Phase II study of dose‐modified busulfan by real‐time targeting in allogeneic hematopoietic stem cell transplantation for myeloid malignancy

Abstract

We aimed to evaluate the efficacy and safety of allogeneic hematopoietic stem cell transplantation with targeted oral busulfan (BU) and cyclophosphamide (CY) in a phase II study. Busulfan (1.0 mg/kg) was given initially in six doses. Based on the estimated concentration at steady state after the first dose of BU, subsequent (7th–16th) doses were adjusted to obtain a targeted overall concentration at steady state of 700–900 ng/mL. The primary endpoint was 1‐year overall survival (OS). Fifty patients were registered and 46 (median age, 53 years; range, 18–62 years) received planned transplant, including 24 with AML, 16 with myelodysplastic syndrome, and six with CML. Fourteen patients were categorized as standard risk. Nineteen patients received transplant from human leukocyte antigen‐identical siblings, 27 from unrelated donors. The BU dose required reduction in 32 patients and escalation in six patients. One‐year OS was 65% (95% confidence interval, 50–77%). Cumulative incidence of hepatic sinusoidal obstruction syndrome was 11%. One‐year transplant‐related mortality was 18%. Both OS and transplant‐related mortality were favorable in this study, including patients of older age and with high risk diseases. Individual dose adjustment based on BU pharmacokinetics was feasible and effective in the current phase II study. This trial is registered in the University Hospital Medical Information Network Clinical Trial Registry System (UMIN‐CTR, ID:C000000156).

Busulfan is an alkylating agent widely used in high‐dose chemotherapy regimens for HSCT.1, 2 The BU level in serum has been shown to be an important factor for graft rejection and regimen‐related toxicity such as SOS.3, 4, 5 Unfavorable profiles of oral BU include delayed and variable absorptive characteristics and high variability in drug metabolism.6 Individualized dose adjustment of BU using the LSM, and its transplantation results, have been investigated widely in Caucasian patients and pediatric populations, but few prospective studies have investigated results in Asian patients.7 Prior to the current study, we carried out a prospective PK study to analyze BU concentration using gas chromatography–mass spectrometry.8 Nine patients were enrolled in the study, and received preparative regimen containing oral BU 1 mg/kg every 6 h for eight or 16 doses. Out of nine patients, only three met the average steady‐state plasma concentration levels in the safety range of 650–1000 ng/mL4, 9 after the first and 13th dose. From the results, we developed LSM to estimate the AUC using two different formulas in order to fit even delayed clearance. Subsequently, we carried out a pilot study that used the same targeting method as the current study, and six patients with myeloid malignancy received tBU+CY conditioning with a targeting AUC of Css 700–900 ng/mL. Four patients received dose reduction after the seventh dose of BU, and overall Css of three patients met the safety range of 786–905 ng/mL (Akio Kohno, Mariko Fukumoto, Hiroto Narimatsu, Kazutaka Ozeki, Masashi Sawa, Shuichi Mizuta, Hitoshi Suzuki, Isamu Sugiura, Seitaro Terakura, Kazuko Kudo, and Yoshihisa Morishita, unpublished data, 2003).

From these results, we carried out a prospective phase II trial in Japanese patients with myeloid malignancies to evaluate the clinical results of allogeneic HSCT undergoing individualized high‐dose oral BU+CY conditioning.

Materials and Methods

Eligibility criteria

Patients from 16 to 65 years old were eligible if they had a diagnosis of AML, CML, or MDS, with an Eastern Cooperative Oncology Group performance status of 0–2, and no previous history of HSCT. Standard risk was defined as AML in first complete remission, MDS in refractory anemia or refractory anemia with ringed sideroblasts, and CML in chronic phase. High risk was defined as the remaining disease type. Patients receiving T cell depletion, or those with clinically significant infection or severe abnormalities of cardiac, pulmonary, and hepatic functions were excluded. Included patient/donor pairs were either related HLA matched by serological typing of A, B, and DR locus, unrelated HLA matched, or HLA DRB1 one locus mismatched by genotypical typing of A, B, and DRB1 locus. Unrelated donors were chosen by coordination with the Japan Marrow Donor Program. Written informed consent was obtained from each patient according to the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of each center.

