Busulfan has a narrow therapeutic range, and in children, pharmacokinetic variability has been found to be high even after the use of intravenous (i.v.) busulfan. Recently, a reduced toxicity myeloablative regimen showed promising results, but the data of busulfan pharmacokinetics in hematopoietic stem cell transplantation (HSCT) using a targeted busulfan/fludarabine regimen in children has not yet been reported. We performed therapeutic drug monitoring (TDM) after once-daily i.v. busulfan combined with fludarabine and analyzed the outcomes. Busulfan (i.v.) was administered once daily for 4 consecutive days. The daily target area under the curve (AUC) was 18,125-20,000 μg*h/L/day (4415-4872 μmol*min/L/day), which was reduced to 18,000-19,000 μg*h/L/day (4384-4628 μmol*min/L/day) after a high incidence of toxicity was observed. A total of 24 patients were enrolled. After infusion of busulfan on the first day, patients showed AUC that ranged from 12,079 to 31,660 μg*h/L (2942 to 7712 μmol*min/L) (median 16,824 μg*h/L, percent coefficient of variation (%CV) = 26.5%), with clearance of 1.74-6.94 mL/min/kg (median 4.03 mL/min/kg). We performed daily TDM in 20 patients, and during the daily TDM, the actual AUC ranged from 73% to 146% of the target AUC, showing high intraindividual variability. The %CV of busulfan clearance of each individual ranged from 7.7% to 38.7%. The total dose of busulfan administered for 4 days ranged from 287.3 mg/m(2) to 689.3 mg/m(2). Graft failure occurred in 3 patients with total AUC less than 74,000 μg*h/L (18,026 μmol*min/L), and 2 patients with relatively high total AUC experienced veno-occlusive disease. Busulfan pharmacokinetics showed high inter- and intraindividual variability in HSCT using a targeted busulfan/fludarabine regimen, which indicates the need for intensive monitoring and dose adjustment to improve the outcome of HSCT. Currently, we are performing a newly designed phase II study to decrease regimen-related toxicities and reduce graft failure by setting an optimal target AUC based on this study.

3492 Introduction: Busulfan has narrow therapeutic range. High exposure is associated with systemic toxicity such as veno-occlusive disease (VOD) while underexposure results in graft failure or relapse. In children, pharmacokinetic variability was found to be high even after the introduction of intravenous busulfan. Busulfan is metabolized via liver through conjugation with glutathione S-transferase (GST) family, and inter-individual variability may be explained by GST polymorphisms. Thus, we investigated the influence of GST polymorphisms on the clinical outcomes of hematopoietic stem cell transplantation (HSCT) with busulfan based conditioning regimen in children. Patients and methods: We studied patients who underwent HSCT at Seoul National University Children's Hospital from November, 2001 to November, 2008. IV busulfan (0.8 mg/kg/dose for patients≥10 kg and 1.1 mg/kg/dose for patients <10 kg, q 6hr for 4 days) was used in combination with cyclophosphamide or fludarabine as conditioning regimen. Etoposide was added for acute lymphocytic leukemia (ALL). GST polymorphisms (GSTA1 375 A > G, -52 G > A, -567 G > T, -631 G > T, -69 C > T, GSTM1 deletion, GSTT1 deletion, and GSTP1 313 A > G) were analyzed by multiplex PCR amplification and SNP genotyping. Results: A total of 70 patients (48 male, 22 female) were analyzed. The diagnoses were ALL in 32, AML in 26, and other diseases in 12 patients. Median age at HSCT was 8.8 years (range 1.0–19.0 years). Bone marrow or peripheral blood stem cell transplantations (BMT/PBSCT) were conducted in 42 patients, and cord blood transplantations (CBT) in 28 patients. Graft failure occurred in 11 (15.7%) patients (1 (2.4%) in BMT/PBSCT, 10 (35.7%) in CBT) and relapse occurred in 15 (21.4%) patients (12 (28.6%) in BMT/PBSCT, 3 (10.7%) in CBT) after HSCT. Patients having GSTA1 *A/*A and GSTP1 313 A/A genotype (N=33) showed higher incidence of graft failure than the others (N=37) (27.3% vs 5.4%, P =.01). Conversely, the incidence of hyperbilirubinemia over grade 3 was significantly lower in the patients with GSTA1 *A/*A and GSTP1 313 A/A genotype than the other patients (6.1% vs 24.3%, P =.05). Event free survival (EFS) of patients with GSTA1 *A/*A and GSTP1 313 A/A genotype was significantly lower than the EFS of the other patients in both BMT/PBSCT and CBT, and the main causes of event were graft failure in CBT and relapse in BMT/PBSCT. Conclusions: In children undergoing HSCT with busulfan based conditioning regimen, GST A1 and P1 polymorphisms seem to have influence on the graft failure, relapse and complications. To confirm our results, further studies about the influence of GST A1 and P1 polymorphisms on the pharmacokinetics of busulfan are needed. Disclosures: No relevant conflicts of interest to declare.

