- Email: [email protected]
Lymphoma Treated with Allogeneic Hematopoietic Stem Cell Transplantation
Accepted Manuscript Title: Risk Assessment in Adult T-Cell Leukemia/Lymphoma Treated with Allogeneic Hematopoietic Stem Cell Transplantation Author: Makoto Yoshimitsu, Ryuji Tanosaki, Koji Kato, Takashi Ishida, Ilseung Choi, Yoshifusa Takatsuka, Takahiro Fukuda, Tetsuya Eto, Michihiro Hidaka, Naoyuki Uchida, Toshihiro Miyamoto, Yasuhiro Nakashima, Yukiyoshi Moriuchi, Koji Nagafuji, Yasuhiko Miyazaki, Tatsuo Ichinohe, Minoko Takanashi, Yoshiko Atsuta, Atae Utsunomiya, ATL Working Group of the Japan Society for Hematopoietic Cell Transplantation PII: DOI: Reference:
S1083-8791(17)30819-4 https://doi.org/10.1016/j.bbmt.2017.11.005 YBBMT 54861
To appear in:
Biology of Blood and Marrow Transplantation
Received date: Accepted date:
18-6-2017 6-11-2017
Please cite this article as: Makoto Yoshimitsu, Ryuji Tanosaki, Koji Kato, Takashi Ishida, Ilseung Choi, Yoshifusa Takatsuka, Takahiro Fukuda, Tetsuya Eto, Michihiro Hidaka, Naoyuki Uchida, Toshihiro Miyamoto, Yasuhiro Nakashima, Yukiyoshi Moriuchi, Koji Nagafuji, Yasuhiko Miyazaki, Tatsuo Ichinohe, Minoko Takanashi, Yoshiko Atsuta, Atae Utsunomiya, ATL Working Group of the Japan Society for Hematopoietic Cell Transplantation, Risk Assessment in Adult T-Cell Leukemia/Lymphoma Treated with Allogeneic Hematopoietic Stem Cell Transplantation, Biology of Blood and Marrow Transplantation (2017), https://doi.org/10.1016/j.bbmt.2017.11.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Risk assessment in adult T-cell leukemia/lymphoma treated with allogeneic
2
hematopoietic stem cell transplantation
3 4
Makoto Yoshimitsu1,†, Ryuji Tanosaki2, Koji Kato3, Takashi Ishida4, Ilseung Choi5, Yoshifusa
5
Takatsuka6, Takahiro Fukuda7, Tetsuya Eto8, Michihiro Hidaka9, Naoyuki Uchida10,
6
Toshihiro Miyamoto3, Yasuhiro Nakashima5, Yukiyoshi Moriuchi11, Koji Nagafuji12, Yasuhiko
7
Miyazaki13, Tatsuo Ichinohe14, Minoko Takanashi15, Yoshiko Atsuta16, and Atae
8
Utsunomiya6; ATL Working Group of the Japan Society for Hematopoietic Cell
9
Transplantation.
10 11
1
12
Japan, 2 Department of Transfusion Medicine and Cell Therapy, Keio University Hospital, Tokyo,
13
Japan, Department of Medicine and Biosystemic Science, Kyushu University Graduate
14
School of Medical Science, Fukuoka, Japan, 4Department of Hematology and Oncology,
15
Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, 5Department
16
of Hematology, National Kyushu Cancer Center, Fukuoka, Japan, 6 Department of
17
Hematology, Imamura Bun-in Hospital, Kagoshima, Japan, 7 Department of Hematopoietc
18
Stem Cell Transplantation, National Cancer Center Hospital, Tokyo, Japan, 8Department of
19
Hematology, Hamanomachi Hospital, Fukuoka, Japan, 9Deparment of Hematology, National
20
Hospital Organization Kumamoto Medical Center, Kumamoto, Japan, 10Department of
21
Hematology, Federation of National Public Service Personnel Mutual Aid Associations
22
Toranomon Hospital, Tokyo, Japan, 11Department of Hematology, Sasebo City General
23
Hospital, 12Division of Hematology and Oncology, Department of Medicine, Kurume
24
University Hospital, 13Department of Hematology, Oita Prefectural Hospital, 14Department of
25
Hematology and Oncology, Research Institute for Radiation Biology and Medicine, Hiroshima
26
University, Hiroshima, Japan,
Department of Hematology and Immunology, Kagoshima University Hospital, Kagoshima,
3
15
Blood Service Headquarters, Japanese Red Cross Society,
1 Page 1 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
16
Japanese Data Center for Hematopoietic Cell Transplantation, Nagoya,
1
Tokyo, Japan,
2
Japan
3
†Corresponding author, mailing address: Department of Hematology and Immunology,
4
Kagoshima University Hospital
5
8-35-1 Sakuragaoka, Kagoshima, Japan, 890-8520.
6
Phone: +81-099-275-5934, Fax: +81-99-275-5947.
7
E-mail: [email protected]
8 9 10
Short title; Pre-transplant comorbidities and HCT in ATL
11
Text word count: 4058
12
Abstract word count: 197
13
The number of figures and tables: 5 Figures, 6 Tables
14
The number of references: 32
15 16
The scientific category: TRANSPLANTATION
17
2 Page 2 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Key Points:
2
HCT-CI and mEBMT are useful pre-transplant prognostic factors for patients with ATL.
3
Newly developed ATL HCT-PI may have better prognostic value for patients with ATL.
4 5
3 Page 3 of 32
YOSHIMITSU et al.
1 2
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Highlights: Previously examined factors associated with HCT, were not prognostic.
3
A total of 824 patients were retrospectively examined in this study.
4
HCT-CI is useful pre-transplant prognostic factors for patients with ATL.
5
Newly developed ATL HCT-PI may have better prognostic value for patients with
6
ATL.
7 8
Summary
9
Disease status at allogeneic hematopoietic cell transplantation (HCT) is an important
10
pre-transplant prognostic factor of HCT in adult T-cell leukemia/lymphoma (ATL); however,
11
other prognostic factors, including comorbidities, were not predictive in small cohort
12
analyses. Several scoring systems (HCT-comorbidity index [CI]/modified European Society
13
for Blood and Marrow Transplantation risk score [mEBMT]) have been adopted to predict
14
HCT outcomes in other hematological malignancies. We retrospectively evaluated HCT-CI
15
and mEBMT to predict non-relapse mortality (NRM) in 824 ATL patients registered in the
16
Japan Society for Hematopoietic Cell Transplantation TRUMP database, from 2008 until
17
2013. A higher HCT-CI was associated with greater NRM when comparing HCT-CI 0 vs.
18
HCT-CI 1–3 and HCT-CI 0 vs. HCT-CI ≥4. A higher mEBMT score was not associated with
19
higher NRM when comparing mEBMT 0–3 with 4–6. As ATL patients are older and
20
consequently at risk of additional complications, we developed an optimized prognostic
21
index for ATL (ATL-HCT-PI) using known risk factors: age, HCT-CI, and donor/recipient sex
22
combination. The ATL-HCT-PI scores effectively predicted the 2-year NRM (22.0%, 27.7%,
23
and 44.4%, respectively). Therefore, the newly developed ATL-HCT-PI, in combination with
24
other risk factors, is more useful for predicting NRM in HCT for ATL patients.
25 26
Keywords: allogeneic hematopoietic cell transplantation; comorbidities; adult T-cell
27
leukemia lymphoma; hematological malignancies; mortality.
4 Page 4 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Introduction
2
Adult T-cell leukemia/lymphoma (ATL) is an intractable hematological malignancy
3
caused by the latent infection of human T-cell leukemia virus-I (HTLV-1) 1. The 3-year
4
overall survival (OS) of ATL patients receiving optimal chemotherapy is approximately
5
24% 2. Allogeneic hematopoietic stem cell transplantation (HCT) is an attractive
6
alternative modality that offers long-term remission in some patients with ATL
7
nationwide retrospective analysis showed that the 3-year OS for the entire cohort was
8
33% 5. Notably, the cumulative incidence of treatment-related mortality (TRM) at 3 years
9
in this cohort was very high (approximately 43%). Therefore, there is an urgent need for
10
clinical solutions to decrease the TRM. One possible solution involves a more accurate
11
prediction of the risk of TRM before allo-HCT. Although several prognostic indices for
12
ATL have been reported, these were either determined in ATL patients without allo-HCT,
13
or have never been validated in terms of OS or TRM for patients with ATL receiving
14
allo-HCT
15
ATL undergoing allo-HCT. The risk factors associated with OS or TRM include the use
16
of cord blood as a donor source, development of grade III–IV acute GVHD, presence of
17
extensive chronic graft versus host disease (GVHD), male sex, performance status (PS)
18
2–4, late transplantation from diagnosis, and old age
19
related to pre-transplantation conditions, including comorbidities, have not been
20
thoroughly evaluated in patients with ATL undergoing allo-HCT.
21
The hematopoietic stem cell transplantation comorbidity index (HCT-CI) is widely used
22
to predict TRM 13. The HCT-CI was established using a large data set and was found to
23
be reproducible in many transplantation settings
24
available on whether HCT-CI is suitable for use among lymphoma patients receiving
25
allo-HCT
26
(n = 63), Pollack et al .reported that the use of HCT-CI is not as beneficial in NHL
3, 4
. A
6-8
. Hence, it is unclear whether these indices can be applied to patients with
5, 9-12
. However, other factors
14-17
. However, limited data are
18
. By analyzing a limited number of non-Hodgkin’s lymphoma (NHL) patients
5 Page 5 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
19
1
patients as in leukemia patients
. In addition, Mori et al. reported that the
2
Pre-transplantation Assessment of Mortality (PAM) score is not as useful in patients with
3
ATL receiving allo-HCT, as compared to that in patients with other hematological
4
malignancies 20.
5
In the present study, we tested the ability of HCT-CI to predict non-relapse mortality
6
(NRM) in patients with ATL receiving allo-HCT, by using the largest ATL allo-HCT
7
registry available worldwide. Moreover, we tested the utility of the modified European
8
Society for Blood and Marrow Transplantation risk score (mEBMT) score and other
9
reported pre-transplant risk factors. In addition, we propose a novel optimized
10
pre-transplant prognostic index for patients with ATL receiving allo-HCT, called the ATL
11
hematopoietic stem cell transplantation prognostic index (ATL-HCT-PI). This new
12
prognostic index better predicts NRM and OS than either the HCT-CI or mEBMT score.
13
To our knowledge, this is the largest corroborative study of HCT-CI and mEBMT scores
14
with a single disease entity.
15
6 Page 6 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Methods
2
Data Collection
3
ATL patients who underwent their first allo-HCT between January 1, 2008, and
4
December 31, 2013 (n = 830), when HCT-CI was introduced into the registry data
5
parameters, were considered for this study. Data were collected from the nationwide
6
survey database of the Japan Society for Hematopoietic Cell Transplantation (JSHCT).
7
The data were updated as of December 2014. Five cases with missing HCT-CI data and
8
1 case receiving autologous HCT were excluded. Therefore, a total of 824 patients were
9
finally included in this analysis. Data collected for analysis included the patients’ clinical
10
characteristics, such as HCT-CI parameters,13 ATL clinical subtype,21 age at HCT, sex,
11
donor sex, survival data at the last follow-up, PS at HCT, disease status at HCT, stem
12
cell sources, HLA typing in serological resolution, ABO blood group, conditioning
13
regimen, date at HCT, donor relationship with the recipient (related or unrelated), date
14
of death, and primary cause of death. The study was approved by the data
15
management committees of the JSHCT, as well as by the institutional ethical committee
16
of Kagoshima University Hospital.
