Pharmacokinetics of intravenous busulfan and evaluation of ... - Nature

Bone Marrow Transplantation, (1998) 22, 241–244  1998 Stockton Press All rights reserved 0268–3369/98 $12.00 http://www.stockton-press.co.uk/bmt

Pharmacokinetics of intravenous busulfan and evaluation of the bioavailability of the oral formulation in conditioning for haematopoietic stem cell transplantation US Schuler1, M Ehrsam1, A Schneider1, H Schmidt1, J Deeg2 and G Ehninger1 1

University Hospital ‘Carl Gustav Carus’, Department of Medicine I, Dresden, Germany; and 2Fred Hutchinson Cancer Research Center, Transplantation Biology Program, Seattle, WA, USA

Summary: Busulfan (BU) is included in many conditioning protocols for haematopoietic stem cell transplantation (HSCT). Pharmacokinetic parameters in individual patients have been related to short-term toxicity and risk of relapse after HSCT. In a series of 11 patients receiving the usual 16 × 1 mg/kg schedule over 4 days, we investigated the pharmacokinetics of replacing one dose with an intravenous formulation (BU in DMSO) which we had previously investigated in dogs. A dose of 0.5–0.6 mg/kg was used. No acute side-effects of BU/DMSO infusions administered over 1 h were observed. Bioavailability of BU powder capsules was on average 70% (range, 44–94%). Interindividual variability of the resulting AUC after intravenous doses was still substantial. Further studies are under way to define the possible role of BU/DMSO infusions in conditioning before HSCT. Keywords: busulfan; dimethylsulphoxide; pharmacokinetics; bioavailability

Oral busulfan (BU) is used in many conditioning protocols before HSCT. The oral bioavailability and pharmacokinetics of BU have been notoriously variable. Furthermore, a correlation between plasma levels and toxicity, in particular veno-occlusive disease (VOD) of the liver has been reported,1 and a recent study found an inverse correlation between plasma levels and probability of relapse in patients with chronic myeloid leukaemia.2 Variations in levels both intrapatient and interpatient, are thought to depend to a large extent on differences in the bioavailability of the oral preparation of BU. An intravenous preparation of BU should overcome these problems at a dose corresponding to that estimated to be the bioavailability of the oral preparation. We previously reported pharmacokinetic data on a water soluble form of BU prepared in dimethylsulfoxide (DMSO) administered to dogs.3 Since then clinical pilot studies using small doses of BU administered i.v. have been reported. However, no study has been conducted using Correspondence: Dr U Schuler, Medizinische Klinik und Poliklinik I, Fetscherstrasse 74, D-01307 Dresden, Germany Received 22 December 1997; accepted 10 March 1998

doses of BU as required in the setting of HSCT. This, however, is necessary to determine the actual bioavailability of the drug given at those doses. For ethical reasons, a classical phase I dose escalation of BU/DMSO is not feasible in the setting of conditioning. Transplantation is always used with curative intent and considering the success rate with oral BU, low dose i.v. administration would not be acceptable. Therefore we chose a different strategy: we designed a protocol in which only one of the customary 16 oral doses of BU would be replaced by an i.v. preparation at a dose corresponding to that estimated to be the bioavailability of the oral preparation. Aims of the present study were to define actual BU bioavailability in the high-dose regimen and to identify possible toxicities that might be associated with single intravenous doses. Materials and methods For the usual 4-day schedule of 4 × 1 mg BU/kg/day in preparation for HSCT, we decided to replace one dose with our i.v. formulation. The starting dose was derived from our previous study,4 in which we measured the area under the curve (AUC) for BU concentration after five oral doses in 20 individual patients. Assuming that the highest AUC in each individual patient represented the maximal bioavailability, we concluded that the average bioavailability would be no higher than 50–80%. In fact, in animal experiments,3 values as low as 20% were observed. Thus, we chose an initial individual i.v. dose of 0.5 mg/kg, ie 50% of the prescribed oral dose. This dose was given as the second dose in the regimen of 16 doses (see Figure 1). Doses 3, 4 and 5 were given as regular oral doses of 1 mg/kg each. Dose 6, 24 h after the i.v. dose was given orally but only at 0.5 mg/kg, ie numerically corresponding to the i.v. dose to allow for pharmacokinetic determinations after such a dose and possibly correct for variables introduced by the modified dose. Blood samples were collected after doses 2, 5 and 6. HPLC analysis was used to determine plasma levels, which were used for possible dose modifications on the last day of treatment. Since bioavailability was unknown, the possibility of both under- and overexposure had to be considered. For example, in the event of unexpected low bioavailability the i.v. BU could have resulted in an excessive AUC and dose 16 would have been reduced or omitted.