1184 Poster Board I-206 Introduction: Busulfan shows narrow therapeutic range. High exposure is associated with systemic toxicity such as veno-occlusive disease (VOD) and underexposure is associated with graft failure or relapse. In children, pharmacokinetic variability was found to be high even after the use of intravenous (IV) busulfan. Recently, reduced toxicity myeloablative regimen showed promising results but the data of busulfan pharmacokinetics when combined with fludarabine in children was not reported yet. We performed therapeutic drug monitoring (TDM) after once-daily IV busulfan combined with fludarabine and analyzed the outcomes. Patients and Methods: Busulfan (once daily ivs for 4 consecutive days) and fludarabine were used for acute myeloid leukemia (AML) and other diseases. For acute lymphocytic leukemia (ALL) etoposide was added. Busulfan (120 mg/m2) was administered on the first day. Blood samplings were taken before administration and 1, 2, 4hr after the end of infusion. Area under the curve (AUC) was analyzed by high pressure liquid chromatography. Target AUC was 18,125-20,000 ug*h/L/day, and dose adjustment was done when AUC was beyond the range. We initially planned to perform TDM on the first, fourth day, and the day when dose modification was needed, but we changed the design to perform TDM daily after the observation of increased AUC with modified dose on the fourth day showing decreased clearance in two patients. After we observed high incidence of toxicities, the target AUC was reduced to 18,000-19,000 ug*h/L/day. Results: A total of 18 patients were enrolled for this study. The diagnoses were AML in 10, ALL in 4, and other diseases in 4 patients. Median age was 9.0 years (range 1.3-18.0 years). After infusion of 120mg/ m2 busulfan on the first day, patients showed AUC ranged from 12,723.0 to 31,659.9 ug*h/L (median 18,638.0 ug*h/L) with clearance of 1.74-6.07 mL/min/kg (median 3.79 mL/min/kg). In 8 patients, busulfan dose increased 1.07-1.53 times compared to the first dose on the second day, and a dose reduction with 0.59-0.95 times compared to the first dose was made in 6 patients. Two patients received same dose of busulfan after target AUC was achieved, but the patients showed increased AUC (27,752.8 and 26,060.1 ug*h/L, respectively) on the fourth day. After then, daily TDM was performed in 15 patients and among the 45 times TDM and dose modification, AUC within the target range was achieved in only 9 times after administration of new targeted modified dose. Higher AUC (104.3-144.6%) than expected was shown in 23 times with decrease of clearance (51.2-97.1% compared to prior day), and lower AUC (79.4-94.2%) with increased clearance (108.8-177.9% compared to prior day) was shown in 13 times. The clearances of each individual changed during the 4 days and further dose modifications were needed on the third and fourth day in 13 and 12 patients, respectively. The total dose of busulfan administered for 4 days ranged from 287.3 mg/m2 to 569.4 mg/m2 (median 467.0 mg/m2) with median total AUC of 77,913.0 ug*h/L (range 72,752.4-83,160.3 ug*h/L). Engraftment was achieved in all patients, and 1 patient developed VOD. Conclusions: Busulfan pharmacokinetics was highly variable and unpredictable in all patients and it also changed during 4 consecutive infusions in each patient when combined with fludarabine. Further studies are needed to determine the factors affecting the variability. Intensive monitoring with daily TDM and dose modification are necessary for optimal therapy in children undergoing hematopoietic stem cell transplantation with busulfan plus fludarabine based conditioning regimen. Disclosures: No relevant conflicts of interest to declare.

3109 Poster Board III-46 Acute myeloid leukemia (AML) is a class of prevalent hematopoietic malignancies, which often remains incurable because of the development of drug resistance. Leukemic cells originated from hypoxic condition in bone marrow that gives a benefit to leukemic cells by protecting them from anti-cancer drugs through physical barrier and by induction of some survival signal mediators. As the development of drug resistance is a key element in the failure of chemotherapy for leukemia, it has been studied about many factors make leukemic cells resist to chemotherapeutic drugs. However, effect of hypoxia on drug induced apoptosis is still poorly understood. Here, we investigated whether hypoxia affects the resistance to drugs in leukemic cells and whether hypoxia has some role in drug-induced apoptosis. To demonstrate whether hypoxia influences the apoptosis of human leukemic cells induced by chemotherapeutic drugs, the human AML cell lines (U937, HL-60, CCRF-CEM and K562) were treated with arsenic trioxide (ATO), actinomycin-D, 17-AAG, valproic acid and cytrabine under hypoxic condition and hypoxiamimetic agent cobalt chloride (CoCl2). Cellular proliferation was evaluated by methyl thiazolyl tetrazolium (MTT) assay. Subsequently, FACS analysis and western blot were performed to investigate the apoptosis related proteins. Among AML cell lines, HL-60 cells became resistant to apoptosis induced by chemotherapeutic drug, especially ATO, under the hypoxic condition. It was demonstrated through FACS analysis that level of Annexin-V staining in hypoxia (13.58%) was lower than in normoxia (29.07%). Also, among many apoptosis related molecules, activation of HIF-1 alpha under hypoxic condition was associated with the cell survival (HSP70) and apoptosis (BAX). Expression of HSP70 was increased and the level of BAX was dramatically down-regulated by HSP70 in ATO treated HL-60 cells. These data show that the hypoxia increases resistance to ATO induced apoptosis and the effect was mediated by HSP70/BAX dependent pathway. Collectively, this results provided several lines of direct evidence for the role of hypoxia on apoptosis of AML cells, in which HSP 70 and BAX elicit a role as an effector downstream to HIF-1alpha. These discoveries would shed new insights for understanding underlying mechanisms of hypoxia and designing new therapeutic strategy for AML. Disclosures No relevant conflicts of interest to declare.