Conditioning regimen, GVHD prophylaxis, and supportive care

Patients received a conditioning regimen consisting of BU 1.0 mg/kg given orally four times a day for six doses on two consecutive days (dose 1–6). Six hours after dose 6, patients received an adjusted dose of BU four times a day for 10 doses (dose 7–16) on three consecutive days (Fig. S1). Cyclophosphamide 60 mg/kg was given i.v. on two successive days. Both BU and CY were dosed based on actual body weight if it was <120% of ideal body weight, and adjusted body weight for those exceeding 120%. Sodium valproate was given as seizure prophylaxis before and during BU treatment. Fluconazole was used as fungal prophylaxis.

Either cyclosporine or tacrolimus in combination with methotrexate was used for GVHD prophylaxis. Cyclosporine was given i.v. at a dose of 3 mg/kg per day in two divided doses starting on day −1. Tacrolimus was given i.v. at a dose of 0.025 mg/kg continuously starting on day −1. Methotrexate was given at a dose of 10 mg/m² on day 1 and 7 mg/m² on days 3 and 6. Oral cyclosporine or tacrolimus was substituted for i.v. administration when tolerated. In the absence of GVHD, the cyclosporine and tacrolimus doses were tapered after day 50. Acute GVHD of grade 2 or more was treated with methylprednisolone 1–2 mg/kg. Chronic GVHD was treated by the protocols of each institute.

Supportive care measures were used according to institutional guidelines. Daily granulocyte colony stimulating factor was started on day 6 and continued until absolute neutrophil count exceeded 500/μL for two consecutive days.

Pharmacokinetic studies of BU

For PK studies of BU, blood samples were obtained 0, 30, 60, 120, 300, and 360 min after the first oral dose. Frozen plasma samples were sent to the laboratory at Kitasato University, and plasma BU concentrations were assayed by gas chromatography–mass spectrometry.8 The AUC was calculated by LSM using the formulas shown in Table 1.

Table 1.

Formulas for limited sample model (LSM) in patients receiving allogeneic hematopoietic stem cell transplantation treated with targeted oral busulfan and cyclophosphamide

i	In cases C 6/C 2 = or <0.5 AUCLSM = 0.5C 0.5 + 0.75C 1 + 2.5C 2 + 2.0C 6 + 4C 6/(LnC 2 − LnC 6)
ii	In cases C 6/C 2 > 0.5 AUCLSM = 0.5C 0.5 + 0.75C 1 + 2.5C 2 + 2.0C 6 + 2C 6/(LnC 2 − LnC 6)

In the previous pilot study, formula (i) bore a strong approximation to actual area under the blood concentration time curve (AUC), but not in patients with an elongated absorption or a delayed elimination of busulfan. The formula of the LSM was modified in the case of C₆/C₂ > 0.5 and formula (ii) was used for those patients. C _x, serum busulfan level obtained at x hours after the first dose.

Average Css levels of BU were determined by the ratio of the BU AUC_LSM over the dosing interval to the time between doses. The BU dose after the sixth dose was adjusted when Css after the first dose was not within 700–900 ng/mL. Dose adjustment was not carried out for patients whose Css after the first dose was 700–900 ng/mL. A targeted dose was calculated to achieve an average Css after all doses of 800 ng/mL. The optimal dose of BU was calculated as follows: optimal single dose of BU (mg/kg) = 800 (ng/mL) × first dose (mg/kg)/Css of first dose (ng/mL).

The dose of the 7th to 16th BU was calculated as follows: revised dose (mg/kg) = [optimal single dose (mg/kg) × 16 (times) − first dose (mg/kg) × 6 (times)]/10 (times).

Definitions of outcomes

The study was designed as a phase II prospective trial. The primary endpoint of the study was 1‐year OS after transplantation. The secondary endpoint was DFS, PK of BU, aGVHD, and cGVHD, and the frequency and severity of SOS, regimen‐related toxicity up to day 28, mortality at day 100, hematological recovery, and DFS and OS of each disease category.

All patients were prospectively monitored for engraftment,10 post‐transplant toxicities, GVHD, hepatic SOS, and infection. Failure to reach an absolute neutrophil count of 0.5 × 10⁹ cells/L by day 28 after transplantation was defined as graft failure, and the patient was withdrawn from the study. The aGVHD was evaluated daily until day 28 and weekly from day 29 to 100 and graded by established criteria.11 The cGVHD was evaluated up to day 365. Treatment and the outcome of aGVHD and cGVHD were also evaluated. Sinusoidal obstruction syndrome was clinically evaluated before day 28, and diagnosed,12, 13, 14 then graded clinically12 according to the published criteria. Liver toxicity that occurred after day 21 and fulfilled the above criteria of SOS was defined as late‐onset SOS. Clinical data after day 29 until day 100 was additionally surveyed to evaluate late‐onset SOS retrospectively.