Intravenous busulfan US Schuler et al

Based on the results in the first patients, the reduced doses 2 and 6 were slightly increased to 60% (0.6 mg/kg) to keep estimates of overall AUCs for the 4 days within the expected range of reported values. Intravenous BU/DMSO solution was prepared immediately before administration as described.3 Heparinized blood samples were obtained at 0, 30, 60 and 90 min, 2, 3, 4, 5 and 6 h after the second and sixth dose. In addition, samples were obtained after the fifth dose at 0, 1, 4, and after 6 h for a limited sampling estimate (LSE) of the AUC. Samples were immediately chilled and centrifuged and plasma stored at −20°C until analysis. HPLC analyses were carried out as described previously.5 AUCs were calculated using the trapezoid rule without extrapolation after the last data point. Half-lives are given as non-parametric estimates, and total clearance after intravenous doses was calculated as the dose divided by AUC.

AUCs, t. and clearance values are given in Table 1, bioavailability ranged from 40–94%, with a mean value of approximately 70%. In one patient the intravenous dose could not be analysed completely because of several missing values. In no patient were modifications of dose 16, or an additional dose 17, required. Estimates of overall AUCs suggested that total BU exposure would have been approximately 3% higher if oral BU had been used exclusively. As observed before, one oral AUC (0.5–0.6 mg/kg) had only limited predictive value for another (full) dose (linear correlation, r2 = 0.57). On the other hand the mean values of oral AUCs (3213 ng × h/ml after 0.5–0.6 mg/kg and 6150 ng × h/ml after 1 mg/kg) show a linear relation between AUC and dose for the group as a whole. In this small group, we could not demonstrate a reduced interpatient variability (expressed as ratio between (standard deviation of mean AUC)/(mean AUC)). Intrapatient variability of the i.v. formulation could not be determined, as each patient received only one intravenous dose.

Patients Eleven patients were included in the study (five AML, one CML, four multiple myeloma, one MDS). Ages ranged from 23–54 years with a median of 43, weight ranged from 58–85 kg with a median of 79 kg. All patients received BU as described, followed by 2 days of cyclophosphamide at a dose of 60 mg/kg. Benzodiazepines were used for seizure prophylaxis; no phenytoin was given. The study protocol was approved by the local ethics committee.

Results Intravenous BU/DMSO was tolerated without clinical toxicity. No seizures were observed. One patient (No. 3) developed veno-occlusive disease (VOD) of the liver but recovered. This patient’s AUC after the intravenous dose was close to the mean of the group; after oral doses 5 and 6, AUCs ranked first and third of the group, respectively.

Discussion We have shown in this study that the bioavailability of oral BU at a dose of 0.5–0.6 mg/kg is approximately 70% (⫾16%) although the range was broad (40–94%). These results are is in good agreement with data by Hassan et al,6 who found a bioavailability of 80% (⫾19%) in seven adults and of 68% (⫾31%) in nine children, using a similar formulation at a much lower dose (median 0.11 mg/kg in children, 0.027 mg/kg in adults). The clearance reported by this group is somewhat higher than the values observed in the present study: for patients over the age of 13 years clearance values (2.4 + 0.5 ml/min/kg) were within 1 s.d. of the mean of the present study. In the studies reporting chronopharmacological effects with BU,7,8 the cause of the observed effects is not entirely clear. If chronopharmacology affects absorption rather than, for example, distribution or elimination, differences in bioavailability might occur

Bioavailability samples taken Dose

p.o.

LSE

1 mg/kg i.v.

242

1

2

5

6 16 dose (dose modifications if necessary)

HPLC analysis

Days Figure 1 Graphic representation of the study design. Reduced doses of BU were given at the same time of the day to exclude effects of chronopharmacology.7,8


Table 1

243

Pharmacokinetic parameters in individual patients

Patient No.

1 2 3 4 5 6 7 8 9 10 11 Mean (±s.d.)

Dose 0.5–0.6 mg/kg

1 mg/kg LSE-AUC (ng × h/ml)

AUC i.v. (ng × h/ml)

AUC oral (ng × h/ml)

bioavailability (%)

t. i.v. (h)

total body clearance i.v. (ml/min/kg)

3195 3172 5116 3716 7359 NA 6735 6066 3932 4231 5608

2807 2144 4789 2001 3102 2703 4356 4520 3133 3192 2595

88 68 94 54 42 NA 65 75 80 75 46

1.8 2.6 3.7 2.2 2.1 NA 3.7 3.4 2.8 3.4 2.5

2.7 2.3 1.8 2.3 1.2 NA 1.5 1.7 2.5 2.2 1.8

5113 NA 7473 5793 7397 4159 7885 7923 5515 6585 3655

4913 (1414)

3213 (901)

69 (16)

2.6 (1.0)