17
Definitions
18
OS was calculated from the date of HCT until the date of death or last follow-up. The
19
registered causes of death were categorized as either ATL-related or Non-relapse
20
mortality (NRM). ATL-related mortality was defined as death with relapse or progression
21
of ATL in patients, based on the judgment of each registered physician. NRM was
22
defined as any death other than ATL-related mortality. If the primary cause of death was
23
GVHD or infection after the relapse or progression of ATL, for example, we classified
24
such a case as ATL-related mortality. The need for a myeloablative conditioning
25
regimen (MAC) and reduced intensity conditioning regimen (RIC) was registered based
26
on the judgment of each physician. An HLA-match was defined when the serologic 7 Page 7 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
resolution typing for HLA-A, HLA-B, and HLA-DR were identical.
2 3
Statistical analysis
4
The probability of OS was estimated according to the Kaplan-Meier method. The
5
incidence of NRM was calculated using cumulative incidence estimates with the
6
competing risk model, while considering ATL-related mortality as a competitive event 22.
7
Univariate Cox proportional regression analysis was used to estimate the risk factors for
8
OS. The variables tested in the univariate analysis for transplant outcomes are PS, sex,
9
donor-recipient sex combination, ABO mismatch, serological HLA status, and age.
10
Multivariate analysis was performed while accounting for the competing risk structure.
11
Pre-transplant variables were included in the multivariate analysis model when a
12
significance level of P ≤ 0.1 was detected on univariate Cox regression analysis and
13
Gray’s test. A significance level of P ≤ .05 was required in the multivariate analysis. Fine
14
and Gray proportional hazard modeling was used to estimate the effect of the same
15
variables used in the univariate analysis of NRM and ATL-related mortality on the
16
cumulative incidence of NRM and ATL-related mortality. A backward selection algorithm
17
was used to select the final model.
18
Development of the new prognostic index
19
In order to develop the new ATL-HCT-PI, integer weights for the ATL-HCT-PI were
20
derived from proportional hazards with NRM over the first 2 years. Therefore, NRM was
21
used for the development of the scores instead of OS, similar to that used for the
22
original HCT-CI. This was chosen since deaths from non-ATL-related causes are more
23
likely to be influenced by pre-transplant comorbidities than deaths from ATL itself, which
24
was treated as a competing risk. The adjusted hazard ratios (HRs) for NRM over the
25
first 2 years after HCT were calculated for each extracted factor from the multivariate
26
analysis. The adjusted HRs were converted to integer weights as follows: factors with 8 Page 8 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
an adjusted HR of ≤1.25 were excluded from consideration, factors with an adjusted HR
2
of 1.26–2.0 were assigned a weight of 1, and factors with an adjusted HR of 2.1–3.0
3
were assigned a weight of 2. The ATL-HCT-PI score was calculated as the sum of these
4
integer weights. The ATL-HCT-PI scores were further collapsed into 3 risk groups: 0
5
(low risk), 1 (intermediate risk), and ≥2 (high risk). C-statistics estimated the predictive
6
capacity of the HCT-CI, mEBMT score, and newly developed ATL HCT-PI, as described
7
23
. All statistical analyses were performed using EZR 24.
8
9 Page 9 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Results
2
Patient characteristics
3
Table 1 summarizes the characteristics of the 824 patients with ATL. At the time of
4
analysis — i.e., after a median follow-up for survivors of 27.8 months (range, 0–80
5
months) and after a median follow-up for all patients of 7.5 months (range, 0–80
6
months) — 311 patients (37.7%) remained alive. The projected OS of the whole cohort
7
at 2 years was 37.5%, with a median OS of 8.6 months (95% CI, 7.1–10.6). The
8
cumulative incidence of NRM at 2 years was 27.5% (95% CI, 24.4–30.7%; Fig 1).
9
HCT-CI
10
The HCT-CI was scored based on the judgment of each registered physician.
11
Co-morbidities, defined by the HCT-CI criteria, were identified in 304 (36.9%) of all
12
patients, with 102 (12.4%), 88 (10.7%), 66 (8.0%), and 48 (5.8%) presenting with scores
13
1, 2, 3, and ≥4, respectively. The main comorbidities in the registered data included
14
moderate pulmonary disease, mild hepatic disease, diabetes, and infection (Table 2).
15
Based on the original HCT-CI risk classification (score 0, low risk; score 1–2,
16
intermediate risk; score ≥3, high risk) 13, 63.1% of the patients were assigned to the
17
low-risk group, 23.1% were assigned to the intermediate-risk group, and 13.8% were
18
assigned to the high-risk group. When the OS was compared among these risk groups,
19
we observed a decreasing trend in OS in patients with a higher HCT-CI. However, this
20
difference was not statistically significant between the intermediate-risk and high-risk
21
groups (2-year OS of 42.8%, 29.8%, and 26.7% for HCT-CI low-, intermediate-, and
22
high-risk patients, respectively; Fig 2A). The HR was 1.52 (95% CI, 1.24–1.87) for
23
intermediate-risk patients and 1.62 (95% CI, 1.27–2.06) for high-risk patients in
24
comparison with the low risk patients. Moreover, among patients with higher HCT-CI,
25
the NRM was greater but the difference was marginal among groups (2-year NRM of
26
25.3%, 29.2%, and 34.2% for HCT-CI low-, intermediate-, and high-risk patients, 10 Page 10 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
respectively; Fig 2B). However, when other cut-off points for the definition of risk groups
2
(score 0, low risk, score 1–3: intermediate risk, score ≥4: high risk) were used, there
3
was a considerable change in the results. When the OS was compared for these
4
re-stratified risk groups (revised HCT-CI), we observed a decreasing trend in OS in
5
patients with higher revised HCT-CI (2-year OS of 42.8%, 32.7%, and 7.8% in the
6
revised HCT-CI low-, intermediate-, and high-risk patients; Fig 2C; Table 3). Moreover,
7
among the patients with higher revised HCT-CI, NRM was greater in the high-risk
8
patients (2-year NRM of 25.3%, 26.8%, and 52.6% for revised HCT-CI low-,
9
intermediate-, and high-risk patients, respectively; Fig 2D; Table 3). The ATL-related
10
mortality was comparable among the 3 risk groups (Table 3).
11
Among the significant HCT-CI factors (cardiac function, infection, liver function,
12
pulmonary function, P<0.10 by univariate analysis), only severe pulmonary dysfunction
13
had a significant influence on NRM in multivariate analysis (HR, 2.06 [95% CI,
14
1.23–3.47]).
15 16 17
mEBMT Thereafter, the mEBMT risk score was determined. The risk score accounts for
18
recipient age, disease stage, donor/host histocompatibility, and donor/recipient gender
19
combination 25.The prevalence of individual factors according to the definition of the
20
mEBMT is shown in Table 4. The median mEBMT risk score was 4 (range, 1–6), with
21
0.4%, 5.9%, 25.0%, 16.1%, 24.0%, 8.7%, and 9.8% of patients scoring 1 or 2, 3, 4, 5, 6,
22
or NA (not available because of lack of data), respectively. Based on the mEBMT risk
23
classification (score 0–3, low; score 4–6, high), 34.7% of patients were assigned to the
24
low-risk group. When the OS was compared among these risk groups, we observed a
25
significant decrease in OS in patients with a higher mEBMT score (2-year OS of 49.6%
26
and 32.6% for mEBMT low-risk and high-risk patients, respectively; Fig 3A, Table 5). 11 Page 11 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
The NRM was comparable between low-risk and high-rsik patients (2-year NRM of
2
28.7% and 27.7% for mEBMT low- and high-risk patients, respectively; Fig 3B, Table 5).
3
The ATL-related mortality was greater in high-risk patients (2-year ATL-related mortality
4
of 21.9% and 39.7% for mEBMT low- and high-risk patients, respectively; Table 5).
5
Among all the mEBMT parameters, donor/recipient gender combination was found to
6
have a significant influence on NRM in multivariate analysis (HR, 1.71 [95% CI,
7
1.28–2.29]. We further analyzed this donor/recipient gender combination in detail, and
8
found that the combination of female donor to male recipient exhibited consistently
9
worse OS (Figure 3C) and NRM (Fig 3D), as compared with the other combinations.
10 11
PS
12
PS is a well-known risk factor for ATL 7, 8, 10. In this cohort, the median PS was 1 (range,
13
0–4). Based on the previous PS risk classification (score 0-1, low risk; score 2–4, high
14
risk)
15
compared among these risk groups, we observed a significant decreasing trend in OS
16
in patients with a higher PS (2-year OS of 40.8% [95% CI, 36.9–44.6%] and 17.6%
17
[10.9–25.7%] for PS low- and high-risk patients, respectively). The HR was 2.26 (95%
18
CI, 1.80–2.837) for high-risk patients in comparison with the low-risk patients.
19
Interestingly, the NRM was comparable between PS low-risk and PS high-risk patients
20
(2-year non-ATL-related mortality of 27.5% [95% CI, 24.1-30.9%] and 27.9%
21
[19.6–36.7%] for PS low- and high-risk patients, respectively). The ATL-related mortality
22
was significantly greater in PS high-risk patients (2-year ATL-related mortality of 31.9%
23
and 54.5% for PS low- and high-risk patients, respectively), with a HR of 2.36 (95% CI,
24
1.73–3.21) for PS high-risk patients in comparison with the low-risk patients.
86.4% of patients were assigned to the low-risk group 8. When the OS was
25 26
Age 12 Page 12 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Age is a major factor for assessing the suitability of allogeneic HCT. The
2
comorbidity-age index, based on the HCT-CI, was recently proposed to facilitate clinical
3
decision making and comparative-effectiveness research of HCT 26. In that cohort of
4
3033 patients, the median age was 45 years. In our cohort of 824 ATL patients, the
5
median age was 57 years (range, 24–72 years), and 96.5% of patients were aged >40
6
years. Although age serves as a parameter for the mEBMT score estimation, we found
7
that the score could be divided into 3 groups (age <19 years, 20–40 years, and >41
8
years), with 95.5% of our cohort was aged >41 years. Therefore, age does not enable
9
pre-transplant risk stratification in patients with ATL. An earlier study of 386 ATL
10
patients with HCT indicated that older age (>50 years) was significantly associated with
11
a lower survival rate 5, whereas another study of 586 ATL patients with HCT showed
12
that older age (>55 years) was not a significant risk factor of therapy-related mortality 10.
13
In our larger cohort, patients were sorted by age and divided into 5 groups based on
14
almost equal number of NRM events (total 229 events) (≤50 years, n = 42; 51–55 years,
15
n = 50; 56–59 years, n = 47; 60–63 years, n = 45; ≥64 years, n = 45). Although
16
increasing age was associated with higher TRM, no difference was observed between
17
the age groups of ≤50 years, 51–55, 56–59, and 60–63 years (Fig. 4). Compared with
18
patients in the age group of ≤50 years, those in the age groups 51-55, 56-59, 60-63, and
19
≥64 years had HRs for NRM of 1.17 (95% CI, 0.78–1.75, P = 0.46), 1.25 (95% CI,
20
0.82–1.89, P = 0.30), 1.10 (95% CI, 0.73–1.69, P = 0.63) and 1.60 (95% CI,1.06–2.41,
21
P =0.025), respectively.
22 23
ATL-HCT-PI
24
Among the valuables tested in the univariate analysis for NRM, HCT-CI, donor/recipient
25
sex combination, and age over 64 were determined as independent risk factors in
26
multivariate analysis (Table 6). Based on these factors, we developed a new composite 13 Page 13 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
pre-transplant prognostic index for ATL, namely the ATL-HCT-PI. We assigned points
2
for each independent risk factor, as described previously. Thereafter, we divided the
3
cohort into 3 risk groups: low risk, intermediate risk, and high risk. The cases with an
4
ATL-HCT-PI score of 0 were assigned to the low-risk group, 1 were assigned to the
5
intermediate-risk group, and ≥2 were assigned to the high-risk group. Based on this
6
stratification, NRM was elevated in the intermediate-risk and high-risk groups (2-year
7
NRM of 22.0%, 27.7%, and 44.4% for ATL-HCT-PI low-, intermediate-, and high-risk
8
patients, respectively; Fig 5A), with a HR of 1.46 (95% CI, 1.06–2.00) for
9
intermediate-risk patients and 2.48 (95% CI, 1.76–3.49) for high-risk patients in
10
comparison with the low-risk patients. This index marginally predict ATL-related
11
mortality, as the HR was 1.18 (95% CI, 0.89–1.55; P = 0.25) and 1.42 (95% CI,
12
1.01–2.01; P = 0.046). When the OS was compared among these risk groups, we
13
observed a decreasing trend in the OS of patients with a higher ATL-HCT-PI (2-year OS
14
of 47.3%, 39.1%, and 15.7% for ATL-HCT-PI low-, intermediate-, and high-risk patients,
15
respectively; Fig 5B). The HR was 1.43 (95% CI, 1.16–1.76) for intermediate-risk
16
patients and 2.48 (95% CI, 1.94–3.16) for high-risk patients, in comparison with the
17
low-risk patients.