Disease monitoring was carried out by bone marrow aspiration within 1 week before or after days 30, 60, and 90 after transplantation. Relapse was defined by hematological recurrence for AML,15, 16 and by hematological or cytogenetic relapse for CML. Deaths in the absence of persistent relapse were categorized as non‐relapse mortality. Additional surveillance was carried out and the onset of SOS and regimen‐related toxicities from days 29 to 100 were collected retrospectively. Long‐term survival data and data of relapse after day 365 were also collected retrospectively.

Statistical analysis

The primary endpoint of the study was 1‐year OS after transplantation. The expected 1‐year OS was estimated to be 60%, and its threshold was estimated to be 40%. With a statistical power of 90% and a one‐sided, type I error of 5%, the number of eligible patients required for this study was calculated to be 46 using a binominal analysis method. The projected sample size was 50 patients, with the expectation that 10% of patients would be deemed ineligible.

Disease‐free survival was calculated from the date of transplantation until the date of relapse or the date of death in complete remission. This trial has been registered in the University Hospital Medical Information Network Clinical Trial Registry System (UMIN‐CTR, ID:C000000156). Data were analyzed with Stata 9.2 statistical software (Stata, College Station, TX, USA).

Results

Patient characteristics

Patients were registered from October 2003 through March 2007. Fifty patients were registered. One patient who developed severe hemorrhagic ulcer of the ileum after registration was considered to be ineligible. One patient developed metastatic breast cancer before receiving the conditioning regimen and was withdrawn. Forty‐eight patients received tBU+CY conditioning. One patient developed systemic convulsion on day −6 before transplantation, and the study was discontinued. Another patient received cord blood transplantation due to unexpected emergent unavailability of the unrelated bone marrow and was included only in the PK analysis. The remaining 46 patients who completed tBU+CY conditioning and received the planned transplantation were analyzed in the subsequent outcome study. Characteristics and a transplantation summary of these 46 patients at the time of registration are shown in Tables 2 and 3, respectively.

Table 2.

Characteristics of patients receiving allogeneic hematopoietic stem cell transplantation (n = 46)

Characteristics
Median age of patients (range), years	53 (18–62)
Sex of recipient (%)
Male	29 (63)
Female	17 (37)
Sex, donor versus recipient (%)
Match	25 (54)
Male to female	11 (24)
Female to male	10 (22)
Disease type (%)
AML	24 (52)
1st CR	5
2nd CR	10
1st relapse	5
No treatment†	4
MDS	16 (35)
RA	4
RAEB	9
CMML	1
RAEB‐t	2
CML	6 (13)
CP	5
AP	1
Disease risk‡ (%)
Standard	14 (30)
High	32 (70)
Performance status§ (%)
0	40 (86)
1	6 (13)
2	0 (0)
Donor (%)
Related	19 (41)
Unrelated	27 (59)
HLA (%)
HLA identical sibling	19 (41)
HLA 6/6 matched, unrelated	23 (50)
HLA mismatched, unrelated	4 (9)

^†Two patients with overt leukemia from myelodysplastic syndrome (MDS) and another two patients with hypoplastic AML did not receive induction chemotherapy before transplantation. ^‡Standard risk was defined as AML in 1st complete remission (CR), MDS in refractory anemia (RA) or RA with ringed sideroblasts, and CML in chronic phase (CP). ^§According to Eastern Cooperative Oncology Group criteria. AP, accelerated phase; CMML, chronic myelomonocytic leukemia; HLA, human leukocyte antigen; RAEB, refractory anemia with excess of blasts; RAEB‐t, RAEB in transformation.

Table 3.

Summary of transplantation in patients with AML (n = 24), myelodysplastic syndrome (n = 16), or CML (n = 6)

Stem cell source
G‐PBMC	7
Bone marrow	39
GVHD prophylaxis†
sMTX+CyA	22
sMTX+FK	21
aGVHD, grade (%)
None	26 (56)
I	4 (9)
II	11 (24)
III	4 (9)
IV	1 (2)
cGVHD, type (%)
None	16 (43)
Lmt	9 (24)
Ext	12 (32)

^†One patient received short‐term methotrexate + tacrolimus prophylaxis and subsequently received short‐term methotrexate + cyclosporine. aGVHD, acute graft versus host disease; chronic GVHD, chronic graft versus host disease; GVHD, graft versus host disease.

