Bone Marrow Transplantation (2000) 25, Suppl. 2, S35–S38  2000 Macmillan Publishers Ltd All rights reserved 0268–3369/00 $15.00 www.nature.com/bmt Allogeneic stem cell transplantation (BMT) for AML and MDS following i.v. busulfan and cyclophosphamide (i.v. BuCy)

BS Andersson1, J Gajewski1, M Donato1, S Giralt1, V Gian2, J Wingard2, S Tarantolo3, H Fernandez4,WWHu5, K Blume5, A Kashyap6, SJ Forman6 and RE Champlin1

1UT MD Anderson Cancer Center, Houston TX; 2UF Shands Cancer Center, Gainesville, FL; 3U Nebraska, Omaha, NE; 4U Miami, Miami, FL; 5Stanford U Medical Center, Palo Alto, CA; and 6City of Hope Natl. Med. Center, Duarte, CA, USA

Summary: plant conditioning therapy prior to allogeneic BMT for patients with AML or MDS. Bone Marrow Transplan- Pretransplant conditioning therapy with i.v. BuCy fol- tation (2000) 25, Suppl. 2, S35–S38. lowed by allogeneic hematopoietic stem cell transplan- Keywords: acute myeloid leukemia; myelodysplastic tation (BMT) was investigated in a phase II trial in syndrome; bone marrow transplantation; stem cell trans- patients with acute myeloid leukemia (AML) or myelod- plantation; i.v. busulfan; conditioning therapy ysplastic syndrome (MDS). We gave i.v. Bu at a dose of mg/kg every 6 h ؋ 16 doses, followed by Cy 0.8 60 mg/kg daily for 2 days. Twenty-six AML patients (18 males/eight females) were treated, only eight of whom High-dose oral busulfan (Bu) in combination with cyclo- were in CR1. The rest were either refractory to induc- phosphamide (Cy) or other cytotoxic agents is often used tion chemotherapy (four patients) or in a more as an alternative to total body irradiation (TBI)-based ther- advanced stage of their disease (14 patients). In apy in pretransplant conditioning regimens, especially in addition, nine patients with MDS (1M/8F) were treated. previously irradiated patients. Bu-based treatment is felt to Their median age was 41 years (range 21–64). be convenient and efficacious, although the few available Engraftment to у500 neutrophils/␮l was reached at 14 randomized studies have failed to conclusively demonstrate 1–5 days (range 10–29 days) post BMT, and the median time an advantage of Bu-based vs TBI-based treatment. of neutropenia was only 11 days (range 4–28 days). The A major shortcoming of oral Bu in high-dose therapy most common regimen-related toxicity was grade 2–3 is its very erratic and unpredictable intestinal nausea. In the post-BMT period (including BMT day absorption/bioavailability. This is a problem, since the ؉30), two patients died, one each from pulmonary resulting plasma concentration pattern has been correlated hemorrhage secondary to CMV pneumonia and hepatic with both toxicity in patients with high/complete drug 6–9 veno-occlusive disease (VOD), for an early treatment- absorption, and graft rejection/recurrent leukemia in related mortality (TRM) of 5.7%. Three patients patients who absorb a suboptimal fraction of the delivered 10,11 developed VOD and two of them died. There was no dose. To overcome this drawback we developed an 12,13 direct regimen-related pulmonary or neurologic tox- intravenous Bu formulation. In a modified phase I dose- icity. Overall, the clinical side-effect spectrum was anal- finding study we determined that a parenteral Bu dose of ogous to what would be expected from a high-dose oral 0.8 mg/kg should give similar pharmacokinetic (PK) para- Bu-based regimen; there was no unique toxicity experi- meters to those obtained after an oral dose of 1.0 mg/kg enced with the used solvent system. The disease-free BW. We here report the use of this Bu formulation in com- survival in the high-risk subgroup (all patients not in bination with Cy in 35 patients undergoing allogeneic stem CR1) at 1 and 2 years post transplant was 44% and cell transplantation (BMT) for acute myeloid leukemia 31%, respectively. The 13 patients still alive in CR have (AML) or myelodysplastic syndromes (MDS). been followed for a median of 24 months (range 18–32). Pharmacokinetic analysis showed very good interdose reproducibility, and limited interpatient variability in Patients and methods area under the plasma concentration vs time curve, peak concentration, and clearance