1.8 (0.7)

6150 (1465)

NA = not available.

depending on the timing of the measurements. Due to the fact that the oral dose was given 1 day later at the same time, we may have reduced variability. Our intravenous starting dose of 50% was close to the measured bioavailability of 70%. No relevant underdosing occurred despite our dose reduction, for two reasons. Firstly because reducing two of 16 doses by 50% is only 1/16 (or 6.25%) of the total dose, secondly BU powder in capsules seems to produce marginally higher AUCs compared to the tablets previously used,4 as estimated by the LSE-AUCs. Thus, the design of our study proved safe, as there was no relevant alteration of drug-exposure in terms of AUC. Dose adjustments for patients with high AUCs have been proposed in an attempt to reduce VOD incidence and severity.1 As in our previous study, the weak intraindividual correlation between the different AUCs does not support attempts to use measured AUCs for dose modifications of later doses. A relation between VOD incidence and AUC has not been found in all series (eg Refs 2 and 4). Increased risk of relapse with low exposure was described by one group.2 When dose adjustments were made, AUCs were not necessarily in the expected range.9 Hepatotoxicity, as a marker of high exposure to an anticancer drug, may even be associated with a more favourable outcome in some settings.10 Therefore we do not regard individual dose adjustments justified outside clinical studies. With these low intravenous doses, we were not able to show a marked reduction of interpatient variability of BU-AUCs. Single daily doses, as proposed by Shaw et al11 for the oral route are now under investigation for our intravenous formulation and may help to reduce variability. Unfortunately, we were not able to prove reduced interpatient variability in this series. Only a series with larger and repeated intravenous dosing will allow the need and possibility of dose individualisation with intravenous doses to be estimated. The total amount of DMSO necessary for a 4-day treatment would be approximately 4 × 150 ␮l/kg, the total dose given on a single day, when large volumes of cryopreserved cells are given. Thus, toxicity of DMSO should be manageable. In conclusion, replacing one out of 16 doses of BU with

an intravenous BU/DMSO solution in a dose of 0.5–0.6 mg/kg was possible without unexpected toxicity. Bioavailability of BU capsules is approximately 70% with wide ranges of observed values. In an ongoing study, oral BU is gradually replaced by four single daily intravenous infusions. In this setting, reliable monitoring should be possible with comparatively few blood samples and without the restrictions of a 6-h dosing interval. This should help to define more clearly the relation between individual BU pharmacokinetic parameters, side-effects and toxicity.

Acknowledgements The research presented in this paper is supported by a grant of the Deutsche Krebshilfe (German Cancer Aid).

References 1 Grochow LB, Jones RJ, Brundrett RB et al. Pharmacokinetics of busulfan: correlation with veno-occlusive disease in patients undergoing bone marrow transplantation. Cancer Chemother Pharmacol 1989; 25: 55–61. 2 Slattery JT, Clift RA, Buckner CD et al. Marrow transplantation for chronic myeloid leukemia: the influence of plasma busulfan levels on the outcome of transplantation. Blood 1997; 89: 3055–3060. 3 Ehninger G, Schuler U, Renner U et al. Use of a water-soluble busulfan formulation – pharmacokinetic studies in a canine model. Blood 1995; 85: 3247–3249. 4 Schuler U, Schroer S, Ku¨hnle A et al. Busulfan pharmacokinetics in BMT patients, is drug monitoring warranted? Bone Marrow Transplant 1994; 14: 759–765. 5 Blanz J, Rosenfeld C, Proksch B et al. Quantitation of busulfan in plasma by high-performance liquid chromatography using postcolumn photolysis. J Chromatogr 1990; 532: 429– 437. 6 Hassan M, Ljungman P, Bolme P et al. Busulfan bioavailability. Blood 1994; 84: 2144–2150. 7 Hassan M, Oberg G, Bekassy AN et al. Pharmacokinetics of


244

high-dose busulphan in relation to age and chronopharmacology. Cancer Chemother Pharmacol 1991; 28: 130–134. 8 Vassal G, Challine D, Koscielny S et al. Chronopharmacology of high-dose busulfan in children. Cancer Res 1993; 53: 1534–1537. 9 Dix SP, Wingard JR, Mullins RE et al. Association of busulfan area under the curve with veno-occlusive disease following BMT. Bone Marrow Transplant 1996; 17: 225–230.

10 Schmiegelow K, Pulcynska M. Prognostic significance of hepatotoxicity during maintenance chemotherapy for childhood acute lymphoblastic leukemia. Br J Cancer 1990; 61: 767–772. 11 Shaw PJ, Scharping CE, Brian RJ et al. Busulfan pharmacokinetics using a single daily high-dose regimen in children with acute leukemia. Blood 1994; 84: 2357–2362.