Treatment‐related toxicity and hepatic veno‐occlusive disease

Forty‐five of 46 patients undergoing tBU+CY conditi‐oning (98%) experienced grade II or higher regimen‐related toxicity, and 38 of 48 patients (79%) experienced grade III or more toxicity within 28 days post‐transplantation (Table S1). Infection (70%), oral mucositis (52%), nausea and vomiting (30%), and diarrhea (30%) were frequent grade III or more adverse reactions. Severe neurological toxicity of grade III or more was observed in five patients (11%). One patient developed subarachnoid hemorrhage and died on day 1 after transplantation. Another patient developed tacrolimus encephalopathy on day 23 after transplantation. This patient died of acute bleeding from gastric ulcer on day 57. Another patient developed neurological toxicity during the course of septic shock and died on day 15. One patient who received dose reduction had delayed engraftment, but subsequently engrafted on day 31.

Among 46 patients undergoing planned transplantation, four patients experienced grade III or IV liver toxicity before day 28 (Table S1). Grade III or more long‐term liver toxicity between days 29 and 100 was observed in nine patients (Table S2). Three patients were reported to have SOS before day 20, and two were reported to have late‐onset SOS from days 21 to 100. Cumulative incidence of overall SOS was 11% (95% CI, 4–22%) at day 100 after transplantation (Fig. 1). Two patients had mild SOS on day 8 and 11 after transplantation, and both improved. One of these patients died of an unrelated cause (acute renal failure and infection). The third patient was reported to have moderate SOS on day 13. This patient died of an unrelated cause (septic shock) on day 15 after transplantation. Two patients developed severe SOS on days 42 and 44. These patients died of hepatic failure on day 64 and 81, respectively.

Figure 1.

Cumulative incidence of sinusoidal obstruction syndrome in patients receiving allogeneic hematopoietic stem cell transplantation treated with targeted oral busulfan and cyclophosphamide. The cumulative incidence of overall sinusoidal obstruction syndrome was 11% (95% confidence interval, 4–22%) at day 100 after transplantation.

Graft versus host disease

The cumulative incidence of grade II–IV and III/IV aGVHD at day 100 were 35% and 11%, respectively. The cumulative incidence of grades II–IV aGVHD in the recipients who underwent transplant from an HLA‐identical related donor or unrelated donor was 26% and 41%, respectively, and those of grades III/IV aGVHD was 11% and 11%, respectively. The cumulative incidence of cGVHD at 1 year after transplantation was 52%. Of the 21 patients who developed cGVHD, 12 had extensive disease and nine had limited disease.

Survival outcome

Twenty‐six patients were alive with a median follow‐up of 43 months (range, 11.9–65 months) after transplant. Overall survival was 65% (95% CI, 50–77%) at 1 year after transplantation, 66% (95% CI, 47–79%) for high risk and 64% (95% CI, 34–83%) for standard risk patients (Fig. 2a). Overall survival of AML was 71% (95% CI, 48–85%; n = 24) 1 year after transplantation, 50% (95% CI, 25–71%; n = 16) for MDS, and 83% (95% CI, 27–97%; n = 6) for CML patients. Two patients died before day 28 as described above. From days 28 to 100, seven patients died due to treatment‐related mortality (four patients), infection (two patients), and relapse (one patient). Of the four patients who died of TRM, two died from hepatic toxicity, one from gastrointestinal bleeding, and one from thrombotic microangiopathy.

Figure 2.

Overall survival and disease‐free survival curves according to disease risk in patients receiving allogeneic hematopoietic stem cell transplantation treated with targeted oral busulfan and cyclophosphamide. Overall survival (a) and disease‐free survival (b), each stratified according to disease risk. Data were analyzed with the Kaplan–Meier method.

Disease‐free survival was 57% (95% CI, 41–69%) 1 year after transplantation, 56% (95% CI, 38–71%) for high risk and 57% (95% CI, 28–78%) for standard risk patients (Fig. 2b). Disease‐free survival of AML was 58% (95% CI, 36–75%) at 1 year, 44% (95% CI, 20–66%) for MDS, and 83% (95% CI, 27–97%) for CML patients.