of Bu after this i.v. Thirty-five patients with AML or MDS were included in formulation. We conclude, that this new i.v. Bu formu- this study (not eligible for a treatment protocol of higher lation is well tolerated; it has an impressive safety pro- institutional priority). This included patients with AML past file, and we suggest that it should be considered as first remission (CR, 14 patients), or in first CR but were appropriate replacement for oral busulfan in pretrans- considered at high risk for recurrent disease (eight patients). It also included those who had failed to achieve complete remission after induction chemotherapy (four patients), and Correspondence: Dr BS Andersson, Dept of Blood and Marrow Transplan- finally nine MDS patients, five of whom had a history of tation, Box 65 UT MD Anderson Cancer Center, 1515 Holcombe Blvd, a previous malignancy. Patients were eligible for study Houston, TX, 77005 USA treatment if they were between the physiologic ages of 15 Pretransplant conditioning with i.v. busulfan and cyclophosphamide BS Andersson et al S36 and 55 years. They must have normal hepatic function tive statistics were calculated for pharmacokinetic para- р Ͻ ϫ (bilirubin 1.0 mg%, SGPT 3 upper reference limit) meters—AUC, T1/2 and CL. and renal function (creatinine р1.5 mg%), and have a car- diac LVEF у50%, and DLCO у50%, and a matched related allogeneic marrow or stem cell donor available. All Results patients volunteered a written informed consent in accord- ance with institutional guidelines. All patients engrafted at a median of 14 days (range 10– The i.v. Bu Busulfex (Busulfan) injection (Orphan Medi- 29 days), and the median time of neutropenia was 11 days cal, Minnetonka, MN, USA) was administered through a (range 4–28 days). However, if only the patients who central venous catheter at 0.8 mg/kg BW over 2 h every 6 h entered the study with a normal absolute neutrophil count for 16 doses. The i.v. Bu was diluted in 5% dextrose in (у1500 cells/␮l) are considered, we observed a delayed water (D5W) or normal saline (NS). This was followed by onset of neutropenia after completion of the chemotherapy. two daily doses of cyclophosphamide at 60 mg/kg BW,14 Thus, the patients would not be neutropenic until day 4–5 and BMT was performed after a day of rest. Supportive post BMT, effectively reducing the median time of chemo- care was provided per each site’s institutional guidelines. therapy-induced neutropenia to 8.5 days prior to This included prophylactic treatment with phenytoin during engraftment. Further, donor-derived hematopoiesis was Bu administration, hydration and Mesna during the cyclo- demonstrated in all 25 evaluable patients who had RFLP phosphamide phase of conditioning and immunosuppres- studies, or who had a sex-mismatched donor. The efficacy sive therapy with low-dose methotrexate and either tacrol- of i.v. BuCy with BMT as intensification/salvage therapy imus or cyclosporin. is demonstrated in Table 1, with 33 of the 35 patients There were 19 males and 16 females with a median age remaining in or achieving a CR when restaged 1 month of 41 years (range 21–64). Sixteen of the 35 patients were after BMT. Three of the eight CR1 patients remain in CR considered heavily pretreated because they had received a at 21, 24 and 24 months. The higher-risk patient population previous therapeutic XRT regimen, у3 chemotherapy regi- (18 AML patients not in CR1, and nine MDS patients), mens or had a previous BMT (three patients). Successful demonstrated an OS of 64% and a DFS of 44%, and 31% engraftment, overall survival (OS) and (long-term) disease- at 1 and 2 years, respectively (Figure 1). The 13 patients free survival (DFS) were the endpoints of the study. who remain in CR have now been followed for a median Engraftment was recorded as the first day that the absolute of 24 months (range 18–32 months). The most common neutrophil count (ANC) exceeded 0.5 ϫ 109/l. Failure to toxicity was grade 2–3 nausea, mostly during the cyclopho- engraft would be defined as a failure to reach an ANC sphamide phase of the treatment. Among the serious treat- у0.5 ϫ 109/l within 100 days after transplantation. Patients ment-related