Bone Marrow Transplantation (2003) 31, 11–15 & 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00 www.nature.com/bmt Non-myeloablative stem cell transplants Fludarabine in combination with cyclophosphamide decreases incidence of GVHD and maintains effective graft-versus-leukemia effect after allogeneic stem cell transplantation in murine lymphocytic leukemia

L Weiss1, A Abdul-Hai1,ROr1, G Amir2 and A Polliack3

1Department of Bone Marrow Transplantation and Cancer Immunobiology Research Laboratory, Jerusalem, Israel; 2Department of Pathology, Jerusalem, Israel; and 3Department of Hematology, Hadassah University Hospital and Hebrew University, Hadassah Medical School, Jerusalem, Israel

Summary: well as in acute leukemia.4–7 Most recently, it has been incorporated into nonmyeloablative cytoreduction regi- Graft-versus-host disease (GVHD) is a severe disorder mens and used as for immunosuppression in the novel and despite therapeutic efforts to decrease its distressing mini-allogeneic transplant procedure used in recent years clinical manifestations, treatment is still not optimal. Here for a variety of hematological disorders.8–10 In fact, we report the results of studies, in which the purine fludarabine is the anchor drug in these regimens. The analogue, fludarabine phosphate, was used in an attempt cytotoxic potential of fludarabine against T lymphocytes, to modify and decrease GVHD after stem cell transplanta- coupled with its immunosuppressive qualities, has made tion, across major histocompatibility barriers for murine it an attractive choice for both chemotherapy and im- leukemia. B-cell leukemia (BCL-1) bearing (BALB/ munomodulation, during allogeneic transplantation for c  C57BL/6) F1 mice received two cycles of fludarabine leukemia.11,12 Indeed, the use of fludarabine-containing (0.8 mg/kg) for 5 days every 2 weeks, followed by 400 mg/ regimens has altered the incidence, and the degree of graft- kg cyclophosphamide i.p. Animals were then transplanted versus-host disease (GVHD) in these patients and this with C57BL/6 precursor cells and the development of regimen is used currently in a number of transplant leukemia and extent of GVHD was monitored both centers.8–13 clinically and histopathologically. In the fludarabine- Despite the impressive results in human hematopoietic treated group, only nine of 28 (32%) mice developed stem cell transplantation, fludarabine has as yet not been leukemia, compared to 25 of 33 (76%) of control animals adequately tested in animal models. In this respect, there is (P ¼ 0.0006 ). Mice treated with fludarabine-containing some recent experimental evidence showing that it may be regimens prior to transplantation also had much less effective in transplantation across histocompatibility GVHD both clinically and at autopsy, while graft-versus- barriers in mice.14 Furthermore, it is also able to induce leukemia appeared to be augmented in the same animals. bilateral tolerance or stable chimerism after marrow or skin Bone Marrow Transplantation (2003) 31, 11–15. transplantation,15 thereby altering the incidence of graft doi:10.1038/sj.bmt.1703775 rejection and the extent of GVHD encountered. In earlier Keywords: fludarabine; cyclophosphamide; graft-versus- pilot murine studies, we were also able to show an anti- host disease incidence; graft-versus-leukemia; hematopoie- GVHD prophylactic effect after marrow transplantation in tic stem cell transplant; murine lymphocytic leukemia mice.16 In the present study, we report the results of a series of experiments using the combination of fludarabine and cyclophosphamide as cytoreduction during allogeneic stem Fludarabine used as a single agent or in combination with cell transplantation for murine leukemia. In these studies, other drugs has been extensively used in recent years for the we attempted to mimic human leukemias clinically and treatment of a variety of hemato-oncologic disorders, and determine whether fludarabine could decrease the intensity has been shown to be most effective in indolent lympho- and incidence of acute GVHD, while preserving or even proliferative disorders, particularly chronic lymphocytic enhancing the simultaneous graft-versus-leukemia (GVL) leukemia (CLL) and follicular lymphoma.1–3 Furthermore, effect in this experimental model. fludarabine-containing combination regimens have also been used in the treatment of aggressive lymphomas as Material and methods