Relapse and TRM

Thirteen patients (28%) experienced disease recurrence. Cumulative incidence of relapse was 24% at 1 year after transplantation. Cumulative incidence of relapse was 22% among patients with high risk disease, and 14% among patients with standard risk disease (Fig. 3a). Cumulative incidence of TRM was 18% at 1 year after transplantation (Fig. 3b).

Figure 3.

Cumulative incidence of relapse and transplant‐related mortality in patients receiving allogeneic hematopoietic stem cell transplantation treated with targeted oral busulfan and cyclophosphamide. Cumulative incidence of relapse with (a) high risk disease (22% at 1 year), standard risk disease (14%), and (b) cumulative incidence of treatment‐related mortality (18%).

Pharmacokinetic studies and dose modification

Among the 47 patients who completed the 16 BU doses, Css of the first dose was 1090 ± 318 ng/mL (range, 593–1673). The mean AUC_inf estimated after the first dose of BU was 6760 μg∙h/L (range, 3656–13058 μg∙h/L). The mean values of oral clearance, distribution volume, and elimination half‐life were 0.159 L/h/kg (0.079–0.263 L/h/kg), 0.55 L/kg (0.178–0.989 L/kg), and 2.54 h (0.98–5.49 h), respectively. Six patients received dose escalation of BU, and 32 received dose reduction (Fig. 4a). Median decreasing dose of BU was 4.5 mg/kg (28% of 16 mg/kg). Mean actual dose of BU was 12.7 ± 3.7 mg/kg (range, 7.6–21.3 mg/kg).

Figure 4.

Pharmacokinetic studies of busulfin (BU) in patients receiving allogeneic hematopoietic stem cell transplantation. (a) Number of the patients reaching concentration at steady state (Css) with the first BU dose. Six patients received dose escalation, nine patients received no modification, and 32 patients received dose reduction. (b) Liver toxicity in the first 28 days, sinusoidal obstruction syndrome (SOS), and BU dose modification. Triangles (▴) indicate patients diagnosed with SOS before day 20. Two of these patients developed grade II liver toxicity, and another developed grade IV liver toxicity. Two of these patients with early onset SOS died of an unrelated cause. Squares (■) indicate two patients who developed grade I liver toxicity in the first 28 days and were later diagnosed with severe late‐onset SOS. (c) Liver toxicity from days 29 to 100, SOS and BU dose modification. Triangles (▴) and squares (■) indicate the same patients as (b). Two patients who had late‐onset SOS and died had received dose escalation.

One patient was excluded from the analysis due to systemic convulsions on day −6, as described above. The Css of the first dose was 683.1 ng/mL in this patient. Although dose escalation was carried out to receive 18.7 mg/kg, the conditioning regimen was not completed.

Busulfan targeting and transplant outcome

Overall survival was not different between patients who received dose reduction, no modification, or escalation of BU (68%, 67%, and 50% at 1 year, respectively). Significantly more grade III–IV toxicities from days 29 to 100 were observed in patients who received dose escalation (Fisher's exact test, P = 0.023) (Table S2). No difference in TRM was observed among these three groups.

All three patients who developed early‐onset SOS within 20 days after transplant had received dose reduction of BU. Two developed grade II liver toxicity, and another developed grade IV liver toxicity before day 28 (Fig. 4b). Two patients who had late‐onset SOS and died had received dose escalation (Fig. 4c).

Discussion

We carried out a phase II study of individualizing the oral BU and CY conditioning regimen for adult allogeneic HSCT for myeloid malignancies. In the current study, 1‐year OS (65%; 95% CI, 50–77%) clearly exceeded the threshold level of 40%.

Oral administration of BU had been associated with erratic gastrointestinal absorption and resulted in unpredictable systemic drug exposure.3, 4, 5, 17 Pharmacokinetic studies of BU and subsequent dose adjustment strategies for the BU and CY conditioning regimen have been reported, mainly among pediatric patients.18, 19, 20, 21, 22 Although no essential difference in PK analysis has been reported between data from Japan and North America,23 survival data and information on the benefit of the tBU+CY regimen for Asian adult populations are limited.7 In this phase II study to target the BU Css range of 700–900 ng/mL, 32 patients received BU dose reduction and the median dose of total BU was reduced. Nevertheless, no increase in relapse was observed and the incidence of TRM was comparable to the BU+CY regimen using the i.v. form.23 Notably, the incidence of SOS (11% at day 100) was relatively lower than in the previous report of an adult population receiving the CY+total body irradiation regimen24 or oral non‐targeted BU+CY.6 Severe SOS was not observed within 20 days after transplantation, and this targeting strategy may contribute to reduce the severity of early‐onset SOS. Our positive results could be a consequence of adjusting the BU dose, considering that 38 of 47 patients (81%) actually had not achieved optimal Css after the first dose. That is, the fixed dose of BU was not optimal in 81% of these Japanese patients.