18
When comparing the OS by ATL-HCT-PI with the OS by HCT-CI using c-statistics, we
19
found that ATL-HCT-PI has a higher c-statistic than HCT-CI, with a better distribution
20
(0.629 and 0.582, respectively; P < 0.001).
21
14 Page 14 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1 2 3
Discussion In
the
present
study,
we
assessed
whether
the
patient-,
disease-,
and
4
transplantation-related variables and pre-transplant comorbidities, represented by
5
HCT-CI and the modified EBMT risk score, enhanced the prognostication of NRM,
6
ATL-related mortality, and OS in a cohort of 824 ATL patients treated in Japan between
7
2008 and 2013. To our knowledge, this score has not been analyzed in a large cohort of
8
ATL patients. Our analysis showed that HCT-CI could identify comorbid conditions in
9
only 36.9% of patients, which is relatively smaller than that reported previously 13, 14
10
(52%–80%)
11
remains a concern, since a brief training program is usually proposed for facilitating
12
reliable assessment of comorbidities
13
study that examined 8115 patients treated with allogeneic SCT reported that 48% of the
14
registered patients had a HCT-CI score of 028. In that cohort, patients were older and
15
were more refractory to chemotherapy prior to allo-HCT. Therefore, it is possible that the
16
patients with more comorbidity were ineligible to undergo allo-HCT.
. The consistency in comorbidity coding across different evaluators
27
. However, a recent large prospective validation
17
Pulmonary disease, hepatic disease, diabetes, and infection had the highest
18
prevalence in our cohort, whereas the other comorbidities were rarely observed. Among
19
the comorbidities included in the HCT-CI, only severe pulmonary comorbidities were
20
found to be predictive of increased NRM; the remaining comorbidities had no significant
21
prognostic value on NRM by multivariate analysis. When analyzing the composite score,
22
we observed a trend for inferior OS and higher NRM in cases with higher HCT-CI.
23
However, prognostic significance was not achieved when risk was stratified according to
24
the original report (score 0, low; score 1–2, intermediate; score ≥3, high; Fig 2A-B).
25
Nevertheless, when the risk was stratified as follows—score 0, low; score 1–3,
26
intermediate; score ≥4, high—we observed a significantly inferior OS and higher NRM in 15 Page 15 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
cases with higher risk (Fig 2C-D). This difference in our population could be attributed to
2
the specific characteristics of our patient cohort. The prevalence of comorbidities, age
3
distribution, and preconditioning regimen were different in our population. In addition, a
4
recent study on the prospective validation of the predictive power of HCT-CI in a large
5
cohort indicated that OS and NRM between cases with HCT-CI scores of 2 and 3 were
6
almost identical in the allogeneic SCT setting
7
sub-analysis was focused on reduced intensity/non-myeloablative conditioning, no
8
difference was observed in the NRM between HCT-CI score 0 and 1–2. Takano et al.
9
also reported that patients with diabetes mellitus as a comorbidity had different
28
. In the present study, when the
29
10
outcomes of NRM in a large Japanese registry study (n = 7626)
11
diabetes mellitus, the outcomes of NRM between an HCT-CI score of 1 and of 2–3 did
12
not differ. However, the reasons underlying these results are unclear. The predictive
13
value of HCT-CI may depend on specific patient characteristics and treatment protocols.
14
Accordingly, it is important to note that median age in our study (57 years) was
15
considerably greater than that in the original report (44.8 years)
16
influence of comorbidities on the overall treatment outcome might depend on the
17
analyzed disease, with a possible loss of predictive value in groups with many high
18
relapse-risk patients. Therefore, the application of this index in a disease- and
19
population-based validation analysis would be warranted. In the present study, the
20
population was older, and only 5.8% of the patients were at high risk. These findings
21
highlight the need for a new pre-transplant risk score that combines age and other
22
known pre-transplantation risk factors in ATL.
. In patients with
13
. Furthermore, the
23
The mEBMT risk score was established for acute lymphoblastic leukemia patients,
24
based on the registry data 25. This score accounts for the recipient’s age, disease status,
25
donor/host histocompatibility, and donor/recipient sex combination. The mEBMT score
26
had a good prognostic value for OS and ATL-related mortality. However, there were no 16 Page 16 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
prognostic value of mEBMT score for NRM. In the analysis of single risk score
2
parameters, although disease status, age, and donor type had a significant influence on
3
OS, we found that only donor/recipient gender combination had a significant influence
4
on NRM in the multivariate analysis. Disease status is a well-known risk factor for OS
5
and ATL-related mortality in patients with ATL receiving all-HCT 5, 10, 30. Older age is also
6
a known risk factor for OS in patients with ATL receiving allo-HCT
7
95.5% of registered patients were aged >40 years, and previous reports stratified age
8
groups at 50 or 55 years for better discrimination. Hence, we decided to re-stratify the
9
age for NRM. Accordingly, we divided the patient cohort into 5 groups. Increasing age
10
was associated with higher NRM; however, no difference was observed between the
11
age groups of ≤50, 51-55, 56–59, and 60–63 years (Fig 4). To our knowledge, this is the
12
first study to analyze the influence of age on NRM in patients with ATL receiving
13
allo-HCT.
14
Interestingly, the donor/recipient gender combination had a significant influence on
15
NRM. We further analyzed this factor in detail, and found that only the combination of
16
female donor and male recipient had a significant impact on the OS and NRM (Fig
17
3D-E). Gahrton reviewed the impact of donor and recipient gender combination in
18
allo-HCT for various hematological disorders and reported that transplantation-related
19
mortality in the female donor and male recipient combination was higher than in other
20
combinations 31. Our result was consistent with those findings. To our knowledge, this is
21
the first study to validate that the NRM in the female donor and male recipient
22
combination is higher in patients with ATL receiving allo-HCT. Previous reports indicated
23
that male sex was a poor risk factor for patients with ATL receiving allo-HCT
24
However, the donor/recipient combination might be more important, since our large
25
cohort study indicated that the NRM in male donor/male recipient is identical to that in
26
the female recipient group.
5, 10
. In this cohort,
5, 10
.
17 Page 17 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Unlike to previous reports, the pre-transplant PS was not a risk factor for predicting 10
. Reliable pre-transplant risk stratification using comorbidities and other factors
2
NRM
3
has been found to be essential for comprehensive risk adjustment in patients with
4
cancer. Fuji et al. recently reported a new prognostic index for ATL patients aged 70
5
years or younger. This prognostic index consists of the clinical subtype, ECOG PS at
6
diagnosis, calcium levels, CRP levels, and soluble interleukin-2 receptor status32. Even
7
though this index is useful for predicting OS of the transplanted group, those factors
8
mainly reflect the aggressiveness of ATL itself. This is in contrast to our report, were we
9
focused on the validation of 2 commonly used predictive models in the HCT setting and
10
developed a new index based on mainly non-ATL disease factors. We incorporated
11
comorbidities with age and donor/recipient gender combination in order to develop a
12
new pre-transplant risk stratification method in patients with ATL (ATL-HCT-PI; Figure
13
5). We also identified the prevalent and prognostically important pre-transplant
14
conditions that should be measured and considered in clinical trials and HCT registries.
15
Furthermore, we noted that ATL-HCT-PI can be widely used by referring hematologists
16
when assessing the risks of HCT in patients with ATL.
17
In summary, this is the first study to assess the utility of 2 commonly used predictive
18
models in the HCT setting (the HCT-CI and the EBMT score) in a large population of
19
allo-HCT recipients with ATL. We found that HCT-CI has validity as a comorbidity
20
instrument in patients with ATL. Furthermore, we developed a simple tool for capturing
21
pre-transplant risk based on comorbidities and other known factors, which can be used
22
to predict outcomes and stratify patients for HCT. In the future, we will further
23
substantiate this index by using a validation cohort.
24
18 Page 18 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Acknowledgement
2
The authors thank all the physicians and data managers at the institutions that contributed
3
valuable data on transplantation for ATL at the JSHCT. The authors also thank all the
4
members of the data management committees of the JSHCT.
5 6
Authorship
7
M.Y., R.T., K.K., T.I., C.I., and A.U. designed the research, organized the project, and wrote
8
the paper; M.Y. performed statistical analysis; Y.T., T.F., T.E., M.H., N.U., T.M., Y.N.,
9
Y.Moriuchi., K.N.,Y.Miyazaki., T.I., M.T., and Y.A. gathered the data; M.Y., R.T, K.K, T.I, C.I.,
10
and A.U. wrote the draft of the paper; and all the authors contributed to the final version.
11 12
Conflict-of-interest disclosure
13
All authors declare no competing financial interest.
14
19 Page 19 of 32
YOSHIMITSU et al.
1
References
2
1.
3 4
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Ishitsuka K, Tamura K. Human T-cell leukaemia virus type I and adult T-cell leukaemia-lymphoma. The Lancet. Oncology. 2014;15:e517-526.
2.
Tsukasaki K, Utsunomiya A, Fukuda H, et al. VCAP-AMP-VECP compared with
5
biweekly CHOP for adult T-cell leukemia-lymphoma: Japan Clinical Oncology Group
6
Study JCOG9801. Journal of clinical oncology : official journal of the American
7
Society of Clinical Oncology. 2007;25:5458-5464.
8
3.
9
leukemia/lymphoma with allogeneic hematopoietic stem cell transplantation. Bone
10 11
Utsunomiya A, Miyazaki Y, Takatsuka Y, et al. Improved outcome of adult T cell
marrow transplantation. 2001;27:15-20. 4.
Okamura J, Utsunomiya A, Tanosaki R, et al. Allogeneic stem-cell transplantation
12
with reduced conditioning intensity as a novel immunotherapy and antiviral
13
therapy for adult T-cell leukemia/lymphoma. Blood. 2005;105:4143-4145.
14
5.
Hishizawa M, Kanda J, Utsunomiya A, et al. Transplantation of allogeneic
15
hematopoietic stem cells for adult T-cell leukemia: a nationwide retrospective study.
16
Blood. 2010;116:1369-1376.
17
6.
Suzumiya J, Ohshima K, Tamura K, et al. The International Prognostic Index
18
predicts outcome in aggressive adult T-cell leukemia/lymphoma: analysis of 126
19
patients from the International Peripheral T-Cell Lymphoma Project. Annals of
20
oncology : official journal of the European Society for Medical Oncology / ESMO.
21
2009;20:715-721.
22
7.
Katsuya H, Yamanaka T, Ishitsuka K, et al. Prognostic index for acute- and
23
lymphoma-type adult T-cell leukemia/lymphoma. Journal of clinical oncology :
24
official journal of the American Society of Clinical Oncology. 2012;30:1635-1640.
25
8.
Fukushima T, Nomura S, Shimoyama M, et al. Japan Clinical Oncology Group
26
(JCOG) prognostic index and characterization of long-term survivors of aggressive
27
adult T-cell leukaemia-lymphoma (JCOG0902A). British journal of haematology.
28
2014;166:739-748.
29
9.