Correspondence: Professor A Polliack, Head of Lymphoma–Leukemia Unit, Department of Hematology, Hadassah University Hospital, Ein All animal procedures utilized in the present study were Kerem, PO Box 12000, Jerusalem 91120, Israel approved by the Institutional Committee for Animal Received 29 May 2002; accepted 29 July 2002 Experimentation. Fludarabine and transplantation in murine leukemia L Weiss et al 12 MHC-BMT model mice. This protocol attempted to mimic the clinical situation of partially responsive disease in humans with Male and female mice 2–6 months old (BALB/c C57BL/ Â leukemia where therapy is followed by allogeneic stem cell 6) F1 (H-2d/b) were transplanted with C57BL/6 (H-2b), transplantation. bone marrow or spleen cells. The mice purchased from the F-1 mice received 105 BCL-1 leukemia cells and 1 day Harlan Breeding Facility (Jerusalem, Israel) were kept in a later 0.8 mg/kg fludarabine was given for 5consecutive standard animal facility with top-filtered cages. Cages, days. After 2 weeks, the mice received another 5days cycle sawdust and water bottles were autoclaved once a week. of 0.8 mg/kg fludarabine, followed by 400 mg/kg i.p. Neomycin sulfate, at a concentration of 0.5%, was given in cyclophosphamide 2 weeks later. A day after the above the drinking water for 2 weeks post-transplantation. treatment with cyclophosphamide, animals were trans- Fludarabine monophosphate (Schering, Berlin, Germany) planted with 10 Â 106 C57BL/6 bone marrow cells and and cyclophosphamide (ASTA medica Ig, Frankfurt/Main, 2 Â 106 spleen cells. Control animals received saline instead Germany) were given intraperitoneally. of fludarabine. The experiments performed with this schedule were repeated on four separate occasions. Stem cell transplantation Experiment 3: This experiment tried to evaluate the effect of fludarabine and allogeneic transplantation com- Bone marrow cells were prepared by flushing RPMI pared to syngeneic transplantation in leukemic mice. F-1 medium through the femora and humeri of donors with a mice were treated with 0.8 mg/kg fludarabine or saline twice 9–25-gauge needle. Spleens were removed aseptically from a week for 2 weeks. The mice were injected with 400 mg/kg donor mice, teased through a nylon mesh into RPMI 1640 cyclo i.p. 10 days later. The mice were transplanted with medium (GIBCO, Grand Island Biological Co., Grand 20 Â 106 C57BL spleen cells or F-1 spleen cells 24 h later. Island, NY, USA) and washed twice. Spleen cells (20 Â 106 The effect of fludarabine on GVHD across major per recipient mouse) were injected intravenously into the histocompatibility (MHC) barriers was assessed in all these lateral tail vein. experiments.

