results of colony-formation experiments, although the number of clones tested was not large, we demonstrated that there was difference in the origin of cells with HTLV-1 provirus. We found MRD in case 5 (as defined in group 1), donor-derived HTLV-1-positive cells in case 1 (group 3) and examples of group 4 in cases 5, 8 and 19. Non-ATLL cells of recipients were shown in case 20 (group 2). In some cases, we assumed that donor CD4-positive T cells were infected de novo with HTLV-1 in the recipient’s body after transplantation as observed in case 1. Virus transmission into donor lymphocytes was described previously and our observation supported this report.17

In case 1, the MRD tests in the peripheral blood were negative in both samples taken at 6 and 7 years from transplantation; however, ATLL relapsed clinically as a subcutaneous tumor after 8 years of continuous CR. With the same integration sites of HTLV-1 in the primary and relapsed tumor cells, it was apparent that the primary ATLL cells remained somewhere in the body for more than 8 years after allo-HSCT and that negative tests for MRD in the peripheral blood did not necessarily indicate eradication of ATLL even long after transplantation.

On the other hand, in case 16, even with the extensive search for HTLV-1 provirus by PCR for various parts of HTLV-1 genome, we failed to demonstrate its presence in the peripheral blood. The antibody against HTLV-1 also became negative only in this case. So far, there has been no evidence to show the presence of HTLV-1 in this case for more than 8 months. There was a previous report of the eradication of HTLV-1 from a carrier who received allo-HSCT for pure red cell aplasia.18 The tests for the virus performed in case 16 were almost the same as used in this report, suggesting that HTLV-1 was cleared off from the body after allo-HSCT in this case, indicating eradication of both ATLL cells and carrier T cells of HTLV-1 simultaneously by allo-HSCT.

Recently, we reported that allo-HSCT would bring about graft-versus-ATLL (GvATLL) effect even without clinically obvious graft-versus-host disease (GVHD).10 GvATLL could be achieved when a specific immune response targeting HTLV-1 was initiated, such as cytotoxic T cells for tax protein as Harashima et al.19 reported. It is also possible that allogeneic immune reaction against recipient cells contributed to GvATLL effect even without HTLV-1-specific immune reactions as seen in transplantations from carrier donors. As most long-term survivors were positive for HTLV-1 provirus and anti-HTLV-1 antibody, our observation suggested that GvATLL had an effect on ATLL cells but not HTLV-1 provirus in most cases. Allogeneic immune reaction without clinically apparent GVHD might be enough to suppress ATLL cells in these situations.

In summary, allo-HSCT for ATLL profoundly reduced provirus load of HTLV-1 in recipients; however, small amounts of HTLV-1-positive cells that remained in long-term survivor were heterogeneous in origin. We also experienced the single case in which HTLV-1 seemed to be eradicated with allo-HSCT. Thus, it was suggested that the way allo-HSCT suppressed and controlled ATLL and HTLV-1 itself was not simple but heterogeneous from case to case. Further analysis is necessary to understand how ATLL is controlled by allo-HSCT through GvATLL effect, and to find how this effect be controlled and enhanced.


  1. 1

    Poiesz BJ, Ruscetti FW, Gazdar AF, Bunn PA, Minna JD, Gallo RC . Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patients with cutaneous T-cell lymphoma. Proc Natl Acad Sci USA 1980; 77: 7419–7451.

  2. 2

    Hinuma Y, Nagata K, Hanaoka M, Nakai M, Matsumoto T, Kinoshita KI 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 USA 1981; 78: 6476–6480.

  3. 3

    Uchiyama T . Human T cell leukemia virus type I (HTLV-I) and human diseases. Annu Rev Immunol 1997; 15: 15–37.

  4. 4

    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: 2534–2537.

  5. 5

    Abott MA, Poiesz BJ, Byrne BC, Kwok S, Sninsky JJ, Ehrlich GD . Enzymatic gene amplification: qualitative and quantitative methods for detecting proviral DNA amplified in vitro. J Infect Dis 1998; 158: 1158–1169.

  6. 6

    Kawase KI, Katamine S, Moriuchi R, Miyamoto T, Kubota K, Igarashi H et al. Maternal transmission of HTLV-1 other than through breast milk: discrepancy between the polymerase chain reaction positivity of cord blood samples for HTLV-1 and the subsequent seropositivity of individuals. Jpn J Cancer Res 1992; 83: 968–977.