Kanda J, Hishizawa M, Utsunomiya A, et al. Impact of graft-versus-host disease on
30
outcomes after allogeneic hematopoietic cell transplantation for adult T-cell
31
leukemia: a retrospective cohort study. Blood. 2012;119:2141-2148.
32
10.
Ishida T, Hishizawa M, Kato K, et al. Allogeneic hematopoietic stem cell
33
transplantation for adult T-cell leukemia-lymphoma with special emphasis on
34
preconditioning regimen: a nationwide retrospective study. Blood.
35
2012;120:1734-1741.
36
11.
Ishida T, Hishizawa M, Kato K, et al. Impact of graft-versus-host disease on
20 Page 20 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
allogeneic hematopoietic cell transplantation for adult T cell leukemia-lymphoma
2
focusing on preconditioning regimens: nationwide retrospective study. Biology of
3
blood and marrow transplantation : journal of the American Society for Blood and
4
Marrow Transplantation. 2013;19:1731-1739.
5
12.
Fuji S, Fujiwara H, Nakano N, et al. Early application of related SCT might improve
6
clinical outcome in adult T-cell leukemia/lymphoma. Bone marrow transplantation.
7
2016;51:205-211.
8
13.
9
(HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic
10 11
Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation HCT. Blood. 2005;106:2912-2919.
14.
Sorror ML, Sandmaier BM, Storer BE, et al. Comorbidity and disease status based
12
risk stratification of outcomes among patients with acute myeloid leukemia or
13
myelodysplasia receiving allogeneic hematopoietic cell transplantation. Journal of
14
clinical oncology : official journal of the American Society of Clinical Oncology.
15
2007;25:4246-4254.
16
15.
Sorror ML, Giralt S, Sandmaier BM, et al. Hematopoietic cell transplantation
17
specific comorbidity index as an outcome predictor for patients with acute myeloid
18
leukemia in first remission: combined FHCRC and MDACC experiences. Blood.
19
2007;110:4606-4613.
20
16.
Sorror M, Storer B, Sandmaier BM, et al. Hematopoietic cell
21
transplantation-comorbidity index and Karnofsky performance status are
22
independent predictors of morbidity and mortality after allogeneic nonmyeloablative
23
hematopoietic cell transplantation. Cancer. 2008;112:1992-2001.
24
17.
Smith AR, Majhail NS, MacMillan ML, et al. Hematopoietic cell transplantation
25
comorbidity index predicts transplantation outcomes in pediatric patients. Blood.
26
2011;117:2728-2734.
27
18.
Farina L, Bruno B, Patriarca F, et al. The hematopoietic cell transplantation
28
comorbidity index (HCT-CI) predicts clinical outcomes in lymphoma and myeloma
29
patients after reduced-intensity or non-myeloablative allogeneic stem cell
30
transplantation. Leukemia. 2009;23:1131-1138.
31
19.
Pollack SM, Steinberg SM, Odom J, Dean RM, Fowler DH, Bishop MR. Assessment
32
of the hematopoietic cell transplantation comorbidity index in non-Hodgkin
33
lymphoma patients receiving reduced-intensity allogeneic hematopoietic stem cell
34
transplantation. Biology of blood and marrow transplantation : journal of the
35
American Society for Blood and Marrow Transplantation. 2009;15:223-230.
36
20.
Mori Y, Teshima T, Kamezaki K, et al. Validation of pretransplantation assessment
21 Page 21 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
of mortality risk score in the outcome of hematopoietic SCT in non-Caucasians. Bone
2
marrow transplantation. 2012;47:1075-1081.
3
21.
Shimoyama M. Diagnostic criteria and classification of clinical subtypes of adult
4
T-cell leukaemia-lymphoma. A report from the Lymphoma Study Group (1984-87).
5
British journal of haematology. 1991;79:428-437.
6
22.
Iacobelli S. Suggestions on the use of statistical methodologies in studies of the
7
European Group for Blood and Marrow Transplantation. Bone marrow
8
transplantation. 2013;48 Suppl 1:S1-37.
9
23.
10 11
strategies for improved prognostic prediction. Statistics in medicine. 1984;3:143-152. 24.
12 13
Harrell FE, Jr., Lee KL, Califf RM, Pryor DB, Rosati RA. Regression modelling Kanda Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone marrow transplantation. 2013;48:452-458.
25.
Terwey TH, Hemmati PG, Martus P, et al. A modified EBMT risk score and the
14
hematopoietic cell transplantation-specific comorbidity index for pre-transplant risk
15
assessment in adult acute lymphoblastic leukemia. Haematologica.
16
2010;95:810-818.
17
26.
Sorror ML, Storb RF, Sandmaier BM, et al. Comorbidity-age index: a clinical
18
measure of biologic age before allogeneic hematopoietic cell transplantation. Journal
19
of clinical oncology : official journal of the American Society of Clinical Oncology.
20
2014;32:3249-3256.
21
27.
22 23
Sorror ML. How I assess comorbidities before hematopoietic cell transplantation.
Blood. 2013;121:2854-2863. 28.
Sorror ML, Logan BR, Zhu X, et al. Prospective Validation of the Predictive Power of
24
the Hematopoietic Cell Transplantation Comorbidity Index: A Center for
25
International Blood and Marrow Transplant Research Study. Biology of blood and
26
marrow transplantation : journal of the American Society for Blood and Marrow
27
Transplantation. 2015;21:1479-1487.
28
29.
Takano K, Fuji S, Uchida N, et al. Pre-transplant diabetes mellitus is a risk factor
29
for non-relapse mortality, especially infection-related mortality, after allogeneic
30
hematopoietic SCT. Bone marrow transplantation. 2015;50:553-558.
31
30.
Kawada H, Yoshimitsu M, Nakamura D, et al. A retrospective analysis of treatment
32
outcomes in adult T cell leukemia/lymphoma patients with aggressive disease
33
treated with or without allogeneic stem cell transplantation: A single-center
34
experience. Biology of blood and marrow transplantation : journal of the American
35
Society for Blood and Marrow Transplantation. 2015;21:696-700.
36
31.
Gahrton G. Risk assessment in haematopoietic stem cell transplantation: impact of
22 Page 22 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
donor-recipient sex combination in allogeneic transplantation. Best practice &
2
research. Clinical haematology. 2007;20:219-229.
3
32.
Fuji S, Yamaguchi T, Inoue Y, et al. Development of a modified prognostic index for
4
patients with aggressive adult T-cell leukemia-lymphoma aged 70 years or younger:
5
possible risk-adapted management strategies including allogeneic transplantation.
6 7
Haematologica. 2017;102:1258-1265.
8
23 Page 23 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Figure 1. Cumulative incidence of non-relapse mortality (NRM) of 824 patients
2
receiving allo-HCT for ATL. A total of 824 ATL patients received allogeneic HCT between
3
2008 and 2013 in Japan.
4 5
Figure 2. HCT-CI is predictive of OS and NRM.
6
(A) OS, and (B) NRM are shown for the HCT-CI risk categories (HCT-CI score 0, low risk;
7
score 1–2, intermediate risk; score ≥3, high risk). The HCT-CI is predictive of OS and NRM.
8
(C) OS, and (D) NRM are shown for the HCT-CI risk categories (HCT-CI score 0, low risk;
9
score 1–3, intermediate risk; score ≥4, high risk).
10 11
Figure 3. mEBMT score is predictive of OS and ATL-related mortality. (A) OS and (B)
12
NRM are shown for the mEBMT risk categories (mEBMT score 0–3, low risk; score 4–6,
13
high risk). The donor/recipient gender combination is predictive of OS and non-ATL-related
14
morality. (C) OS, and (D) NRM are shown for the donor/recipient sex combination.
15 16
Figure 4 Age is predictive of NRM. NRM is shown for the age categories (≤50, 51–55,
17
56–59, 60–63, and ≥64 years).
18 19
Figure 5 ATL-HCT-PI is predictive of NRM and OS, but not of ATL-related mortality. (A)
20
NRM, and (B) OS are shown for the ATL-HCT-PI categories (low, 0; intermediate, 1; high,
21
≥2).
22 23 24
24 Page 24 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 1. Patient and disease characteristics. Characteristics
N = 824
(%)
Acute
522
63.3
Lymphoma
267
32.4
Chronic
25
3
Smoldering
3
0.4
NA
7
0.8
>0–19
0
0
20–39
29
3.5
40–49
122
14.8
50–59
366
44.4
≥60
307
37.3
Median (range)
57
(24–72)
Male
449
54.5
Female
375
45.5
ATL clinical subtype
Age of patients at transplantation years
Sex
Follow-up, months Median (range)
7.5 (0.0–80.0)
PS at transplantation 0
354
43
1
358
43.4
2
85
10.3
3
18
2.2
25 Page 25 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
4
8
1
NA
1
0.1
CR1
305
37
≥CR2
10
1.2
non-CR
501
60.8
NA
8
1
BM
476
57.8
PBSC
135
16.4
BM+PBSC
1
0.1
CB
212
25.7
Match
474
57.5
Mismatch
350
42.5
Myeloablative
302
36.7
RIST
520
63.1
NA
2
0.2
Alive
311
37.7
ATL-related mortality
283
34.3
Non-relapse mortality
229
27.8
NA
1
0.1
Disease status at transplantation
Stem cell source
Serological HLA
Conditioning regimen
Survival/cause of death
1 2
26 Page 26 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 2. Parameters associated with HCT-CI score. HCT-CI Parameter
Score Patients, %
Arrhythmia
1
1.6
Cardiac
1
3.2
IBD
1
0.1
Diabetes
1
7.0
Cerebrovascular
1
0.6
Psychiatric disturbance
1
1.7
Mild hepatic
1
7.8
Obesity
1
1.3
Infection
1
5.5
Rheumatological
2
1.3
Peptic ulcer
2
0.4
Moderate/severe renal
2
1.1
Moderate pulmonary
2
10.3
Prior solid tumor
3
4.6
Heart valve disease
3
0.8
Severe pulmonary
3
2.9
Moderate/severe hepatic
3
2.2
2 3
27 Page 27 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 3. Revised HCT-CI in ATL: univariate analysis of OS, NRM, and ATL-related mortality.
Variables
Patients %
OS HR (95% CI)
NRM P
HR (95% CI)
ATL-related mortality HR
P
(95% CI)
P
HC-CI Low
63.1
1
1
Intermediate
31.1
1.4 (1.16–1.69)
4.43E-04
1.16 (0.87–1.56)
High
5.8
2.75 (2.01–3.77)
3.31E-10
2.52 (1.68–3.78)
1 0.31 8.50E-06
1.36 (1.04-1.70)
0.025
1.42 (0.85–2.36)
0.18
28 Page 28 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 4. Parameters associated with mEBMT score. mEBMT Parameter
Score Patients, %
Age, years
<20
0
0
20–40
1
4.5
>40
2
95.5
CR1
0
37.4
>CR1
1
1.2
No CR
2
61.4
MRD
0
19.7
Other
1
80.3
Other
0
79.3
Female donor-male recipient
1
20.7
Disease stage
Donor type
Gender combination
2 3
29 Page 29 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 5. mEBMT in ATL: univariate analysis of OS, NRM, and ATL-related mortality.