Statistical evaluation The t-test was used for comparison between fludarabine- Results treated and control animals. Experiment 1: Four of the 10 mice receiving 0.3 and 0.5mg/kg fludarabine and cyclophosphamide survived Murine B-cell leukemia (BCL-1) more than 2 months. Mice that received 0.5mg/kg BCL-1, as previously described in BALB/c female mice,17 fludarabine only developed leukemia within 45days. Two was maintained in vivo by intravenous (i.v.) passage of 106– of the five mice treated with cyclophosphamide only 107 peripheral blood lymphocytes (PBLs) obtained from survived more than 2 months. In the group of six mice tumor-bearing mice. All untreated recipients of BCL-1 cells receiving 0.7 mg/kg fludarabine and the same dose of (X100 cells) consistently develop splenomegaly followed by cyclophosphamide, three (50%) survived, all control mice marked lymphocytosis in the peripheral blood and all developed leukemia within 34 days (Figure 1). animals die from leukemia.17 PBL counts of all experi- Experiment 2: Figure 2 summarizes the results of four mental groups were carried out weekly. Clinical onset of different experiments. A total of 25out of 33 (76%) control leukemia was defined as PBL counts exceeding 20 000/mm3. animals treated with saline and transplanted with allogeneic stem cells (C57) developed leukemia and died within 60 days, while in the fludarabine-treated group only nine out Monitoring of GVHD of 28 mice (32%) had leukemia within this period Mice were observed daily for survival and clinical signs of (P ¼ 0.006). GVHD, manifested as diarrhoea, weight loss and ruffled skin. In 12 surviving mice treated with fludarabine regimens, autopsy was performed. In these animals without clinical Experimental design evidence of leukemia, histopathology revealed that one had obvious leukemia in the spleen and liver, four had minor Basically, three different protocols were used. minimal residual disease, while seven of 12 had no signs of Experiment 1: This attempted to assess the effect of the leukemia at all. GVHD was either totally absent or minor fludarabine-containing regimens on BCL-1 leukemia in F-1 in the examined group (Figures 3–5). mice without transplant. Of the 33 control mice, only eight had no clinical Mice were injected with 104 BCL-1 leukemia cells and 1 evidence of leukemia. Six of these underwent autopsy, day later received 3 doses of fludarabine (0.3, 0.5, or 0.7 mg/ which showed that five had obvious leukemia and one had kg) i.p. for 3 days. A day after the administration of tumorous extramedullary hematopoiesis. Five of the six fludarabine, mice also received 100 mg/kg cyclophos- controls had GVHD. phamide. The mice were followed for 2 months and the Experiment 3: All six control mice without any incidence of leukemia and survival were recorded. conditioning regimens developed leukemia within 19–31 Experiment 2: Designed to assess the effect of fludara- days. In addition, nine out of 10 mice (90%) transplanted bine followed by cyclophosphamide and stem cell trans- with syngeneic (F1) spleen cells post-fludarabine treatment plantation given as treatment for BCL1 1eukemia in F-1 developed leukemia within 22–59 days. However, after

Bone Marrow Transplantation Fludarabine and transplantation in murine leukemia L Weiss et al 13

Figure 3 Histological section of the spleen showing numerous nodular leukemic infiltrates.

Figure 1 Leukemia-free survival rate of BCL-1 leukemia bearing (BALB/ c  C57BL/6) F1 nontransplant mice treated with various doses of fludarabine with or without cyclophosphamide. Control animals were treated with saline only.

Figure 4 High-power view of sheets of leukemic cells involving the spleen.

Figure 2 Leukemia-free survival rate of transplanted BCL-1 leukemic (BALB/c  C57BL/6) F1 mice treated with fludarabine or saline followed by infusion of C57BL/6 20  106 spleen cells. The figure represents four different experiments.

allogeneic transplantation with C57 spleen cells (post- fludarabine treatment and Cy) only three of the 11 (27.3%) mice developed leukemia (Figure 6) (P ¼ 0.0034). Figure 5 Histologic section of the skin showing GVHD and apoptotic keratinocytes.

Discussion human observations recorded in leukemic patients after allotransplantation using fludarabine-containing regimens. The results of the above studies provide the first experi- Clinical data in humans have shown that fludarabine in mental data in a murine leukemia model, which support the combination with other agents is highly effective as

Bone Marrow Transplantation Fludarabine and transplantation in murine leukemia L Weiss et al 14 GVHD and its sequelae still remain a major complica- tion in allotransplantation and the results of our study are indeed encouraging. The results of the present study and those of our earlier experiments16 imply that effective and successful immunomodulation and immunosuppression can be achieved in allogeneic transplantation when fludarabine is used together with cyclophosphamide. These properties of fludarabine have already been successfully exploited not only in the treatment of indolent lymphocytic leukemias/lymphomas, but also as a major part of the nonmyeloablative conditioning regimen used for stem cell transplant in humans. After using fludarabine-containing regimens, a potential state of host-versus-graft tolerance is achieved with the induction of partial chimerism. The latter serves as a basis for more successful engraftment, more effective GVL and less clinical and histopathological manifestations of GVHD.13,19–22 Experimental data cer- tainly justify the use of fludarabine-containing regimens in human transplantation.

