Conditioning Regimens Intravenous Busulfan for Allogeneic Hematopoietic Stem Cell Transplantation in Infants: Clinical and Pharmacokinetic Results

Home , Busulfan

Conditioning Regimens Intravenous busulfan for allogeneic hematopoietic stem cell transplantation in infants: clinical and pharmacokinetic results

JH Dalle1, D Wall2, Y Theoret1, M Duval1, L Shaw3, D. Larocque1, C Taylor2, J Gardiner3, MF Vachon1 and MA Champagne1

1Service d’He´matologie et Oncologie Pe´diatrique, Hoˆpital Sainte Justine, Montre´al QC, Canada; 2Methodist Children’s Hospital of South Texas, San Antonio, TX,USA; and 3Department of Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, Philadelphia, PA, USA

Summary: tion, very wide inter- and intrapatient systemic exposure is observed with two- to sixfold of coefficient variability. High-dose busulfan is an important component of This wide bioavailability range may be linked to erratic myeloablative regimens. Variable drug exposure may intestinal absorption (7 emesis), variable hepatic metabo- occur following oral administration. Therefore, the use lism, circadianrhythm, geneticpolymorphism of a-glu- of intravenous busulfan has been advocated. Previous tathione-S-transferase, initial diagnosis, previous work has suggested a cumulative dosage of 16 mg/kg for treatment, drug–drug interaction and/or patient age. haematopoietic transplantation in children less than 3 Hepatic and renal clearance mechanisms are generally years of age, but only limited data are available in infants. underdeveloped and inefficient in the neonate, but they Pharmacokinetics of intravanous busulfan administered at may change dramatically in the months following birth. the suggested dosage were studied in 14 infants (median Thus, pharmacokinetically guided dosage adjustment age 4.7 months). Busulfan plasma concentrations were appears mandatory, particularly in children.5,7–12 measured by either GC-MS or HPLC-UV. In seven Recently, different intravenous busulfan formulations patients, the dose was decreased to target an area- under- have been developed in order to minimize variations of the- curve of 600–1300 lmol min. The median total dose inter- and intrapatient systemic exposure and to provide given was 13.8 mg/kg. All patients engrafted. Severe veno- complete dose assurance.13,14 Inadults, the recommended occlusive disease occurred in one patient. Our study dosage was 0.8 mg/kg/dose for 16 consecutive doses.13,15,16 demonstrates that a cumulative dosage of 16 mg/kg is Wall et al17 suggested a dosage of 1.0 mg/kg/dose every 6 h associated with higher exposure than expected in infants. for 16 doses for childrenless than4 years of age, whereas We suggest an initial dose of 0.8 mg/kg followed by Cremers et al18 recommended 0.8 mg/kg/dose. However, pharmacokinetically guided dose adjustment. data in infants, that is, children under 1 year of age, are Bone Marrow Transplantation (2003) 32, 647–651. scarce, precluding a recommended dosage for that parti- doi:10.1038/sj.bmt.1704209 cular age group. Inthis retrospective analysis,we report Keywords: busulfan; infants; pharmacokinetics; allogeneic clinical and pharmacokinetic results of intravenous busulfan, prescribed off-label, in infants undergoing SCT at our institutions High-dose busulfan is widely used in conditioning regimens prior to haematopoietic stem cell transplantation (HSCT), as an alternative to total body irradiation. Although Study design busulfan may be administered once or twice daily,1,2 the most commondosage schedule is 1 mg/kg orally every 6 h Patients for 4 days to a total dosage of 16 mg/kg.3 Busulfanhas a narrow therapeutic index and veno-occlusive disease Between October 2000 and August 2002, 14 patients under (VOD) – one of the major dose-dependent toxicities – has 1 year of age and less than 10 kg received intravenous a mean incidence of 20–30% and an associated mortality busulfan as