Variables
Patients %
OS HR (95% CI)
NRM P
HR (95% CI)
ATL-related mortality HR
P
(95% CI)
P
mEBMT Low
34.7
High
65.3
1 1.76 (1.43–2.16)
1 7.50E-08
1.03 (0.78–1.37)
1 0.82
2.25 (1.67–3.02)
8.3E-08
30 Page 30 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1 2
Table 6. Multivariate analysis of the effect of pre-transplant variables on
3
non-ATL-related mortality. Variables
Patients
non-ATL mortality
%
HR (95% CI)
ATL-HCT-PI P
points
Revised HCT-CI Low (0) 63.1 Intermediate (1-3) 31.1 High (4≥) 5.8
1
0
1.27
(0.94–1.71)
0.12
1
2.47
(1.64–3.73)
1.7E-05
2
Sex Combination Other
79.3
FDMR
20.7
≤63
84.3
≥64
15.7
1 1.65 (1.23–2.21)
7.4E-04
1
0.047
1
Age, years 1 1.39 (1.00–1.92)
4
31 Page 31 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
32 Page 32 of 32
S1083-8791(17)30819-4 https://doi.org/10.1016/j.bbmt.2017.11.005 YBBMT 54861
To appear in:
Biology of Blood and Marrow Transplantation
Received date: Accepted date:
18-6-2017 6-11-2017
Please cite this article as: Makoto Yoshimitsu, Ryuji Tanosaki, Koji Kato, Takashi Ishida, Ilseung Choi, Yoshifusa Takatsuka, Takahiro Fukuda, Tetsuya Eto, Michihiro Hidaka, Naoyuki Uchida, Toshihiro Miyamoto, Yasuhiro Nakashima, Yukiyoshi Moriuchi, Koji Nagafuji, Yasuhiko Miyazaki, Tatsuo Ichinohe, Minoko Takanashi, Yoshiko Atsuta, Atae Utsunomiya, ATL Working Group of the Japan Society for Hematopoietic Cell Transplantation, Risk Assessment in Adult T-Cell Leukemia/Lymphoma Treated with Allogeneic Hematopoietic Stem Cell Transplantation, Biology of Blood and Marrow Transplantation (2017), https://doi.org/10.1016/j.bbmt.2017.11.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Risk assessment in adult T-cell leukemia/lymphoma treated with allogeneic
2
hematopoietic stem cell transplantation
3 4
Makoto Yoshimitsu1,†, Ryuji Tanosaki2, Koji Kato3, Takashi Ishida4, Ilseung Choi5, Yoshifusa
5
Takatsuka6, Takahiro Fukuda7, Tetsuya Eto8, Michihiro Hidaka9, Naoyuki Uchida10,
6
Toshihiro Miyamoto3, Yasuhiro Nakashima5, Yukiyoshi Moriuchi11, Koji Nagafuji12, Yasuhiko
7
Miyazaki13, Tatsuo Ichinohe14, Minoko Takanashi15, Yoshiko Atsuta16, and Atae
8
Utsunomiya6; ATL Working Group of the Japan Society for Hematopoietic Cell
9
Transplantation.
10 11
1
12
Japan, 2 Department of Transfusion Medicine and Cell Therapy, Keio University Hospital, Tokyo,
13
Japan, Department of Medicine and Biosystemic Science, Kyushu University Graduate
14
School of Medical Science, Fukuoka, Japan, 4Department of Hematology and Oncology,
15
Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, 5Department
16
of Hematology, National Kyushu Cancer Center, Fukuoka, Japan, 6 Department of
17
Hematology, Imamura Bun-in Hospital, Kagoshima, Japan, 7 Department of Hematopoietc
18
Stem Cell Transplantation, National Cancer Center Hospital, Tokyo, Japan, 8Department of
19
Hematology, Hamanomachi Hospital, Fukuoka, Japan, 9Deparment of Hematology, National
20
Hospital Organization Kumamoto Medical Center, Kumamoto, Japan, 10Department of
21
Hematology, Federation of National Public Service Personnel Mutual Aid Associations
22
Toranomon Hospital, Tokyo, Japan, 11Department of Hematology, Sasebo City General
23
Hospital, 12Division of Hematology and Oncology, Department of Medicine, Kurume
24
University Hospital, 13Department of Hematology, Oita Prefectural Hospital, 14Department of
25
Hematology and Oncology, Research Institute for Radiation Biology and Medicine, Hiroshima
26
University, Hiroshima, Japan,
Department of Hematology and Immunology, Kagoshima University Hospital, Kagoshima,
3
15
Blood Service Headquarters, Japanese Red Cross Society,
1 Page 1 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
16
Japanese Data Center for Hematopoietic Cell Transplantation, Nagoya,
1
Tokyo, Japan,
2
Japan
3
†Corresponding author, mailing address: Department of Hematology and Immunology,
4
Kagoshima University Hospital
5
8-35-1 Sakuragaoka, Kagoshima, Japan, 890-8520.
6
Phone: +81-099-275-5934, Fax: +81-99-275-5947.
7
E-mail: [email protected]
8 9 10
Short title; Pre-transplant comorbidities and HCT in ATL
11
Text word count: 4058
12
Abstract word count: 197
13
The number of figures and tables: 5 Figures, 6 Tables
14
The number of references: 32
15 16
The scientific category: TRANSPLANTATION
17
2 Page 2 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Key Points:
2
HCT-CI and mEBMT are useful pre-transplant prognostic factors for patients with ATL.
3
Newly developed ATL HCT-PI may have better prognostic value for patients with ATL.
4 5
3 Page 3 of 32
YOSHIMITSU et al.
1 2
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Highlights: Previously examined factors associated with HCT, were not prognostic.
3
A total of 824 patients were retrospectively examined in this study.
4
HCT-CI is useful pre-transplant prognostic factors for patients with ATL.
5
Newly developed ATL HCT-PI may have better prognostic value for patients with
6
ATL.
7 8
Summary
9
Disease status at allogeneic hematopoietic cell transplantation (HCT) is an important
10
pre-transplant prognostic factor of HCT in adult T-cell leukemia/lymphoma (ATL); however,
11
other prognostic factors, including comorbidities, were not predictive in small cohort
12
analyses. Several scoring systems (HCT-comorbidity index [CI]/modified European Society
13
for Blood and Marrow Transplantation risk score [mEBMT]) have been adopted to predict
14
HCT outcomes in other hematological malignancies. We retrospectively evaluated HCT-CI
15
and mEBMT to predict non-relapse mortality (NRM) in 824 ATL patients registered in the
16
Japan Society for Hematopoietic Cell Transplantation TRUMP database, from 2008 until
17
2013. A higher HCT-CI was associated with greater NRM when comparing HCT-CI 0 vs.
18
HCT-CI 1–3 and HCT-CI 0 vs. HCT-CI ≥4. A higher mEBMT score was not associated with
19
higher NRM when comparing mEBMT 0–3 with 4–6. As ATL patients are older and
20
consequently at risk of additional complications, we developed an optimized prognostic
21
index for ATL (ATL-HCT-PI) using known risk factors: age, HCT-CI, and donor/recipient sex
22
combination. The ATL-HCT-PI scores effectively predicted the 2-year NRM (22.0%, 27.7%,
23
and 44.4%, respectively). Therefore, the newly developed ATL-HCT-PI, in combination with
24
other risk factors, is more useful for predicting NRM in HCT for ATL patients.
25 26
Keywords: allogeneic hematopoietic cell transplantation; comorbidities; adult T-cell
27
leukemia lymphoma; hematological malignancies; mortality.
4 Page 4 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Introduction
2
Adult T-cell leukemia/lymphoma (ATL) is an intractable hematological malignancy
3
caused by the latent infection of human T-cell leukemia virus-I (HTLV-1) 1. The 3-year
4
overall survival (OS) of ATL patients receiving optimal chemotherapy is approximately
5
24% 2. Allogeneic hematopoietic stem cell transplantation (HCT) is an attractive
6
alternative modality that offers long-term remission in some patients with ATL
7
nationwide retrospective analysis showed that the 3-year OS for the entire cohort was
8
33% 5. Notably, the cumulative incidence of treatment-related mortality (TRM) at 3 years
9
in this cohort was very high (approximately 43%). Therefore, there is an urgent need for
10
clinical solutions to decrease the TRM. One possible solution involves a more accurate
11
prediction of the risk of TRM before allo-HCT. Although several prognostic indices for
12
ATL have been reported, these were either determined in ATL patients without allo-HCT,
13
or have never been validated in terms of OS or TRM for patients with ATL receiving
14
allo-HCT
15
ATL undergoing allo-HCT. The risk factors associated with OS or TRM include the use
16
of cord blood as a donor source, development of grade III–IV acute GVHD, presence of
17
extensive chronic graft versus host disease (GVHD), male sex, performance status (PS)
18
2–4, late transplantation from diagnosis, and old age
19
related to pre-transplantation conditions, including comorbidities, have not been
20
thoroughly evaluated in patients with ATL undergoing allo-HCT.
21
The hematopoietic stem cell transplantation comorbidity index (HCT-CI) is widely used
22
to predict TRM 13. The HCT-CI was established using a large data set and was found to
23
be reproducible in many transplantation settings
24
available on whether HCT-CI is suitable for use among lymphoma patients receiving
25
allo-HCT
26
(n = 63), Pollack et al .reported that the use of HCT-CI is not as beneficial in NHL
3, 4
. A
6-8
. Hence, it is unclear whether these indices can be applied to patients with
5, 9-12
. However, other factors
14-17
. However, limited data are
18
. By analyzing a limited number of non-Hodgkin’s lymphoma (NHL) patients
5 Page 5 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
19
1
patients as in leukemia patients
. In addition, Mori et al. reported that the
2
Pre-transplantation Assessment of Mortality (PAM) score is not as useful in patients with
3
ATL receiving allo-HCT, as compared to that in patients with other hematological
4
malignancies 20.
5
In the present study, we tested the ability of HCT-CI to predict non-relapse mortality
6
(NRM) in patients with ATL receiving allo-HCT, by using the largest ATL allo-HCT
7
registry available worldwide. Moreover, we tested the utility of the modified European
8
Society for Blood and Marrow Transplantation risk score (mEBMT) score and other
9
reported pre-transplant risk factors. In addition, we propose a novel optimized
10
pre-transplant prognostic index for patients with ATL receiving allo-HCT, called the ATL
11
hematopoietic stem cell transplantation prognostic index (ATL-HCT-PI). This new
12
prognostic index better predicts NRM and OS than either the HCT-CI or mEBMT score.
13
To our knowledge, this is the largest corroborative study of HCT-CI and mEBMT scores
14
with a single disease entity.
15
6 Page 6 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Methods
2
Data Collection
3
ATL patients who underwent their first allo-HCT between January 1, 2008, and
4
December 31, 2013 (n = 830), when HCT-CI was introduced into the registry data
5
parameters, were considered for this study. Data were collected from the nationwide
6
survey database of the Japan Society for Hematopoietic Cell Transplantation (JSHCT).
7
The data were updated as of December 2014. Five cases with missing HCT-CI data and
8
1 case receiving autologous HCT were excluded. Therefore, a total of 824 patients were
9
finally included in this analysis. Data collected for analysis included the patients’ clinical
10
characteristics, such as HCT-CI parameters,13 ATL clinical subtype,21 age at HCT, sex,
11
donor sex, survival data at the last follow-up, PS at HCT, disease status at HCT, stem
12
cell sources, HLA typing in serological resolution, ABO blood group, conditioning
13
regimen, date at HCT, donor relationship with the recipient (related or unrelated), date
14
of death, and primary cause of death. The study was approved by the data
15
management committees of the JSHCT, as well as by the institutional ethical committee
16
of Kagoshima University Hospital.
17
Definitions
18
OS was calculated from the date of HCT until the date of death or last follow-up. The
19
registered causes of death were categorized as either ATL-related or Non-relapse
20
mortality (NRM). ATL-related mortality was defined as death with relapse or progression
21
of ATL in patients, based on the judgment of each registered physician. NRM was
22
defined as any death other than ATL-related mortality. If the primary cause of death was
23
GVHD or infection after the relapse or progression of ATL, for example, we classified
24
such a case as ATL-related mortality. The need for a myeloablative conditioning
25
regimen (MAC) and reduced intensity conditioning regimen (RIC) was registered based
26
on the judgment of each physician. An HLA-match was defined when the serologic 7 Page 7 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
resolution typing for HLA-A, HLA-B, and HLA-DR were identical.