Figure 6 Comparison of syngeneic and allogeneic transplantation in References fludarabine-treated BCL-1 bearing (BALB/c  C57BL/6) F1 mice. 1 Sorensen JM, Vena DA, Fallavollita A et al. Treatment of refractory chronic lymphocytic leukemia with fludarabine phosphate via the group C protocol mechanism of the National Cancer Institute: five year follow-up report. J Clin Oncol 1997; 15: 458–465. cytoreductive treatment for lymphoid leukemias, and can 2 Foran JM, Oscier D, Orchard J et al. Pharamacokinetic study also cause host immunomodulation as well as suffi- of single doses of oral fludarabine phosphate in patients with cient immunosuppression for successful allotransplan- ‘‘low-grade’’ non-Hodgkin’s lymphoma and B-cell chronic tation.15,18,19 Indeed fludarabine has already become lymphocytic leukemia. J Clin Oncol 1999; 17: 1574–1579. accepted as the anchor drug in these combination regimens 3 Johnson S, Smith AG, Loffler H et al. Multicentre prospective for successful non-myeloablative stem cell transplantation randomised trial of fludarabine versus cyclophosphamide in leukemias, lymphoma and in other malignancies.19–22 doxorubicin, and prednisone (CAP) for treatment of advanced Despite the successful use of fludarabine in humans, until stage chronic lymphocyte leukemia The French Cooperative Lancet recently there is very little experimental evidence-based Group on CLL. 1996; 347: 1432–1438. 4 Ghandi V, Estey E, Keating MJ, Plunkett W. Fludarabine available data in animals that can be used as a parallel for potentiates metabolism of cytarabine in patients with acute the successful experience in human leukemia and trans- myelogenous leukemia during therapy. J Clin Oncol 1993; 11: plantation. 116–124. In the studies reported here, fludarabine used in 5Ghandi V, Plunkett W, Weller S et al. Evaluation of the combination with cyclophosphamide achieved longer sur- combination of nelarabine and fludarabine in leukemias: vival in allotransplanted mice with lymphocytic leukemia clinical response, pharmacokinetics in leukemia cells. J Clin compared to syngeneic-transplanted animals or controls Oncol 2001; 19: 2142–2152. transplanted without fludarabine. In addition, these experi- 6 Yalman N, Sarper N, Devecioglu O et al. Fludarabine, mental animals also had less GVHD while the GVL effect cytarabine, G-CSF and idarubicin (FLAG-IDA) for the was still maintained. These findings support our earlier treatment of relapsed or poor risk childhood acute leukemia. Turk J Pediatr 2000; 42: 198–204. observations with nonleukemic mice, showing that flu- 7 Koller CA, Kantarjian HM, Feldman EJ et al. A phase I–II darabine impressively altered the incidence of GVHD trial of escalating doses of mitoxantrone with fixed doses of 16 transplanted across MHC barriers. These earlier studies cytarabine plus fludarabine as salvage therapy for patients with also showed that fludarabine can produce bilateral acute leukemia and the blastic phase of chronic myelogenous tolerance and stable chimerism after transplantation. leukemia. Cancer 1999; 86: 2246–2251. To the best of our knowledge, this experience is the only 8 Slavin S, Nagler A, Naparstek E et al. Nonmyeloablative stem murine experimental data available providing information cell transplantation and cell therapy as an alternative to on the successful use of fludarabine, combined with conventional bone marrow transplantation with lethal cytor- cyclophosphamide for allogeneic transplant of murine eduction for the treatment of malignant and nonmalignant hematologic diseases. Blood 1998; 91: 756–763. leukemia. Most recently, Petrus et al14 have shown that a 9 Aversa F, Tabilio A, Velardi A et al. Treatment of high-risk fludarabine-containing regimen prevents mismatched mur- acute leukemia with T-cell-depleted stem cells from related 23 ine marrow graft rejection, while Luznik et al have also donors with one fully mismatched HLA haplotype. New Engl J shown convincingly that a fludarabine-containing regimen Med 1998; 339: 1186–1193. can result in durable engraftment in incompatible murine 10 Borhnauser M, Thiede C, Schuler U et al. Dose-reduced donor transplants. conditioning for allogeneic blood stem transplantation: dur-