part of a conditioning regimen prior to rate from 3 to 67%.4,5 However, low drug exposure has allogeneic HSCT at the Methodist Children’s Hospital of been associated with higher recurrence rates and graft South Texas (San Antonio, TX; n ¼ 8) and at Sainte Justine failures.6,7 Following administration of the oral formula- Hospital (Montreal, Quebec; n ¼ 6). The underlying dis- eases were inherited syndromes in 10 cases, haematologic malignancies in three cases and amegakaryocytosis in one case. Grafts were one or two mismatched-unrelated cord Correspondence: Dr M Champagne, Service d’He´ matologie et Oncologie Pe´ diatrique, Hoˆ pital Sainte Justine, Coˆ te Sainte Catherine, Montreal blood (n ¼ 9), sibling bone marrow donor (n ¼ 3), matched- QC, Canada H3T 1C5. E-mail: [email protected] unrelated bone marrow donor (n ¼ 1) or haplotype Received 18 December 2002; accepted 17 March 2003 mismatch-related bone marrow donor (n ¼ 1) (Table 1). Intravenous busulfan for HSCT in infants JH Dalle et al 648 The median recipient age and body weight were 4.8 months steady state. The busulfaninfusionwascarried out over 2 h (range: 0.7–12) and 5.9 kg (range: 3.5–10), respectively via a central venous catheter using a controlled-rate (Table 2). infusion pump to mimic the time of maximum plasma concentration (about 1–2 h) observed following oral Treatment regimen administration. According to initial diagnosis and centre, different conditioning regimens were used: (1) busulfan– s An intravenous busulfan (Busulfex , OrphanMedical, cyclophosphamide association, where busulfan was given Minnetonka, MN, USA) dose of 1 mg/kg was administered ondays À9toÀ6 and cyclophosphamide 50 mg/kg once every 6 h for 16 doses in all patients but three (nos. 1, 3, 8), daily i.v. ondays À5toÀ2 (total dose 200 mg/kg) to seven inwhom a dose of 0.8 mg/kg was selected. Pharmacokine- patients (nos. 5, 9–14); (2) busulfan–fludarabine, with tically guided adjustment was usually performed , if needed, busulfan given on days –9 to –6 and fludarabine given at at dose 7, to target anarea underthe curve (AUC) of 600– 30 mg/m2 once daily i.v. on days –5 to À2 (total dose 900 mM min in children undergoing HLA-matched related 120 mg/m2) to three patients (nos. 1, 2 and 8); (3) busulfan– transplantation and 900–1300 mM mininthe others, in melphalan, with busulfan given on days –8 to –5 and melphalangivenat 45 mg/m 2 once daily i.v. on days –4 to –2 (total dose 135 mg/m2) to four patients (nos. 3, 4, 6 Table 1 Patients characteristics and 7). Intravenous busulfan was always administered as Pt no./ Age Body Weight Diagnosis Type of Seizure the first i.v. chemotherapy component of the preparative Gender (months) (Kg) Graft prophylaxis regimen; interaction with other chemotherapy agents that may interfere with PK characteristics had been avoided. All 1/F 5 5.3 SCID UCB Phenytoin patients received intravenous anticonvulsant prophylaxis as 2/M 3 6 SCID Sib Phenytoin 3/F 1 4.2 SCID UCB Phenytoin either midazolam (1.2 mg/kg/day) or phenytoin infusion 4/M 4 5.9 SCID UCB Phenytoin (dose adjusted to maintain therapeutic levels) from the day 5/M 8 6.9 Amega- UCB Phenytoin before to the day after the administration of busulfan. karyocytocis Ursodiol was administered as VOD prophylaxis. Hepatic 6/M 8 7.3 AML UCB Phenytoin VOD was defined and graded according to criteria 7/F 7 7.2 ALL UCB Phenytoin 19,20 8/M 4 6.2 SCID UCB Phenytoin described by McDonald et al. GVHD prophylaxis 9/F 0.7 3.5 Krabbe UCB Midazolam consisted of antithymocyte globulin (except for patient no. 10/M 9 9.8 AML MUD Midazolam 14), cyclosporinA andshort-course methotrexate or 11/F 4.5 4.5 SCID Haplo-Id Midazolam methylprednisolone. Patients were given standard suppor- 12/F 6 5.4 Omenn Sib Midazolam 13/M 2.9 3.8 Omenn UCB Midazolam tive care. Patients received HSCT after 1 day of rest at 14/F 12 10 FEL Sib Midazolam completionof the preparative regimen.