In our previous study, SOS was not observed among patients whose Css range was within the target dose or when the BU dose was reduced (Akio Kohno, Mariko Fukumoto, Hiroto Narimatsu, Kazutaka Ozeki, Masashi Sawa, Shuichi Mizuta, Hitoshi Suzuki, Isamu Sugiura, Seitaro Terakura, Kazuko Kudo, and Yoshihisa Morishita, unpublished data, 2003). In the current study, three patients in the BU reduction group developed early‐onset SOS, although the estimated cumulative Css remained within the targeted range. Liver toxicity in these patients might also be related to increased exposure to toxic metabolites of CY.24 A dose‐escalation study using test dose PK also showed that patients who showed a high level of AUC in the first dose developed severe toxicity, including hepatic SOS.25

Two of the six patients who received dose escalation experienced late‐onset severe SOS. We may need to be cautious of possible late‐onset severe SOS after dose escalation of BU. However, the causal relationship between dose escalation of BU based on low initial Css and SOS needs to be further evaluated, because individual oral BU PK are influenced by many factors. Glutathione S‐transferase‐mediated conjugation with GSH is the main mechanism to detoxify BU. Accumulation of the active metabolite of CY through depletion of the cellular GSH pool may contribute hepatic toxicity.26 Hepatic GST activity and GST gene polymorphisms have been shown to be associated with BU clearance as well as transplant outcome. Polymorphism of GSTM1 is reported as a risk factor of SOS.27 The heterozygous variant of GSTA1 (GSTA1*A/*B), which is observed in 26% of the Japanese population, resulted in slower elimination of BU than the wild‐type.28 Analysis using the Japan Marrow Donor Program showed a higher risk of TRM among recipients with the GSTM1‐positive genotype, which was different from the Caucasian population.29 We are currently investigating gene polymorphisms reported to be related with the risk factors of transplantation, such as GST genes and the UDP glucosyltransferase gene family30 in a prospective trial.

Dose targeting possibly improves the OS by alleviating the variable absorptive characteristics among individuals. However, our results also suggest that dose modification might increase the chance of toxicity after day 28, especially in the case of dose escalation, although we should be careful of this interpretation. Dose reduction could generally lead to rejection of the graft. However, in this study, only one patient had 3 days' delay of engraftment in spite of the large number of patients in the study who received a dose reduction of BU. Busulfan in i.v. form has enabled us to accomplish narrow‐ranged dose adjustment.31 Careful validation of the clinical efficacy of PK‐based targeting using i.v. BU is warranted.

In conclusion, individual dose adjustment based on BU PK was feasible and effective in the current phase II study.

Disclosure Statement

The authors have no conflicts of interest.

Abbreviations

aGVHD

acute graft versus host disease

AUC

area under the blood concentration time curve

BU

busulfan

cGVHD

chronic graft versus host disease

CI

cumulative incidence

Css

concentration at steady state

CY

cyclophosphamide

DFS

disease‐free survival

GSH

glutathione

GVHD

graft versus host disease

HLA

human leukocyte antigen

HSCT

hematopoietic stem cell transplantation

LSM

limited sample model

MDS

myelodysplastic syndrome

OS

overall survival

PK

pharmacokinetic

SOS

sinusoidal obstruction syndrome

tBU+CY

targeting BU+CY

TRM

transplant‐related mortality

Supporting information

Fig. S1. Study scheme. 
 Table S1. Regimen‐related toxicity before day 28 in patients receiving allogeneic hematopoietic stem cell transplantation treated with targeted oral busulfan and cyclophosphamide. 
 Table S2. Regimen‐related toxicity (day 29–100) in patients receiving allogeneic hematopoietic stem cell transplantation treated with targeted oral busulfan and cyclophosphamide.

(Cancer Sci, 2012; 103: 1688–1694)