Bone Marrow Transplantation Fludarabine and transplantation in murine leukemia L Weiss et al 15 able engraftment without antithymocyte globulin. Bone 17 Slavin S, Strober S. Spontaneous murine B-cell leukemia. Marrow Transplant 2000; 26: 119–125. Nature 1978; 272: 624–626. 11 Priebe T, Platsoucas CD, Seki H et al. Purine nucleoside 18 Briz M, Cabrera R, Sanjuan I et al. Diagnosis of transfusion- modulation of functions of human lymphocytes. Cell Immunol associated graft-versus-host disease by polymerase chain 1990; 129: 321–328. reaction in fludarabine-treated B-chronic lymphocytic leuke- 12 Chun HG, Leyland-Jones B, Cheson BD. Fludarabine mia. Br J Haematol 1995; 91: 409–411. phosphate: a synthetic purine antimetabolite with significant 19 Wijermans PW, Gerrits WB, Haak HL. Severe immunodefi- activity against lymphoid malignancies. J Clin Oncol 1991; 9: ciency in patients treated with fludarabine monophosphate. 175–188. Eur J Haematol 1993; 50: 292–296. 13 Khouri IF, Keating M, Korbling M et al. Transplant-lite: 20 Nagler A, Slavin S, Varadi G et al. Allogeneic peripheral blood induction of graft-versus-malignancy using fludarabine-based stem cell transplantation using a fludarabine-based low nonablative chemotherapy and allogeneic blood progenitor- intensity conditioning regimen for malignant lymphoma. Bone cell transplantation as treatment for lymphoid malignancies. J Marrow Transplant 2000; 25: 1021–1028. Clin Oncol 1998; 16: 2817–2824. 21 Childs R, Chernoff A, Contentin N et al. Regression of 14 Petrus MJ, Williams JF, Eckhaus MA et al. An immuno- metastatic renal-cell carcinoma after nonmyeloablative allo- ablative regimen of fludarabine and cyclophosphamide pre- geneic peripheral blood stem-cell transplantation. New Engl J vents fully MHC-mismatched murine marrow graft rejection Med 2000; 343: 750–758. independent of GVHD. Biol Blood Marrow Transplant 2000; 6: 22 Wasch R, Reisser S, Hahn J et al. Rapid achievement of 182–189. complete donor chimerism and low regimen-related toxicity 15Goodman ER, Fiedor PS, Fein S et al. Fludarabine phosphate: after reduced conditioning with fludarabine, carmustine, a DNA synthesis inhibitor with potent immunosuppressive melphalan and allogeneic transplantation. Bone Marrow activity and minimal clinical toxicity. Am Surg 1996; 62: 435– Transplant 2000; 26: 243–250. 442. 23 Luznik L, Jalla S, Engstrom LW et al. Durable engraftment of 16 Or R, Weiss L, Amir G et al. The prophylactic potential of major histocompatibility complex-incompatible cells after fludarabine monophosphate in graft-versus-host disease after nonmyeloablative conditioning with fludarabine, low-dose bone marrow transplantation in murine models. Bone Marrow total body irradiation, and posttransplantation cyclopho- Transplant 2000; 25: 263–266. sphamide. Blood 2001; 98: 3456–3464.

Bone Marrow Transplantation