Median4.8 6 Pharmacokinetic study Blood samples were withdrawn from central venous lines F ¼ female; M ¼ male; UCB ¼ unrelated cord blood; Sib ¼ sibling donor; MUD ¼ match-unrelated donor; Haplo-Id ¼ haplo-identical donor; into heparinized tubes immediately before and after the FEL ¼ familial lympho-erythrophagocytic histiocytosis. administration of i.v. busulfan, and at 15, 30, 60, 120, 180

Table 2 Summary of the results of i.v. busulfanpharmacokinetics

Pt /Gender Age (months)/ Initial Bu dose Clearance ofBu AUC0-N Dose Adjusted AUCt Total received dose weight (kg) (mg/kg) (ml/kg/min) (mmol min) adjustment dose (mmol min) (mg/kg) after the after the 1st dose 9th dose

1/F 5/5.3 0.8 3.95 825 (1031)a No change NA ND 12.8 2/M 3/6 1 4.87 833 No change NA ND 16 3/F 1/4.2 0.8 2.98 1103 (1378)a Decrease 2.8 ND 11.52 4/M 4/5.9 1 3.09 1315 Decrease 4 ND 12.8 5/M 8/6.9 1 3.40 1203 No change NA ND 16 6/M 8/7.3 1 3.77 1077 No change NA ND 16 7/F 0.7/3.5 1 NA 1986 Decrease 2 ND 11.68 8/M 9/10.0 1 3.10 1369 Decrease 9 ND 16 9/F 4.5/4.5 1 3.30 1264 Decrease 3 768 12.8 10/F 6/5.4 1 2.80 1461 Decrease 4 806 13.3 11/M 2.9/3.8 1 3.10 1264 Decrease 3 950 13.6 12/F 12/10.0 1 3.72 1069 No change NA ND 16 13/F 7/7.2 1 4.57 889 Increase NA ND 18.25 14/M 4/6.2 0.8 4.12 794 (992) Increase NA ND 13.9 Median4.75/5.9 3.4 1234 13.75 Mean7s.d. 3.670.64 12247290 14.372.02

AUC ¼ area under concentration–time curve; ND ¼ not done; NA ¼ not available; s.d.: standard deviation; aCalculated values from normalized AUC for 1 mg/kg/dose.