  7. 7

    Yamada Y, Tomonaga M . The current status of therapy for adult T-cell leukaemia–lymphoma in Japan. Leuk Lymphoma 2003; 44: 611–618.

  8. 8

    Tsukasaki K, Maeda T, Arimura K, Taguchi J, Fukushima T, Miyazaki Y et al. Poor outcome of autologous stem cell transplantation for adult T cell leukemia/lymphoma: a case report and review of the literature. Bone Marrow Transplant 1999; 23: 87–89.

  9. 9

    Kami M, Hamaki T, Miyakoshi S, Musashige N, Kanda Y, Tanosaki Y et al. Allogeneic haematopoietic stem cell transplantation for the treatment of adult T-cell leukaemia/lymphoma. Br J Haematol 2003; 120: 304–309.

  10. 10

    Utsunomiya A, Miyazaki Y, Takatsuka Y, Hanada S, Uozumi K, Yashiki S et al. Improved outcome of adult T cell leukemia/lymphoma with allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 2001; 27: 15–20.

  11. 11

    Fukushima T, Miyazaki Y, Honda S, Kawano F, Moriuchi Y, Masuda M et al. Allogeneic hematopoietic stem cell transplantation provides sustained long-term survival for patients with adult T-cell leukemia/lymphoma. Leukemia 2005; 19: 829–834.

  12. 12

    Shimoyama M, members of the Lymphoma Study Group. Diagnostic criteria and classification of clinical subtypes of adult T-cell leukaemia/lymphoma. Br J Haematol 1991; 79: 428–437.

  13. 13

    Kamihira S, Dateki N, Sugahara K, Yamada Y, Tomonaga M, Maeda T et al. Real-time polymerase chain reaction for quantification of HTLV-1 proviral load: application for analyzing aberrant integration of the proviral DNA in adult T-cell leukemia. Int J Hematol 2000; 72: 79–84.

  14. 14

    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: 3080–3085.

  15. 15

    Hata T, Fujimoto T, Tsushima H, Murata K, Tsukasaki K, Atogami S et al. Multi-clonal expansion of unique human T-lymphotropic virus type-I-infected T cells with high growth potential in response to interleukin-2 in prodoromal phase of adult T cell leukemia. Leukemia 1999; 13: 215–221.

  16. 16

    Hishizawa M, Imada K, Ishikawa T, Uchiyama T . Kinetics of proviral DNA, soluble interleukin-2 receptor level and tax expression in patients with adult T-cell leukemia receiving allogeneic stem cell transplantation. Leukemia 2004; 18: 167–169.

  17. 17

    Ljungman P, Lawler M, Asjo B, Bogdanovic G, Karlsson K, Malm C et al. Infection of donor lymphocytes with human T lymphotrophic virus type 1 (HTLV-I) following allogeneic bone marrow transplantation for HTLV-I positive adult T-cell leukaemia. Br J Haematol 1994; 88: 403–405.

  18. 18

    Kawa K, Nishiuchi R, Okamura T, Igarashi H . Eradication of human T-lymphotropic virus type 1 by allogeneic bone-marrow transplantation. Lancet 1998; 352: 1034–1035.

  19. 19

    Harashima N, Kurihara K, Utsunomiya A, Tanosaki R, Hanabuchi S, Masuda M et al. Graft-versus-Tax response in adult T-cell leukemia patients after hematopoietic stem cell transplantation. Cancer Res 2004; 64: 391–399.

Download references


This work was supported in part by grant from the Ministry of Health, Labour and Welfare of Japan.

Author information

Correspondence to Y Miyazaki.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Yamasaki, R., Miyazaki, Y., Moriuchi, Y. et al. Small number of HTLV-1-positive cells frequently remains during complete remission after allogeneic hematopoietic stem cell transplantation that are heterogeneous in origin among cases with adult T-cell leukemia/lymphoma. Leukemia 21, 1212–1217 (2007) doi:10.1038/sj.leu.2404678

Download citation


  • ATLL
  • transplantation
  • MRD
  • HTLV-1

Further reading