ORIGINAL ARTICLE
Long-term outcomes after hematopoietic SCT for adult T-cell
leukemia/lymphoma: results of prospective trials
I Choi1, R Tanosaki2, N Uike1, A Utsunomiya3, M Tomonaga4, M Harada5, T Yamanaka6,
M Kannagi7 and J Okamura6, on behalf of the ATLL allo-HSCT Study Group
1Department of Hematology, National Kyushu Cancer Center, Fukuoka, Japan; 2Stem Cell Transplantation Unit, National Cancer
Center Hospital, Tokyo, Japan; 3Department of Hematology, Imamura Bun-in Hospital, Kagoshima, Japan; 4Molecular Medicine
Unit, Department of Hematology, Atomic Bomb Disease Institute, School of Medicine, Nagasaki University, Nagasaki, Japan;
5Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan; 6National Kyushu Cancer Center, Institute for Clinical
Research, Fukuoka, Japan and 7Medical Research Division, Department of Immunotherapeutics, Tokyo Medical and Dental
University, Tokyo, Japan
We have previously conducted clinical trials of allogeneic
hematopoietic SCT with reduced-intensity conditioning
regimen (RIC) for adult T-cell leukemia/lymphoma
(ATLL)—a disease caused by human T-lymphotropic
virus type 1 (HTLV-1) infection and having a dismal
prognosis. Long-term follow-up studies of these trials
revealed that 10 of the 29 patients have survived for a
median of 82 months (range, 54–100 months) after RIC,
indicating a possible curability of the disease by RIC.
However, we have also observed that the patterns of post-
RIC changes in HTLV-1 proviral load over time among
the 10 survivors were classified into three patterns. This is
the first report to clarify the long-term outcomes after
RIC for ATLL patients.
Bone Marrow Transplantation (2011) 46, 116–118;
doi:10.1038/bmt.2010.92; published online 19 April 2010
Keywords: adult T-cell leukemia/lymphoma; allogeneic
hematopoietic SCT; reduced-intensity conditioning regi-
men; HTLV-1 proviral load
Introduction
Adult T-cell leukemia/lymphoma (ATLL) is a peripheral
T-cell malignancy that is caused by human T-lymphotropic
virus type 1 (HTLV-1) infection and commonly affects
individuals at an average age of 60 years. It has been
reported that the 4-year survival rate was only 10.3%; in
particular, patients with an acute or lymphoma subtype
showed a dismal prognosis with a 4-year survival rate of
approximately 5.0%.1 Several retrospective studies for
ATLL patients younger than 50 years have suggested the
possible usefulness of allogeneic hematopoietic SCT
(allo-HSCT) with a conventional conditioning chemothe-
rapy regimen. However, the treatment-related mortality by
conventional allo-HSCT was high (40–60%), probably due
to the disease-specific immune deficiency at diagnosis.2–4
This unacceptable level of mortality, even in the case
of young patients, critically deters the applicability of
conventional allo-HSCT for the general population of
ATLL.
To permit the application of allo-HSCT for ATLL in
patients aged more than 50 years, we can consider
allo-HSCT for ATLL conditioned with reduced-intensity
regimen (hereafter, allo-HSCT conditioned with reduced-
intensity regimen is referred to as ‘RIC’). Few retrospective
studies have reported the results of RIC for ATLL so far;
Shiratori et al.5 followed up 15 patients after allo-HSCT
(including 10 who received RIC) whose median age was 57
years and reported that the OS rate at 3 years reached 73%.
Kato et al.6 investigated the results of 33 patients with allo-
HSCT from unrelated donors but this study included only 6
patients receiving RIC. However, our study group had
previously activated the first clinical trials of RIC in 2001.
These were two trials to clarify the feasibility of RIC: one
studied RIC administered with immunosuppressant antithy-
mocyte globulin (ATG) and the other studied RIC without
ATG. The results have been already published elsewhere7,8
and the treatment-related mortality in both trials collec-
tively decreased to the 20% level, showing that RIC is a
promising procedure for ATLL patients more than 50 years
of age. In this report, we present the results of long-term
follow-up of the two trials and discuss the longitudinal
patterns of changes in HTLV-1 proviral load in survivors.
Patients and methods
The patient characteristics have been described in the
previous reports.7,8 Briefly, patients were eligible if they had
ATLL of acute or lymphoma type and were aged between
Received 21 January 2010; revised 24 February 2010; accepted 3 March
2010; published online 19 April 2010
Correspondence: Dr J Okamura, National Kyushu Cancer Center,
Institute for Clinical Research, 3-1-1 Notame, Minami-ku, Fukuoka
811-1395, Japan.