Bone Marrow Transplantation Intravenous busulfan for HSCT in infants JH Dalle et al 649 and 240 min after the end of infusion, for all patients. When drug assures delivered dose precision. In addition, its use is possible, blood samples were also withdrawnbefore and more convenient in infants. Nevertheless, this variability after the ninth dose and at 15, 30, 120 and 240 min after the remains substantial,9 and does not obviate the need for end of infusion. Plasma busulfan values were determined by pharmacokinetically guided adjustment. GC-MS or HPLC-UV, as described elsewhere.21,22 Phar- There was no infusion-related toxicity. All patients macokinetic modelling was performed using WinNonlin engrafted as defined above. All surviving patients had Professional 3.1 software (Pharsight Corporation) and documented full donor chimerism; patients transplanted noncompartmental analysis was used to determine phar- for SCIDs, who had had a submyeloablative preparative macokinetic parameters such as the elimination rate regimen, had lymphoid donor chimerism. In all, 11 patients constant, clearance and AUC. From these, the individua- are alive with a medianfollow-up of 13.9 (0.2–28.2) lized maintenance dose that results in an AUC near the months. Kaplan–Meier 24 month-overall survival and day target was calculated. 100-treatment-related mortality were 78.6711 and 14.379.3%, respectively. This outcome compares favour- Definition of end points ably with international registry reports of paediatric SCT recipients. Patient no. 2 died at day þ 8 from sepsis and Engraftment: primary measures for haematopoietic pulmonary haemorrhage, and marrow engraftment was recovery were the time to achieve anabsolute neutrophil identified in a post-mortem marrow sample. Patient no. 8 count (ANC) of 0.5 Â 109/l for 3 consecutive days and time died at day þ 58 from multiple organ failure. Patient no. 13 to achieve a platelet count of 20 Â 109/l, independent of experienced intraventricular cerebral haemorrhage fol- platelet transfusion, for 3 consecutive days. In cases of lowed by a compressive peri-medullar haematoma at day death before haematopoietic recovery, engraftment docu- þ 60 whenhis platelet countwas more than50 Â 109/ml. mentation was performed by pathology sampling. There is no evidence that this neurological toxicity was Transplant-related mortality: deaths during the first 28 attributable to intravenous busulfan. The patient even- days after HSCT or deaths more than28 days after HSCT tually died at day þ 231 of an unrelated cause. Patient inremission. no. 9, who received a haplo-identical graft, developed Overall survival : time to death from any cause. severe VOD complicated by portal hypertension and oesophageal varices. None of the other patients experienced Statistical analysis any busulfan-related complications. In adults transplanted for CML, a recent study showed a Overall survival and treatment-related mortality censored significantly decreased risk of death for patients having a at 24 months and 100 days, respectively, were estimated per-dose AUC between950 and1500 mmol.min.27 The using the method of Kaplan and Meier.23.Analysis was initial dosage prescribed in our study was 1 mg/kg to target performed using GraphPad Prism 3.03 software (Graph- anAUC of 600–900 mM.min for children undergoing HLA- Pad, SanDiego, CA, USA). matched related transplantation and 900–1300 mM.minfor the others. According to pharmacokinetic results and targeted AUC, the dose was decreased inoneof the three Results and discussion patientsinthe former group, andinsix of 11 inthe latter. The dose was increased to more than 1 mg/kg in only one Using the AUC normalized for 1 mg/kg for patients nos. 1, patient. The median total dose of busulfan was 13.75 mg/kg 3 and 8 who had received 0.8 mg/kg, median and mean (range: 11.5–18.25). There were no failures of engraftment. (7s.d.) AUC of busulfanat steady state was 1234 mmol min Three patients who had a dosage modification had repeated (range: 992–1986) and 12247290 mmol.min(95% CI 1056– AUC determinations after the 9th dose of i.v. busulfan. The 1391), respectively (Table 2). Using the real AUC for these results were inthe targeted rangefor all the three (Table 2). three patients did not significantly modify the median AUC The meanpharmacokineticallyguided adjusted dose was (1153 mmol.min), and only slightly changed the mean: 0.86 mg/kg (range 0.57–1.25). These data support a dosage 11757317 mmol.min (95% CI 992–1358). One patient in of 0.8 mg/kg given as a 2-h intravenous infusion every 6 h 14 had anAUC greater than1500 mm.min, which can be for 16 doses as the initial dosage in infants. Even though no considered as the lower limit for life-threatening toxicity absolute data are available, according to Hassan et al,28 since in a study published by Dix et al 24 this was a phenytoin may be responsible for higher clearance, lower significant risk for VOD development when patients had an AUC and a shorter elimination half-life of busulfan than AUC 41500 mmol.min. In pharmacokinetic studies of oral benzodiazepines, presumably by induction of a-GST and busulfaninchildrenolder than4 years receiving1 mg/kg/ busulfan intestinal wall transport modifications. Here, both dose, published ranges of AUC are 402–1491 with a mean intravenous phenytoin and busulfan were used. Alteration of 9527285 mmol.min(ie 1652–6120 and3918 71170 ng h/ of busulfan intestinal wall transport cannot be incrimi- ml as reported by the authors).25 A similarly wide range nated, but the induction of a-GST canstill play a role in was reported from infants included in another study.26 The busulfan pharmacokinetics. In our cohort, only two of presented range of AUC and the standard deviation, eight patients given phenytoin, but five of six given although lower, still reflect wide pharmacokinetic varia- midazolam, needed busulfan reduction. However, the data bility. Our data suggest that in infants, in spite of are too limited to allow conclusion on to whether the interindividual busulfan hepatic metabolism variations as phenytoin–busulfan interaction was significant. No corre- seen with oral busulfan, the intravenous formulation of this lationwas observed betweenweight or age andAUC inthis