events we had two deaths in the first 30 days; who engrafted within the first 100 days after transplan- one patient died from respiratory failure after pulmonary tation, but who later developed an ANC Ͻ0.5 ϫ 109/l hemorrhage as a complication of CMV pneumonia, and one would be classified as having late graft failure, unless the patient developed lethal VOD (incidence 2/35, ie 5.7%). neutropenia was caused by recurrence of the underlying Between days BMT ϩ31 and ϩ100, five additional patients disease. Relapse was recorded as the day of detection. Sur- died: one patient from VOD; one developed interstitial vival was recorded as the day of death with the cause of pneumonitis with ARDS, and then died in a clinical picture death noted. Patients were monitored for engraftment, of progressive ARDS vs bronchiolitis obliterans (BOOP). relapse and adverse events from BMT day Ϫ7 to day ϩ28, This patient had previously received approximately 40 Gy or until discharge from the hospital, and thereafter a mini- mantle XRT for Hodgkin’s disease. Finally, one patient mum of twice weekly until day ϩ28. All patients were developed hemorrhagic complications after intensified monitored for clinical chemistry laboratory parameters immunosuppressive therapy for graft-versus-host disease, twice weekly. Bone marrow examination with cytogenetics and two patients died of leukemic progression. The overall were performed at 1 month after BMT or as clinically indi- mortality in the first 100 days post BMT was 7/35 = 20%. cated. Restriction fragment length polymorphism (RFLP) Aside from the patient who had pulmonary hemorrhage studies were performed to document engraftment. Plasma samples for pharmacokinetic analysis were drawn at doses 1, 9 and 13. Bu plasma concentrations were determined for Table 1 Response/outcome at 1 month each time point by a GC-MSD assay.15 Pharmacokinetic (PK) parameters were determined for Status at BMT CR PD ED Total each patient’s plasma Bu concentration profiles after doses 1, 9 and 13 using the non-compartmental technique CR1 8 0 0 8 Primary Refr. or ϾCR1 16 0 2 18 (WinNonlin Version 1.1). The peak Bu concentrations MDS 9 0 0 9 (Cmax) were observed values after doses 1, 9 and 13. The Total 33 0 2 35 area under the plasma concentration vs time curve (AUC) was calculated by the linear trapezoidal rule. Apparent Summary of outcome assessed 30 days after BMT. At the left is the plasma clearance (CL) was determined using the dose– patient’s status at the start of conditioning therapy, at right the outcome area relationship. assessed on day 30. CR1 = first complete remission; Primary refr. = refractory to primary Descriptive statistics (mean, median, standard deviation) induction chemotherapy; ϾCR1 = relapse after CR, or more advanced were calculated for pertinent efficacy parameters: remission; PD = progressive disease; ED = early death, defined as death engraftment, relapse, survival, and DFS. The same descrip- prior to or on day 30 after BMT.

Bone Marrow Transplantation Pretransplant conditioning with i.v. busulfan and cyclophosphamide BS Andersson et al S37

Figure 1 The probability of surviving after allogeneic transplantation following i.v. BuCy for high-risk patients with AML and MDS. OS, overall survival; DFS, disease-free survival. For details, see Patients and methods, and Results. after CMV pneumonia and the patient who had interstitial dictable intestinal Bu absorption yields highly variable pneumonitis ARDS vs BOOP, we did not encounter any inter- and intra-individual plasma concentrations. High pulmonary toxicity, and there was no serious neurological plasma Bu concentrations have been correlated with serious toxicity encountered. Hepatic veno-occlusive disease hepatic and neurologic toxicity,6–9 although not all investi- (VOD) was seen in three patients (8.6%): two of them died, gators have been able to confirm these observations.16 one had previously failed an allogeneic BMT after Ctx- Further, it has been reported that low plasma Bu levels are TBI. All regimen-related toxicity had been previously associated with an increased risk for both graft rejection described with the use of oral Bu. There was no new and and