2 3
Statistical analysis
4
The probability of OS was estimated according to the Kaplan-Meier method. The
5
incidence of NRM was calculated using cumulative incidence estimates with the
6
competing risk model, while considering ATL-related mortality as a competitive event 22.
7
Univariate Cox proportional regression analysis was used to estimate the risk factors for
8
OS. The variables tested in the univariate analysis for transplant outcomes are PS, sex,
9
donor-recipient sex combination, ABO mismatch, serological HLA status, and age.
10
Multivariate analysis was performed while accounting for the competing risk structure.
11
Pre-transplant variables were included in the multivariate analysis model when a
12
significance level of P ≤ 0.1 was detected on univariate Cox regression analysis and
13
Gray’s test. A significance level of P ≤ .05 was required in the multivariate analysis. Fine
14
and Gray proportional hazard modeling was used to estimate the effect of the same
15
variables used in the univariate analysis of NRM and ATL-related mortality on the
16
cumulative incidence of NRM and ATL-related mortality. A backward selection algorithm
17
was used to select the final model.
18
Development of the new prognostic index
19
In order to develop the new ATL-HCT-PI, integer weights for the ATL-HCT-PI were
20
derived from proportional hazards with NRM over the first 2 years. Therefore, NRM was
21
used for the development of the scores instead of OS, similar to that used for the
22
original HCT-CI. This was chosen since deaths from non-ATL-related causes are more
23
likely to be influenced by pre-transplant comorbidities than deaths from ATL itself, which
24
was treated as a competing risk. The adjusted hazard ratios (HRs) for NRM over the
25
first 2 years after HCT were calculated for each extracted factor from the multivariate
26
analysis. The adjusted HRs were converted to integer weights as follows: factors with 8 Page 8 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
an adjusted HR of ≤1.25 were excluded from consideration, factors with an adjusted HR
2
of 1.26–2.0 were assigned a weight of 1, and factors with an adjusted HR of 2.1–3.0
3
were assigned a weight of 2. The ATL-HCT-PI score was calculated as the sum of these
4
integer weights. The ATL-HCT-PI scores were further collapsed into 3 risk groups: 0
5
(low risk), 1 (intermediate risk), and ≥2 (high risk). C-statistics estimated the predictive
6
capacity of the HCT-CI, mEBMT score, and newly developed ATL HCT-PI, as described
7
23
. All statistical analyses were performed using EZR 24.
8
9 Page 9 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Results
2
Patient characteristics
3
Table 1 summarizes the characteristics of the 824 patients with ATL. At the time of
4
analysis — i.e., after a median follow-up for survivors of 27.8 months (range, 0–80
5
months) and after a median follow-up for all patients of 7.5 months (range, 0–80
6
months) — 311 patients (37.7%) remained alive. The projected OS of the whole cohort
7
at 2 years was 37.5%, with a median OS of 8.6 months (95% CI, 7.1–10.6). The
8
cumulative incidence of NRM at 2 years was 27.5% (95% CI, 24.4–30.7%; Fig 1).
9
HCT-CI
10
The HCT-CI was scored based on the judgment of each registered physician.
11
Co-morbidities, defined by the HCT-CI criteria, were identified in 304 (36.9%) of all
12
patients, with 102 (12.4%), 88 (10.7%), 66 (8.0%), and 48 (5.8%) presenting with scores
13
1, 2, 3, and ≥4, respectively. The main comorbidities in the registered data included
14
moderate pulmonary disease, mild hepatic disease, diabetes, and infection (Table 2).
15
Based on the original HCT-CI risk classification (score 0, low risk; score 1–2,
16
intermediate risk; score ≥3, high risk) 13, 63.1% of the patients were assigned to the
17
low-risk group, 23.1% were assigned to the intermediate-risk group, and 13.8% were
18
assigned to the high-risk group. When the OS was compared among these risk groups,
19
we observed a decreasing trend in OS in patients with a higher HCT-CI. However, this
20
difference was not statistically significant between the intermediate-risk and high-risk
21
groups (2-year OS of 42.8%, 29.8%, and 26.7% for HCT-CI low-, intermediate-, and
22
high-risk patients, respectively; Fig 2A). The HR was 1.52 (95% CI, 1.24–1.87) for
23
intermediate-risk patients and 1.62 (95% CI, 1.27–2.06) for high-risk patients in
24
comparison with the low risk patients. Moreover, among patients with higher HCT-CI,
25
the NRM was greater but the difference was marginal among groups (2-year NRM of
26
25.3%, 29.2%, and 34.2% for HCT-CI low-, intermediate-, and high-risk patients, 10 Page 10 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
respectively; Fig 2B). However, when other cut-off points for the definition of risk groups
2
(score 0, low risk, score 1–3: intermediate risk, score ≥4: high risk) were used, there
3
was a considerable change in the results. When the OS was compared for these
4
re-stratified risk groups (revised HCT-CI), we observed a decreasing trend in OS in
5
patients with higher revised HCT-CI (2-year OS of 42.8%, 32.7%, and 7.8% in the
6
revised HCT-CI low-, intermediate-, and high-risk patients; Fig 2C; Table 3). Moreover,
7
among the patients with higher revised HCT-CI, NRM was greater in the high-risk
8
patients (2-year NRM of 25.3%, 26.8%, and 52.6% for revised HCT-CI low-,
9
intermediate-, and high-risk patients, respectively; Fig 2D; Table 3). The ATL-related
10
mortality was comparable among the 3 risk groups (Table 3).
11
Among the significant HCT-CI factors (cardiac function, infection, liver function,
12
pulmonary function, P<0.10 by univariate analysis), only severe pulmonary dysfunction
13
had a significant influence on NRM in multivariate analysis (HR, 2.06 [95% CI,
14
1.23–3.47]).
15 16 17
mEBMT Thereafter, the mEBMT risk score was determined. The risk score accounts for
18
recipient age, disease stage, donor/host histocompatibility, and donor/recipient gender
19
combination 25.The prevalence of individual factors according to the definition of the
20
mEBMT is shown in Table 4. The median mEBMT risk score was 4 (range, 1–6), with
21
0.4%, 5.9%, 25.0%, 16.1%, 24.0%, 8.7%, and 9.8% of patients scoring 1 or 2, 3, 4, 5, 6,
22
or NA (not available because of lack of data), respectively. Based on the mEBMT risk
23
classification (score 0–3, low; score 4–6, high), 34.7% of patients were assigned to the
24
low-risk group. When the OS was compared among these risk groups, we observed a
25
significant decrease in OS in patients with a higher mEBMT score (2-year OS of 49.6%
26
and 32.6% for mEBMT low-risk and high-risk patients, respectively; Fig 3A, Table 5). 11 Page 11 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
The NRM was comparable between low-risk and high-rsik patients (2-year NRM of
2
28.7% and 27.7% for mEBMT low- and high-risk patients, respectively; Fig 3B, Table 5).
3
The ATL-related mortality was greater in high-risk patients (2-year ATL-related mortality
4
of 21.9% and 39.7% for mEBMT low- and high-risk patients, respectively; Table 5).
5
Among all the mEBMT parameters, donor/recipient gender combination was found to
6
have a significant influence on NRM in multivariate analysis (HR, 1.71 [95% CI,
7
1.28–2.29]. We further analyzed this donor/recipient gender combination in detail, and
8
found that the combination of female donor to male recipient exhibited consistently
9
worse OS (Figure 3C) and NRM (Fig 3D), as compared with the other combinations.
10 11
PS
12
PS is a well-known risk factor for ATL 7, 8, 10. In this cohort, the median PS was 1 (range,
13
0–4). Based on the previous PS risk classification (score 0-1, low risk; score 2–4, high
14
risk)
15
compared among these risk groups, we observed a significant decreasing trend in OS
16
in patients with a higher PS (2-year OS of 40.8% [95% CI, 36.9–44.6%] and 17.6%
17
[10.9–25.7%] for PS low- and high-risk patients, respectively). The HR was 2.26 (95%
18
CI, 1.80–2.837) for high-risk patients in comparison with the low-risk patients.
19
Interestingly, the NRM was comparable between PS low-risk and PS high-risk patients
20
(2-year non-ATL-related mortality of 27.5% [95% CI, 24.1-30.9%] and 27.9%
21
[19.6–36.7%] for PS low- and high-risk patients, respectively). The ATL-related mortality
22
was significantly greater in PS high-risk patients (2-year ATL-related mortality of 31.9%
23
and 54.5% for PS low- and high-risk patients, respectively), with a HR of 2.36 (95% CI,
24
1.73–3.21) for PS high-risk patients in comparison with the low-risk patients.
86.4% of patients were assigned to the low-risk group 8. When the OS was
25 26
Age 12 Page 12 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Age is a major factor for assessing the suitability of allogeneic HCT. The
2
comorbidity-age index, based on the HCT-CI, was recently proposed to facilitate clinical
3
decision making and comparative-effectiveness research of HCT 26. In that cohort of
4
3033 patients, the median age was 45 years. In our cohort of 824 ATL patients, the
5
median age was 57 years (range, 24–72 years), and 96.5% of patients were aged >40
6
years. Although age serves as a parameter for the mEBMT score estimation, we found
7
that the score could be divided into 3 groups (age <19 years, 20–40 years, and >41
8
years), with 95.5% of our cohort was aged >41 years. Therefore, age does not enable
9
pre-transplant risk stratification in patients with ATL. An earlier study of 386 ATL
10
patients with HCT indicated that older age (>50 years) was significantly associated with
11
a lower survival rate 5, whereas another study of 586 ATL patients with HCT showed
12
that older age (>55 years) was not a significant risk factor of therapy-related mortality 10.
13
In our larger cohort, patients were sorted by age and divided into 5 groups based on
14
almost equal number of NRM events (total 229 events) (≤50 years, n = 42; 51–55 years,
15
n = 50; 56–59 years, n = 47; 60–63 years, n = 45; ≥64 years, n = 45). Although
16
increasing age was associated with higher TRM, no difference was observed between
17
the age groups of ≤50 years, 51–55, 56–59, and 60–63 years (Fig. 4). Compared with
18
patients in the age group of ≤50 years, those in the age groups 51-55, 56-59, 60-63, and
19
≥64 years had HRs for NRM of 1.17 (95% CI, 0.78–1.75, P = 0.46), 1.25 (95% CI,
20
0.82–1.89, P = 0.30), 1.10 (95% CI, 0.73–1.69, P = 0.63) and 1.60 (95% CI,1.06–2.41,
21
P =0.025), respectively.
22 23
ATL-HCT-PI
24
Among the valuables tested in the univariate analysis for NRM, HCT-CI, donor/recipient
25
sex combination, and age over 64 were determined as independent risk factors in
26
multivariate analysis (Table 6). Based on these factors, we developed a new composite 13 Page 13 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
pre-transplant prognostic index for ATL, namely the ATL-HCT-PI. We assigned points
2
for each independent risk factor, as described previously. Thereafter, we divided the
3
cohort into 3 risk groups: low risk, intermediate risk, and high risk. The cases with an
4
ATL-HCT-PI score of 0 were assigned to the low-risk group, 1 were assigned to the
5
intermediate-risk group, and ≥2 were assigned to the high-risk group. Based on this
6
stratification, NRM was elevated in the intermediate-risk and high-risk groups (2-year
7
NRM of 22.0%, 27.7%, and 44.4% for ATL-HCT-PI low-, intermediate-, and high-risk
8
patients, respectively; Fig 5A), with a HR of 1.46 (95% CI, 1.06–2.00) for
9
intermediate-risk patients and 2.48 (95% CI, 1.76–3.49) for high-risk patients in
10
comparison with the low-risk patients. This index marginally predict ATL-related
11
mortality, as the HR was 1.18 (95% CI, 0.89–1.55; P = 0.25) and 1.42 (95% CI,
12
1.01–2.01; P = 0.046). When the OS was compared among these risk groups, we
13
observed a decreasing trend in the OS of patients with a higher ATL-HCT-PI (2-year OS
14
of 47.3%, 39.1%, and 15.7% for ATL-HCT-PI low-, intermediate-, and high-risk patients,
15
respectively; Fig 5B). The HR was 1.43 (95% CI, 1.16–1.76) for intermediate-risk
16
patients and 2.48 (95% CI, 1.94–3.16) for high-risk patients, in comparison with the
17
low-risk patients.