Bone Marrow Transplantation Intravenous busulfan for HSCT in infants JH Dalle et al 650 3000 AUC/age kinetic individualization of busulfan dosage regimens. Bone AUC/weight Marrow Transplant 2001; 28: 743–751. 6 Yeager AM et al. Optimizationof busulfandosage inchildren undergoing bone marrow transplantation: a pharmacokinetic 2000 r2<0.5 study of dose escalation. Blood 1992; 80: 2425–2428. 7 TranHT et al. Individualizing high-dose oral busulfan: M.min)

µ prospective dose adjustment in a pediatric population under- 1000 going allogeneic stem cell transplantation for advanced

AUC( hematologic malignancies. Bone Marrow Transplant 2000; 26: 463–470. 8 Sandstrom M et al. Populationpharmacokineticanalysis 0 resulting in a tool for dose individualization of busulphan in 012345678910111213 bone marrow transplantation recipients. Bone Marrow Trans- Age (months) or Weight (Kg) plant 2001; 28: 657–664. 9 Vassal G. Pharmacologically-guided dose adjustment of Figure 1 No correlationbetweenAUC andweight or age. busulfan in high-dose chemotherapy regimens: rationale and pitfalls (review). Anticancer Res 1994; 14: 2363–2370. 10 Bolinger AM et al. An evaluation of engraftment, toxicity and busulfan concentration in children receiving bone marrow complete cohort of infants (Figure 1). There appears to be a transplantation for leukemia or genetic disease. Bone Marrow trend for correlation between age and AUC in the female Transplant 2000; 25: 925–930. subgroup. 11 Dupuis LL, Najdova M, Saunders EF. Retrospective appraisal of busulfan dose adjustment in children. Bone Marrow Transplant 2000; 26: 1143–1147. Conclusion 12 Baker KS et al. Busulfan pharmacokinetics do not predict relapse inacute myeloid leukemia. Bone Marrow Transplant Our data suggest that intravenous busulfan is suitable as an 2000; 26: 607–614. 13 Andersson BS et al. Conditioning therapy with intravenous alternative to oral busulfan in conditioning before haema- busulfanandcyclophosphamide (IV BuCy2) for hematologic topoietic transplantation in infants, including high-risk malignancies prior to allogeneic stem cell transplantation: transplantation and obviates the difficulties of oral intake a phase II study. Biol Blood Marrow Transplant 2002; 8: in these patients. The use of an intravenous formulation of 145–154. busulfan was followed by reducing intra- and interindivi- 14 HassanZ et al. Pharmacokinetics of liposomal busulphan in dual variations. Pharmacokinetically guided dose adjust- man. Bone Marrow Transplant 2001; 27: 479–485. ment is still required, however, given the narrow 15 Andersson BS et al. Acute safety and pharmacokinetics of therapeutic range of busulfan. An initial dosage of intravenous busulfan when used with oral busulfan and 0.8 mg/kg every 6 h, with 16 planned doses is suggested cyclophosphamide as pretransplantation conditioning therapy: for future studies in infants. a phase I study. Biol Blood Marrow Transplant 2000; 6: 548–554. 16 Olavarria E et al. A phase I/II study of multiple-dose intravenous busulfan as myeloablation prior to stem cell transplantation. Leukemia 2000; 14: 1954–1959. Acknowledgements 17 Wall D et al. Phase II trial of intravenous busulfan (Busulfex) with cyclophosphamide inpediatric allogeneichematopoietic JH Dalle is a fellow of La Fondation Charles Bruneau. cell transplantation (HSC): pharmacokinetics, toxicity and efficacy (A Pediatric Blood and Marrow Transplant Con- sortium study). in: 42nd Annual Meeting of the American References Society of Hematology Vol 97 edn. Hematology, A.S.o. (Blood, SanFrancisco,CA 2000. 