disease recurrence post BMT.10,11 To circumvent the unique toxicity identified that could be associated with the unpredictable intestinal drug absorption and individual vari- formulation vehicle of dimethylacetamide and PEG- ations in drug metabolism, individualized Bu dosing based 400.12,13 on first-dose Bu plasma PK has been advocated.6,17 Acute graft-versus-host disease was encountered in five Although intellectually very appealing, such an individ- patients, one of whom died on BMT day ϩ62 of ualized dosing approach does not take into account that hemorrhagic complications after intensified immunosup- many patients vomit after the first dose, and neither does pressive therapy. it consider that some patients have a delayed intestinal drug The plasma PK analysis at doses 1, 9 and 13 revealed absorption, resulting in Bu blood levels that are still an excellent interdose reproducibility of all parameters increasing at the end of the customary 6-h dosing inter- (Table 2). The mean AUC value after dose 1 was 1130 val.9,18 Therefore, we developed the present parenteral, Ϯ ␮ Ϯ Ϯ ␮ 353 mol-min s.d., and it was 1169 225 mol-min pharmaceutically acceptable formulation12,13 for use in pre- after dose 9, demonstrating excellent inter-dose reproduci- transplant conditioning therapy. Assessed by peripheral bility. blood counts all patients engrafted, and donor-derived hem- atopoiesis was confirmed in all 25 patients who had RFLP and/or cytogenetic data available at 1 or 3 months post Discussion BMT. Our study illustrates that the new formulation is well tolerated and is an effective conditioning therapy for AML Oral Bu is a popular alternative to TBI in high-dose con- and MDS. ditioning therapy prior to BMT. Unfortunately, the unpre- The toxicity profile of the new i.v. Bu was analogous to what could be expected with an oral Bu-based conditioning Table 2 Pharmacokinetic parameters regimen when used in a high-risk patient population (for example see Refs 4, 19). However, the low overall inci- Dose AUC, ␮mol- Cmax, ng/ml ± s.d. CL ml/min ± s.d. min ± s.d. dence of serious VOD and absence of serious CNS toxicity in our patients compares favorably with other published 1 1130 Ϯ 353 931 Ϯ 181 200 Ϯ 55 studies using oral Bu-based conditioning therapy.4,19 In 9 1169 Ϯ 224 1218 Ϯ 192 187 Ϯ 46 Ϯ addition, the absence of any toxicity that could be attributed 13 n.a. 1122 282 NA to the used solvent vehicle is encouraging and supports the use of the DMA/PEG400 formulation. The PK analysis AUC = area under plasma concentration vs time curve; Cmax = peak con- centration; CL = clearance; NA = not available. showed similar clearance of our i.v. Bu to previously pub- For details see Patients and methods. lished data obtained with the standard oral preparation (cf.

Bone Marrow Transplantation Pretransplant conditioning with i.v. busulfan and cyclophosphamide BS Andersson et al S38 Ref. 8). This indicates that the solvent vehicle does not comparing busulfan with total body irradiation as conditioning alter human Bu metabolism in the used regimen. The PK in allogeneic marrow transplant recipients with leukemia: a parameters’ high inter-dose reproducibility was illustrated report from the Nordic Marrow Transplantation Group. Blood 1994; 83: 2723–2730. by the mean values of AUC, Cmax, and clearance after doses 1, 9 and 13 (Table 2). The lack of solvent system toxicity, 5 Devergie A, Blaise D, Attal M et al. Allogeneic bone marrow transplantation for chronic myeloid leukemia in first chronic the reproducible PK profile, and the DFS of 44% and 31% phase: a randomized trial of busulfan-cytoxan vs cytoxan-total at 1 and 2 years post transplant, respectively, in this high- body irradiation as preparative regimen: a report from the risk patient subgroup suggests, that the new i.v. Bu formu- French Society of Bone Marrow Graft (SFGM). Blood 1995; lation is at least equivalent to the standard tablets. In 85: 2263–2268. addition, the i.v. formulation lacks the problems associated 6 Grochow LB, Jones JJ, Brundrett RB et al. Pharmacokinetics with unpredictable, erratic intestinal absorption and poss- of busulfan: correlation with veno-occlusive disease in ible hepatic first-pass effect experienced with the tablets. patients undergoing bone marrow transplantation. Cancer In summary, the good clinical tolerance and reproducible Chemother Pharmacol 1989; 25: 55–61. 