18
When comparing the OS by ATL-HCT-PI with the OS by HCT-CI using c-statistics, we
19
found that ATL-HCT-PI has a higher c-statistic than HCT-CI, with a better distribution
20
(0.629 and 0.582, respectively; P < 0.001).
21
14 Page 14 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1 2 3
Discussion In
the
present
study,
we
assessed
whether
the
patient-,
disease-,
and
4
transplantation-related variables and pre-transplant comorbidities, represented by
5
HCT-CI and the modified EBMT risk score, enhanced the prognostication of NRM,
6
ATL-related mortality, and OS in a cohort of 824 ATL patients treated in Japan between
7
2008 and 2013. To our knowledge, this score has not been analyzed in a large cohort of
8
ATL patients. Our analysis showed that HCT-CI could identify comorbid conditions in
9
only 36.9% of patients, which is relatively smaller than that reported previously 13, 14
10
(52%–80%)
11
remains a concern, since a brief training program is usually proposed for facilitating
12
reliable assessment of comorbidities
13
study that examined 8115 patients treated with allogeneic SCT reported that 48% of the
14
registered patients had a HCT-CI score of 028. In that cohort, patients were older and
15
were more refractory to chemotherapy prior to allo-HCT. Therefore, it is possible that the
16
patients with more comorbidity were ineligible to undergo allo-HCT.
. The consistency in comorbidity coding across different evaluators
27
. However, a recent large prospective validation
17
Pulmonary disease, hepatic disease, diabetes, and infection had the highest
18
prevalence in our cohort, whereas the other comorbidities were rarely observed. Among
19
the comorbidities included in the HCT-CI, only severe pulmonary comorbidities were
20
found to be predictive of increased NRM; the remaining comorbidities had no significant
21
prognostic value on NRM by multivariate analysis. When analyzing the composite score,
22
we observed a trend for inferior OS and higher NRM in cases with higher HCT-CI.
23
However, prognostic significance was not achieved when risk was stratified according to
24
the original report (score 0, low; score 1–2, intermediate; score ≥3, high; Fig 2A-B).
25
Nevertheless, when the risk was stratified as follows—score 0, low; score 1–3,
26
intermediate; score ≥4, high—we observed a significantly inferior OS and higher NRM in 15 Page 15 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
cases with higher risk (Fig 2C-D). This difference in our population could be attributed to
2
the specific characteristics of our patient cohort. The prevalence of comorbidities, age
3
distribution, and preconditioning regimen were different in our population. In addition, a
4
recent study on the prospective validation of the predictive power of HCT-CI in a large
5
cohort indicated that OS and NRM between cases with HCT-CI scores of 2 and 3 were
6
almost identical in the allogeneic SCT setting
7
sub-analysis was focused on reduced intensity/non-myeloablative conditioning, no
8
difference was observed in the NRM between HCT-CI score 0 and 1–2. Takano et al.
9
also reported that patients with diabetes mellitus as a comorbidity had different
28
. In the present study, when the
29
10
outcomes of NRM in a large Japanese registry study (n = 7626)
11
diabetes mellitus, the outcomes of NRM between an HCT-CI score of 1 and of 2–3 did
12
not differ. However, the reasons underlying these results are unclear. The predictive
13
value of HCT-CI may depend on specific patient characteristics and treatment protocols.
14
Accordingly, it is important to note that median age in our study (57 years) was
15
considerably greater than that in the original report (44.8 years)
16
influence of comorbidities on the overall treatment outcome might depend on the
17
analyzed disease, with a possible loss of predictive value in groups with many high
18
relapse-risk patients. Therefore, the application of this index in a disease- and
19
population-based validation analysis would be warranted. In the present study, the
20
population was older, and only 5.8% of the patients were at high risk. These findings
21
highlight the need for a new pre-transplant risk score that combines age and other
22
known pre-transplantation risk factors in ATL.
. In patients with
13
. Furthermore, the
23
The mEBMT risk score was established for acute lymphoblastic leukemia patients,
24
based on the registry data 25. This score accounts for the recipient’s age, disease status,
25
donor/host histocompatibility, and donor/recipient sex combination. The mEBMT score
26
had a good prognostic value for OS and ATL-related mortality. However, there were no 16 Page 16 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
prognostic value of mEBMT score for NRM. In the analysis of single risk score
2
parameters, although disease status, age, and donor type had a significant influence on
3
OS, we found that only donor/recipient gender combination had a significant influence
4
on NRM in the multivariate analysis. Disease status is a well-known risk factor for OS
5
and ATL-related mortality in patients with ATL receiving all-HCT 5, 10, 30. Older age is also
6
a known risk factor for OS in patients with ATL receiving allo-HCT
7
95.5% of registered patients were aged >40 years, and previous reports stratified age
8
groups at 50 or 55 years for better discrimination. Hence, we decided to re-stratify the
9
age for NRM. Accordingly, we divided the patient cohort into 5 groups. Increasing age
10
was associated with higher NRM; however, no difference was observed between the
11
age groups of ≤50, 51-55, 56–59, and 60–63 years (Fig 4). To our knowledge, this is the
12
first study to analyze the influence of age on NRM in patients with ATL receiving
13
allo-HCT.
14
Interestingly, the donor/recipient gender combination had a significant influence on
15
NRM. We further analyzed this factor in detail, and found that only the combination of
16
female donor and male recipient had a significant impact on the OS and NRM (Fig
17
3D-E). Gahrton reviewed the impact of donor and recipient gender combination in
18
allo-HCT for various hematological disorders and reported that transplantation-related
19
mortality in the female donor and male recipient combination was higher than in other
20
combinations 31. Our result was consistent with those findings. To our knowledge, this is
21
the first study to validate that the NRM in the female donor and male recipient
22
combination is higher in patients with ATL receiving allo-HCT. Previous reports indicated
23
that male sex was a poor risk factor for patients with ATL receiving allo-HCT
24
However, the donor/recipient combination might be more important, since our large
25
cohort study indicated that the NRM in male donor/male recipient is identical to that in
26
the female recipient group.
5, 10
. In this cohort,
5, 10
.
17 Page 17 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Unlike to previous reports, the pre-transplant PS was not a risk factor for predicting 10
. Reliable pre-transplant risk stratification using comorbidities and other factors
2
NRM
3
has been found to be essential for comprehensive risk adjustment in patients with
4
cancer. Fuji et al. recently reported a new prognostic index for ATL patients aged 70
5
years or younger. This prognostic index consists of the clinical subtype, ECOG PS at
6
diagnosis, calcium levels, CRP levels, and soluble interleukin-2 receptor status32. Even
7
though this index is useful for predicting OS of the transplanted group, those factors
8
mainly reflect the aggressiveness of ATL itself. This is in contrast to our report, were we
9
focused on the validation of 2 commonly used predictive models in the HCT setting and
10
developed a new index based on mainly non-ATL disease factors. We incorporated
11
comorbidities with age and donor/recipient gender combination in order to develop a
12
new pre-transplant risk stratification method in patients with ATL (ATL-HCT-PI; Figure
13
5). We also identified the prevalent and prognostically important pre-transplant
14
conditions that should be measured and considered in clinical trials and HCT registries.
15
Furthermore, we noted that ATL-HCT-PI can be widely used by referring hematologists
16
when assessing the risks of HCT in patients with ATL.
17
In summary, this is the first study to assess the utility of 2 commonly used predictive
18
models in the HCT setting (the HCT-CI and the EBMT score) in a large population of
19
allo-HCT recipients with ATL. We found that HCT-CI has validity as a comorbidity
20
instrument in patients with ATL. Furthermore, we developed a simple tool for capturing
21
pre-transplant risk based on comorbidities and other known factors, which can be used
22
to predict outcomes and stratify patients for HCT. In the future, we will further
23
substantiate this index by using a validation cohort.
24
18 Page 18 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Acknowledgement
2
The authors thank all the physicians and data managers at the institutions that contributed
3
valuable data on transplantation for ATL at the JSHCT. The authors also thank all the
4
members of the data management committees of the JSHCT.
5 6
Authorship
7
M.Y., R.T., K.K., T.I., C.I., and A.U. designed the research, organized the project, and wrote
8
the paper; M.Y. performed statistical analysis; Y.T., T.F., T.E., M.H., N.U., T.M., Y.N.,
9
Y.Moriuchi., K.N.,Y.Miyazaki., T.I., M.T., and Y.A. gathered the data; M.Y., R.T, K.K, T.I, C.I.,
10
and A.U. wrote the draft of the paper; and all the authors contributed to the final version.
11 12
Conflict-of-interest disclosure
13
All authors declare no competing financial interest.
14
19 Page 19 of 32
YOSHIMITSU et al.
1
References
2
1.
3 4
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Ishitsuka K, Tamura K. Human T-cell leukaemia virus type I and adult T-cell leukaemia-lymphoma. The Lancet. Oncology. 2014;15:e517-526.
2.
Tsukasaki K, Utsunomiya A, Fukuda H, et al. VCAP-AMP-VECP compared with
5
biweekly CHOP for adult T-cell leukemia-lymphoma: Japan Clinical Oncology Group
6
Study JCOG9801. Journal of clinical oncology : official journal of the American
7
Society of Clinical Oncology. 2007;25:5458-5464.
8
3.
9
leukemia/lymphoma with allogeneic hematopoietic stem cell transplantation. Bone
10 11
Utsunomiya A, Miyazaki Y, Takatsuka Y, et al. Improved outcome of adult T cell
marrow transplantation. 2001;27:15-20. 4.
Okamura J, Utsunomiya A, Tanosaki R, et al. Allogeneic stem-cell transplantation
12
with reduced conditioning intensity as a novel immunotherapy and antiviral
13
therapy for adult T-cell leukemia/lymphoma. Blood. 2005;105:4143-4145.
14
5.
Hishizawa M, Kanda J, Utsunomiya A, et al. Transplantation of allogeneic
15
hematopoietic stem cells for adult T-cell leukemia: a nationwide retrospective study.
16
Blood. 2010;116:1369-1376.
17
6.
Suzumiya J, Ohshima K, Tamura K, et al. The International Prognostic Index
18
predicts outcome in aggressive adult T-cell leukemia/lymphoma: analysis of 126
19
patients from the International Peripheral T-Cell Lymphoma Project. Annals of
20
oncology : official journal of the European Society for Medical Oncology / ESMO.
21
2009;20:715-721.
22
7.
Katsuya H, Yamanaka T, Ishitsuka K, et al. Prognostic index for acute- and
23
lymphoma-type adult T-cell leukemia/lymphoma. Journal of clinical oncology :
24
official journal of the American Society of Clinical Oncology. 2012;30:1635-1640.
25
8.
Fukushima T, Nomura S, Shimoyama M, et al. Japan Clinical Oncology Group
26
(JCOG) prognostic index and characterization of long-term survivors of aggressive
27
adult T-cell leukaemia-lymphoma (JCOG0902A). British journal of haematology.
28
2014;166:739-748.
29
9.
Kanda J, Hishizawa M, Utsunomiya A, et al. Impact of graft-versus-host disease on
30
outcomes after allogeneic hematopoietic cell transplantation for adult T-cell
31
leukemia: a retrospective cohort study. Blood. 2012;119:2141-2148.
32
10.
Ishida T, Hishizawa M, Kato K, et al. Allogeneic hematopoietic stem cell
33
transplantation for adult T-cell leukemia-lymphoma with special emphasis on
34
preconditioning regimen: a nationwide retrospective study. Blood.
35
2012;120:1734-1741.
36
11.