1 Fernandez HF et al. Evaluationof safety andpharmacoki- 18 Cremers S et al. Pharmacokinetics of intravenous busulfan in netics of administering intravenous busulfan in a twice-daily or children prior to stem cell transplantation. Br J Clin Pharmacol daily schedule to patients with advanced hematologic malig- 2002; 53: 386–389. nant disease undergoing stem cell transplantation. Biol Blood 19 McDonald GB et al. Veno-occlusive disease of the liver and Marrow Transplant 2002; 8: 486–492. multiorgan failure after bone marrow transplantation: a cohort 2 Russell JA et al. Once-daily intravenous busulfan given with study of 355 patients. Ann Intern Med 1993; 118: 255–267. fludarabine as conditioning for allogeneic stem cell transplan- 20 McDonald GB, Sharma P, Matthews DE, Shulman HM, tation: study of pharmacokinetics and early clinical outcomes. Thomas ED. The clinical course of 53 patients with Biol Blood Marrow Transplant 2002; 8: 468–476. venocclusive disease of the liver after marrow transplantation. 3 Santos GW et al. Marrow transplantation for acute non- Transplantation 1985; 39: 603–608. lymphocytic leukemia after treatment with busulfan and 21 Lai WK et al. Routine analysis of plasma busulfan by gas cyclophosphamide. N Engl J Med 1983; 309: 1347–1353. chromatography-mass fragmentography. Clin Chem 1998; 44: 4 HartmanAR, Williams S, DillonJJ. Survival, disease-free 2506–2510. survival and adverse effects of conditioning for allogeneic bone 22 Quernin MH et al. Quantification of busulfan in plasma by gas marrow transplantation with busulfan/cyclophosphamide vs chromatography-mass spectrometry following derivatization total body irradiation: a meta-analysis. Bone Marrow Trans- with tetrafluorothiophenol. J Chromatogr B Biomed Sci Appl plant 1998; 22: 439–443. 1998; 709: 47–56. 5 Bleyzac N et al. Improved clinical outcome of paediatric bone 23 Kaplan EL, Meir P. Nonparametric estimation from incom- marrow recipients using a test dose and Bayesian pharmaco- plete observations. J Am Stat Assoc 1958; 53: 457–481.

Bone Marrow Transplantation Intravenous busulfan for HSCT in infants JH Dalle et al 651 24 Dix SP et al. Associationof busulfanarea underthe curve with 27 Andersson BS et al. Busulfansystemic exposure relative to veno-occlusive disease following BMT. Bone Marrow Trans- regimen-related toxicity and acute graft-versus-host disease: plant 1996; 17: 225–230. deﬁning a therapeutic window for i.v. BuCy2 in chronic 25 Vassal G, Gouyette A, Hartmann O, Pico JL, Lemerle J. myelogenous leukemia. Biol Blood Marrow Transplant 2002; 8: Pharmacokinetics of high-dose busulfan in children. Cancer 477–485. Chemother Pharmacol 1989; 24: 386–390. 28 HassanM, Oberg G, Bjorkholm M, WallinI, LindgrenM. 26 Vassal G et al. Busulfandispositionbelow the age of three: Inﬂuence of prophylactic anticonvulsant therapy on high-dose alterationinchildrenwith lysosomal storage disease. Blood busulphankinetics. Cancer Chemother Pharmacol 1993; 33: 1993; 82: 1030–1034. 181–186.

Bone Marrow Transplantation