2 PK data should encourage the use of i.v. Bu in prospective 7 Vassal G, Deroussent A, Challine D et al. Is 600 mg/m the investigations of how to optimize Bu dosing schedules to appropriate dosage of busulfan in children undergoing bone marrow transplantation? Blood 1992; 79: 2475–2479. further improve both treatment safety and long-term DFS. 8 Hassan M, Ljungman P, Bolme P et al. Busulfan bioavail- Our observations suggest that the new i.v. Bu should be ability. Blood 1994; 84: 2144–2150. further evaluated as replacement for oral Bu in high-dose 9 Dix SP, Wingard JR, Mullins RE et al. Association of busul- pretransplant conditioning therapy. fan area under the curve with veno-occlusive disease follow- ing BMT. Bone Marrow Transplant 1996; 17: 225–230. 10 Slattery JT, Sanders JE, Buckner CD et al. Graft-rejection and toxicity following bone marrow transplantation in relation to Acknowledgements busulfan pharmacokinetics. Bone Marrow Transplant 1995; 16: 31–32. This work was supported by the US Food and Dug Administration 11 Slattery JT, Clift RA, Buckner CD et al. Marrow transplan- through Grants FD-R-001650-02-01, FD-R-0111202, and the tation for chronic myeloid leukemia: the influence of plasma National Cancer Institute through Grants 3PO1-CA49639, and busulfan levels on the outcome of transplantation. Blood 1997; 3P30-CA16672. The authors also wish to acknowledge the excel- 89: 3055–3060. lent work of the medical, nursing and clinical pharmacy staff of 12 Andersson BS, Bhagwatwar H, Chow D. Parenteral busulfan the blood and marrow transplantation units of the Centers for treatment of malignant disease. US Patents No. 5,430,057 involved in this study. Finally, BS Andersson is a consultant to and No. 5,559,148. Orphan Medical, Inc. 13 Bhagwatwar H, Phadungpojna S, Chow DS-L, Andersson BS. Formulation and stability of busulfan for intravenous adminis- tration in high-dose chemotherapy. Cancer Chemother Pharm- acol 1996; 37: 401–408. References 14 Tutschka PJ, Copelan EA, Klein JP. Bone marrow transplan- tation for leukemia following a new busulfan and cyclophos- 1 Blaise D, Maraninchi D, Archimbaud E et al. Allogeneic bone phamide regimen. Blood 1987; 70: 1382–1388. marrow transplantation for acute myeloid leukemia in first 15 Vassal G, Re M, Gouyette A. Gas chromotography-mass spec- remission: a randomized trial of a busulfan-cytoxan vs trometry assay for busulfan in biological fluids using a deuter- cytoxan-total body irradiation as preparative regimen: a report ated internal standard. J Chromatogr 1988; 428: 357–361. from the Group d’Etudes de la Greffe Moelle Osseuse. Blood 16 Schuler U, Schroern A, Kuhnle A et al. Busulfan pharmacoki- 1992; 79: 2578–2582. netics in BMT patients, is drug monitoring warranted? Bone 2 Blume KG, Kopecky KJ, Hensley-Downey JP et al. A pro- Marrow Transplant 1994; 14: 759–765. spective randomized comparison of total body irradiation-eto- 17 Grochow LB. Busulfan disposition: the role of therapeutic poside vs busulfan-cyclophosphamide as preparative regimens monitoring in bone marrow transplantation induction regi- for bone marrow transplantation in patients with leukemia who mens. Semin Oncol 1993: 20: 18–25. were not in first remission: a Southwest Oncology Group 18 Tran HT, Madden T, Felix E et al. Evaluation of a preparative Study. Blood 1993; 81: 2187–2193. regimen containing adjusted-dose busulfan with stem cell sup- 3 Clift RA, Buckner CD, Thomas ED et al. Marrow transplan- port for the treatment of advanced hematologic malignancies tation for chronic myeloid leukemia: a randomized study com- in pediatric patients. Blood 1996; 88: 3767a. paring cyclophosphamide and total body irradiation with bus- 19 Przepiorka D, Anderlini P, Ippolitti C et al. Allogeneic blood ulfan and cyclophosphamide. Blood 1994; 84: 2036–2043. stem cell transplantation in advanced hematologic cancers. 4 Ringden O, Ruutu T, Remberger M et al. A randomized trial Bone Marrow Transplant 1997; 19: 455–460.

Bone Marrow Transplantation