Ishida T, Hishizawa M, Kato K, et al. Impact of graft-versus-host disease on
20 Page 20 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
allogeneic hematopoietic cell transplantation for adult T cell leukemia-lymphoma
2
focusing on preconditioning regimens: nationwide retrospective study. Biology of
3
blood and marrow transplantation : journal of the American Society for Blood and
4
Marrow Transplantation. 2013;19:1731-1739.
5
12.
Fuji S, Fujiwara H, Nakano N, et al. Early application of related SCT might improve
6
clinical outcome in adult T-cell leukemia/lymphoma. Bone marrow transplantation.
7
2016;51:205-211.
8
13.
9
(HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic
10 11
Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation HCT. Blood. 2005;106:2912-2919.
14.
Sorror ML, Sandmaier BM, Storer BE, et al. Comorbidity and disease status based
12
risk stratification of outcomes among patients with acute myeloid leukemia or
13
myelodysplasia receiving allogeneic hematopoietic cell transplantation. Journal of
14
clinical oncology : official journal of the American Society of Clinical Oncology.
15
2007;25:4246-4254.
16
15.
Sorror ML, Giralt S, Sandmaier BM, et al. Hematopoietic cell transplantation
17
specific comorbidity index as an outcome predictor for patients with acute myeloid
18
leukemia in first remission: combined FHCRC and MDACC experiences. Blood.
19
2007;110:4606-4613.
20
16.
Sorror M, Storer B, Sandmaier BM, et al. Hematopoietic cell
21
transplantation-comorbidity index and Karnofsky performance status are
22
independent predictors of morbidity and mortality after allogeneic nonmyeloablative
23
hematopoietic cell transplantation. Cancer. 2008;112:1992-2001.
24
17.
Smith AR, Majhail NS, MacMillan ML, et al. Hematopoietic cell transplantation
25
comorbidity index predicts transplantation outcomes in pediatric patients. Blood.
26
2011;117:2728-2734.
27
18.
Farina L, Bruno B, Patriarca F, et al. The hematopoietic cell transplantation
28
comorbidity index (HCT-CI) predicts clinical outcomes in lymphoma and myeloma
29
patients after reduced-intensity or non-myeloablative allogeneic stem cell
30
transplantation. Leukemia. 2009;23:1131-1138.
31
19.
Pollack SM, Steinberg SM, Odom J, Dean RM, Fowler DH, Bishop MR. Assessment
32
of the hematopoietic cell transplantation comorbidity index in non-Hodgkin
33
lymphoma patients receiving reduced-intensity allogeneic hematopoietic stem cell
34
transplantation. Biology of blood and marrow transplantation : journal of the
35
American Society for Blood and Marrow Transplantation. 2009;15:223-230.
36
20.
Mori Y, Teshima T, Kamezaki K, et al. Validation of pretransplantation assessment
21 Page 21 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
of mortality risk score in the outcome of hematopoietic SCT in non-Caucasians. Bone
2
marrow transplantation. 2012;47:1075-1081.
3
21.
Shimoyama M. Diagnostic criteria and classification of clinical subtypes of adult
4
T-cell leukaemia-lymphoma. A report from the Lymphoma Study Group (1984-87).
5
British journal of haematology. 1991;79:428-437.
6
22.
Iacobelli S. Suggestions on the use of statistical methodologies in studies of the
7
European Group for Blood and Marrow Transplantation. Bone marrow
8
transplantation. 2013;48 Suppl 1:S1-37.
9
23.
10 11
strategies for improved prognostic prediction. Statistics in medicine. 1984;3:143-152. 24.
12 13
Harrell FE, Jr., Lee KL, Califf RM, Pryor DB, Rosati RA. Regression modelling Kanda Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone marrow transplantation. 2013;48:452-458.
25.
Terwey TH, Hemmati PG, Martus P, et al. A modified EBMT risk score and the
14
hematopoietic cell transplantation-specific comorbidity index for pre-transplant risk
15
assessment in adult acute lymphoblastic leukemia. Haematologica.
16
2010;95:810-818.
17
26.
Sorror ML, Storb RF, Sandmaier BM, et al. Comorbidity-age index: a clinical
18
measure of biologic age before allogeneic hematopoietic cell transplantation. Journal
19
of clinical oncology : official journal of the American Society of Clinical Oncology.
20
2014;32:3249-3256.
21
27.
22 23
Sorror ML. How I assess comorbidities before hematopoietic cell transplantation.
Blood. 2013;121:2854-2863. 28.
Sorror ML, Logan BR, Zhu X, et al. Prospective Validation of the Predictive Power of
24
the Hematopoietic Cell Transplantation Comorbidity Index: A Center for
25
International Blood and Marrow Transplant Research Study. Biology of blood and
26
marrow transplantation : journal of the American Society for Blood and Marrow
27
Transplantation. 2015;21:1479-1487.
28
29.
Takano K, Fuji S, Uchida N, et al. Pre-transplant diabetes mellitus is a risk factor
29
for non-relapse mortality, especially infection-related mortality, after allogeneic
30
hematopoietic SCT. Bone marrow transplantation. 2015;50:553-558.
31
30.
Kawada H, Yoshimitsu M, Nakamura D, et al. A retrospective analysis of treatment
32
outcomes in adult T cell leukemia/lymphoma patients with aggressive disease
33
treated with or without allogeneic stem cell transplantation: A single-center
34
experience. Biology of blood and marrow transplantation : journal of the American
35
Society for Blood and Marrow Transplantation. 2015;21:696-700.
36
31.
Gahrton G. Risk assessment in haematopoietic stem cell transplantation: impact of
22 Page 22 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
donor-recipient sex combination in allogeneic transplantation. Best practice &
2
research. Clinical haematology. 2007;20:219-229.
3
32.
Fuji S, Yamaguchi T, Inoue Y, et al. Development of a modified prognostic index for
4
patients with aggressive adult T-cell leukemia-lymphoma aged 70 years or younger:
5
possible risk-adapted management strategies including allogeneic transplantation.
6 7
Haematologica. 2017;102:1258-1265.
8
23 Page 23 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
Figure 1. Cumulative incidence of non-relapse mortality (NRM) of 824 patients
2
receiving allo-HCT for ATL. A total of 824 ATL patients received allogeneic HCT between
3
2008 and 2013 in Japan.
4 5
Figure 2. HCT-CI is predictive of OS and NRM.
6
(A) OS, and (B) NRM are shown for the HCT-CI risk categories (HCT-CI score 0, low risk;
7
score 1–2, intermediate risk; score ≥3, high risk). The HCT-CI is predictive of OS and NRM.
8
(C) OS, and (D) NRM are shown for the HCT-CI risk categories (HCT-CI score 0, low risk;
9
score 1–3, intermediate risk; score ≥4, high risk).
10 11
Figure 3. mEBMT score is predictive of OS and ATL-related mortality. (A) OS and (B)
12
NRM are shown for the mEBMT risk categories (mEBMT score 0–3, low risk; score 4–6,
13
high risk). The donor/recipient gender combination is predictive of OS and non-ATL-related
14
morality. (C) OS, and (D) NRM are shown for the donor/recipient sex combination.
15 16
Figure 4 Age is predictive of NRM. NRM is shown for the age categories (≤50, 51–55,
17
56–59, 60–63, and ≥64 years).
18 19
Figure 5 ATL-HCT-PI is predictive of NRM and OS, but not of ATL-related mortality. (A)
20
NRM, and (B) OS are shown for the ATL-HCT-PI categories (low, 0; intermediate, 1; high,
21
≥2).
22 23 24
24 Page 24 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 1. Patient and disease characteristics. Characteristics
N = 824
(%)
Acute
522
63.3
Lymphoma
267
32.4
Chronic
25
3
Smoldering
3
0.4
NA
7
0.8
>0–19
0
0
20–39
29
3.5
40–49
122
14.8
50–59
366
44.4
≥60
307
37.3
Median (range)
57
(24–72)
Male
449
54.5
Female
375
45.5
ATL clinical subtype
Age of patients at transplantation years
Sex
Follow-up, months Median (range)
7.5 (0.0–80.0)
PS at transplantation 0
354
43
1
358
43.4
2
85
10.3
3
18
2.2
25 Page 25 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
4
8
1
NA
1
0.1
CR1
305
37
≥CR2
10
1.2
non-CR
501
60.8
NA
8
1
BM
476
57.8
PBSC
135
16.4
BM+PBSC
1
0.1
CB
212
25.7
Match
474
57.5
Mismatch
350
42.5
Myeloablative
302
36.7
RIST
520
63.1
NA
2
0.2
Alive
311
37.7
ATL-related mortality
283
34.3
Non-relapse mortality
229
27.8
NA
1
0.1
Disease status at transplantation
Stem cell source
Serological HLA
Conditioning regimen
Survival/cause of death
1 2
26 Page 26 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 2. Parameters associated with HCT-CI score. HCT-CI Parameter
Score Patients, %
Arrhythmia
1
1.6
Cardiac
1
3.2
IBD
1
0.1
Diabetes
1
7.0
Cerebrovascular
1
0.6
Psychiatric disturbance
1
1.7
Mild hepatic
1
7.8
Obesity
1
1.3
Infection
1
5.5
Rheumatological
2
1.3
Peptic ulcer
2
0.4
Moderate/severe renal
2
1.1
Moderate pulmonary
2
10.3
Prior solid tumor
3
4.6
Heart valve disease
3
0.8
Severe pulmonary
3
2.9
Moderate/severe hepatic
3
2.2
2 3
27 Page 27 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 3. Revised HCT-CI in ATL: univariate analysis of OS, NRM, and ATL-related mortality.
Variables
Patients %
OS HR (95% CI)
NRM P
HR (95% CI)
ATL-related mortality HR
P
(95% CI)
P
HC-CI Low
63.1
1
1
Intermediate
31.1
1.4 (1.16–1.69)
4.43E-04
1.16 (0.87–1.56)
High
5.8
2.75 (2.01–3.77)
3.31E-10
2.52 (1.68–3.78)
1 0.31 8.50E-06
1.36 (1.04-1.70)
0.025
1.42 (0.85–2.36)
0.18
28 Page 28 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 4. Parameters associated with mEBMT score. mEBMT Parameter
Score Patients, %
Age, years
<20
0
0
20–40
1
4.5
>40
2
95.5
CR1
0
37.4
>CR1
1
1.2
No CR
2
61.4
MRD
0
19.7
Other
1
80.3
Other
0
79.3
Female donor-male recipient
1
20.7
Disease stage
Donor type
Gender combination
2 3
29 Page 29 of 32
YOSHIMITSU et al.
1
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
Table 5. mEBMT in ATL: univariate analysis of OS, NRM, and ATL-related mortality.
Variables
Patients %
OS HR (95% CI)
NRM P
HR (95% CI)
ATL-related mortality HR
P
(95% CI)
P
mEBMT Low
34.7
High
65.3
1 1.76 (1.43–2.16)
1 7.50E-08
1.03 (0.78–1.37)
1 0.82
2.25 (1.67–3.02)
8.3E-08
30 Page 30 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1 2
Table 6. Multivariate analysis of the effect of pre-transplant variables on
3
non-ATL-related mortality. Variables
Patients
non-ATL mortality
%
HR (95% CI)
ATL-HCT-PI P
points
Revised HCT-CI Low (0) 63.1 Intermediate (1-3) 31.1 High (4≥) 5.8
1
0
1.27
(0.94–1.71)
0.12
1
2.47
(1.64–3.73)
1.7E-05
2
Sex Combination Other
79.3
FDMR
20.7
≤63
84.3
≥64
15.7
1 1.65 (1.23–2.21)
7.4E-04
1
0.047
1
Age, years 1 1.39 (1.00–1.92)
4
31 Page 31 of 32
YOSHIMITSU et al.
PRE-TRANSPLAT-COMORBIDITIES AND HCT IN ATL
1
32 Page 32 of 32