E-mail: jyokamur@nk-cc.go.jp
Bone Marrow Transplantation (2011) 46, 116–118
& 2011 Macmillan Publishers Limited All rights reserved 0268-3369/11
www.nature.com/bmt

50 and 70 years. The patients were required to be in either
CR or PR at the time of trial registration, and to have a
HLA-identical sibling donor. The conditioning regimen
consisted of fludarabine (30mg/m2 per day) for 5 days and
BU (1mg/kg orally per day) for 2 days. The patients in the
first study also received low-dose ATG (2.5mg/kg per day)
for 2 days, whereas those in the second study did not. On
day 0, G-CSF-mobilized peripheral blood grafts from their
HLA-identical sibling donors were transplanted. To pre-
vent GVHD, we continuously infused CYA (3mg/kg per
day) starting on day 1. The degree of donor–recipient
chimerism in peripheral blood mononuclear cells was
examined according to the previously reported method.9
The HTLV-1 proviral load was estimated using blood
samples obtained before and at 1, 2, 3, 6, 12 months and
every year after transplantation. HTLV-1 proviral DNA
was measured by the quantitative PCR amplification of
HTLV-1 pX DNA.10 The detection limit of the HTLV-1
proviral load was 0.5 copies per 1000 cells. The OS curve
was estimated by the Kaplan–Meier method.
Results and discussion
Long-term survivors after RIC
In all, 15 and 14 patients were registered in the first and
second studies, respectively. Eleven (six and five in the first
and second studies, respectively) and eight (four in each
study) patients died because of ATLL and the treatment,
respectively. The last treatment-related death occurred 26
months after RIC. Characteristics of the remaining 10
patients (5 in each study) are summarized in Table 1. They
are currently alive with a median follow-up period of 82
months after RIC (range, 54–100 months). Of the surviving
patients, six and four patients had the acute and lymphoma
types of ATLL. Of 10 patients, 5 received the grafts from
HTLV-1-positive sibling donors. The OS rate at 60 months
(5 years) was 34% (95% confidence interval, 18–51). No
death was reported beyond 36 months after RIC (Figure 1).
Of the 10 survivors, 3 developed nonhematological
relapse in the skin and/or lymph nodes within a half year
after RIC (Table 1). However, remission was achieved
again in these patients after the discontinuation of CYA,
immunosuppressive agent, and the administration of
additional treatments. In one of these patients, remission
was achieved with the cessation of CYA alone. Two other
patients were treated with systemic chemotherapy as well as
local irradiation or donor lymphocyte infusion after the
discontinuation of CYA, and thereafter obtained remission.
These three patients survived for 100, 88 and 54 months
after RIC, respectively. Because disease recurrence is
usually fatal, the clinical course for the three patients was
unique. It is suggested that the newly established immuno-
logical environment after RIC might have contributed to
the eradication of ATLL lesions after early relapse.
All the 10 survivors developed acute GVHD (9 grades
I–II and 1 grade III). Chronic GVHD was observed in all
but one patient. Although immunosuppressive treatment
was discontinued in 9 of the 10 patients, 1 patient is still
receiving treatment due to active chronic GVHD. The
development of chronic GVHD may suggest the presence
of the graft-vs-ATLL effect. Of note is that 8 of 10
survivors received RIC when they were in PR after
induction chemotherapy.
Kinetic patterns of HTLV-1 proviral load in long-term
survivors
Serial changes in the HTLV-1 proviral load after RIC in the
10 long-term survivors are shown in Figure 2. The changes
in the proviral load are heterogeneous but can be roughly
classified into three patterns. In the first pattern, the
proviral load became undetectable after RIC and continued
to remain so; this pattern was seen in three patients. In the
second pattern, the proviral load had become undetectable
but returned to detectable levels thereafter; this pattern was
also seen in three patients, all of whom had received RIC
from HTLV-1-negative donors. Finally, in the third
pattern, the proviral load had remained at the carrier level
in four patients; these patients received the grafts from
donors who were HTLV-1 carriers. All the 10 survivors
continue to show complete donor chimera during the
observation period regardless of the HTLV-1 proviral load
level.
We noted that one survivor who was donated graft from
an HTLV-1 carrier showed a strikingly high proviral load
(nearly 1000 copies) during the first year after RIC; this
Table 1 Characteristics of long-term survivors
Age
(years)
Gender ATL
subtype
Donor status
of HTLV-1
Status
at RIC
Acute
GVHD
Chronic
GVHD
Relapse Treatment after
relapse
Current Karnofsky
PS score (%)
Survival after
RIC (months)
62 Male Acute (+) PR I Yes Lynd, skin (day 28) d/c CsA 490 100
66 Female Acute (+) PR II Yes No 490 98
51 Male Acute () PR II Yes No 490 98
53 Male Lymph () PR II Yes No 490 91
54 Male Lymph () CR II Yes Lynd (day 171) d/c CsA, Rx, Cx 490 88
55 Male Lymph (+) PR II Yes No 490 75
62 Male Acute (+) CR II Yes No 490 74
50 Female Lymph () PR I Yes No 490 62
56 Male Acute () PR II Yes Skin (day 29) d/c CsA, DLI, steroid 490 54
53 Female Acute (+) PR III No No 490 54
Abbreviations: Cx¼ chemotherapy; d/c¼ discontinued; DLI¼ donor lymphocyte infusion; lynd¼ lymph node; PS¼ performance status;
RIC¼hematopoietic stem cell transplantation conditioned with reduced-intensity regimen; Rx¼ radiation therapy.
Long-term outcomes after RIC allo-HSCT for ATLL
I Choi et al
117
Bone Marrow Transplantation