Bone Marrow Transplantation (2000) 25, 263–266  2000 Macmillan Publishers Ltd All rights reserved 0268–3369/00 $15.00 www.nature.com/bmt The prophylactic potential of fludarabine monophosphate in graft- versus-host disease after bone marrow transplantation in murine models

ROr1, L Weiss1, G Amir2, S Tejman1 and A Polliack3

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

Summary: impairing the graft-versus-leukemia (GVL) effect, because these two alloimmune-mediated reactivities appear to be Fludarabine phosphate, a currently mutually dependent.2,4–8 It follows, therefore, that an agent used in the therapy of hematological malignancies, is capable of modifying, or even eliminating, GVHD without known to cause immunosuppression and long-lasting T interfering with GVL could be of great clinical value. cell lymphopenia. In this study, the effect of fludarabine One therapeutic option to be explored in this respect is on murine graft-versus-host disease occurring after fludarabine monophosphate, a novel purine analogue which marrow transplantation across major and minor histo- has been extensively used in the past few years in the treat- compatibility barriers was evaluated. Survival of ment of chronic lymphocytic leukemia (CLL) and low- BALB/c ؋ C57BL/6)F1 mice irradiated and trans- grade non-Hodgkin’s (NHL) with impressive) planted across the major histocompatibility barrier with success.9–11 Lately, fludarabine has also been shown to be C57BL/6 spleen cells, and subsequently treated with effective in acute when used in a variety of com- fludarabine was significantly longer than that of the bination regimens.12 During these clinical control animals (P Ͻ 0.0001). On the other hand, fluda- trials, long-lasting T and B cell lymphopenia has been rabine had no effect on the graft-versus-host disease and observed after a single cycle of therapy, suggesting that this survival of CBA mice transplanted by B10.BR and of agent possesses immunosuppressive activity.13,14 The latter BALB/c mice transplanted by B10.D2 spleen cells across property of fludarabine has led us to investigate whether it the minor histocompatability barrier. The results indi- is feasible to utilize the drug for GVHD prevention post cate that in certain murine models, particularly a major allogeneic marrow transplantation in murine models. mismatch, fludarabine has the potential to induce bilateral tolerance and stable chimerism after marrow transplantation. Bone Marrow Transplantation (2000) 25, 263–266. Materials and methods Keywords: fludarabine monophosphate; bone marrow transplantation; graft-versus-host disease; murine model Animals All animal procedures utilized in the present study were approved by the Institutional Committee for Animal Exper- One of the most frequent complications of allogeneic bone imentation. One major histocompatibility (MHC) antigen marrow transplantation (BMT) is the development of graft- mismatch and two minor histocompatibility (MiHC) mis- versus-host disease (GVHD). The incidence of the acute match mouse BMT models were set up. Two- to 6-month- form of GVHD in transplant patients ranges between 50%– old male and female mice were used in various donor– 1–3 80%, with a mortality of up to 30%. This unacceptably recipient combinations. high incidence of mortality associated with GVHD occurs MHC-BMT model (BALB/c ϫ C57BL/6)F1 (H-2d/b) despite optimal immunosuppressive preventive treatment transplanted with C57BL/6 (H-2b) marrow: the mice were routinely given to patients undergoing allogeneic BMT. purchased from the Harlem Breeding Facility (Jerusalem, Among the narrow spectrum of prophylactic GVHD agents Israel). available today, the most frequently used are cyclosporine, MiHC-BMT model CBA (H-2K) transplanted with tacrolimus (FK506), and steroids. Although B10.BR marrow and BALB/c (H-2d) with B10.D2 (H-2d) T cell depletion of the allograft is an effective preventive marrow: the mice were obtained from Jackson Laboratories measure, it carries the inherent disadvantage of severely (Bar Harbor, ME, USA). The mice were kept in a standard animal facility with top-filtered cages. Cages, sawdust and water bottles were Correspondence: Dr R Or, Department of Bone Marrow Transplantation, Hadassah University Hospital, Ein Kerem, PO Box 12000, Jerusalem autoclaved once a week. Neomycin sulfate, at a concen- 91120, Israel tration of 0.5%, was given in the drinking water for 2 weeks Received 24 August 1998; accepted 16 June 1999 post transplantation. Prophylactic fludarabine in GVHD after BMT in mice ROret al 264 Experimental design 1.00 Transplant candidate mice were exposed to total body irradiation (TBI) 1 day prior to BMT. After BMT, fludara- bine monophosphate (Schering, Berlin, Germany) was 0.80 administered intraperitoneally in a dose of 0.025 mg/mouse on days ϩ1, ϩ4, ϩ8 and twice weekly for 3 weeks there- after. The doses of fludarabine for the current study were 0.60 chosen after initial pilot experiments, testing 0.03 mg/kg, 0.08 mg/kg and 0.25 mg/kg for cytotoxicity (data not shown). These preliminary studies were an attempt to estab- lish the minimal drug dosage having the potential of an 0.40 immunosuppressive/immunomodulatory effect without Cumulative Survival associated myeloablation, a drawback of full therapeutic dosages in the animals. The dosages of fludarabine applied 0.20 in humans ranges between 25 mg/m2 and 30 mg/m2, the equivalent of which was found to be toxic in the present murine models. Control animals received saline only on the same days. In each BMT model, experimental and control 0 groups consisted of eight mice, respectively. Each experi- 0510 15 20 25 30 35 ment was performed three times. Two to three weeks post Days transplant the number of donor-type cells present in the blood was determined in vitro by the complement- Figure 1 Survival rate of irradiated (BALB/c ϫ C57BL/6) F1 mice infused with C57BL/6 20 ϫ 106 spleen cells followed by administration dependent microcytotoxic test. of fludarabine ࡯, or saline b. The Figure represents the Kaplan–Meier survival analysis of one out of three similar experiments. Total body irradiation (TBI) ing animals (Figure 1). Survival rates of the fludarabine- Mice were placed in radiation chambers and exposed to a treated animals were consistently higher than those of the total lethal dose of 600 cGy, delivered from a linear accel- controls during the experimental period. Four weeks after erator (Varian Clinac 6X; Varian Oncology Systems, Zug, BMT, the survival rate of the fludarabine-treated mice was Switzerland) at a dose rate of 170 cGy/min, at a source-to- 45% (P Ͻ 0.0001) (Figure 1). GVHD was the cause of skin distance of 80 cm. death in all non-survivors.

Effect of fludarabine on GVHD across the minor Bone marrow transplantation histocompatibility (MiHC) barrier Spleens were removed aseptically from donor mice, teased In the CBA and BALB/c, both the experimental and control through a nylon mesh into RPMI 1640 medium (Gibco, marrow recipients developed mild clinical signs of GVHD. Grand Island Biological Co, Grand Island, NY, USA) and As shown in Figure 2, survival of the fludarabine and con- washed twice. Spleen cells (20 ϫ 106 per recipient mouse) were injected intravenously into the lateral tail vein. Monitoring of GVHD was performed by recording 100 weight loss and survival twice weekly in all groups.

80 Statistical evaluation The log-rank test was used for comparison of survival 60 between fludarabine-treated and control animals.

40 % Survival

Results 20 Effect of fludarabine on GVHD across the major histocompatibility (MHC) barrier 0 Thirty-four of 36 F1 controls (saline) developed GVHD as 0 10 20 30 manifested by severe diarrhea, skin ruffling and consider- Days able weight loss. In the fludarabine-treated counterparts, Figure 2 Survival of CBA mice inoculated with B10.BR 20 ϫ 106 16/35 mice survived with minimal clinical manifestations spleen cells followed by administration of fludarabine ࡯, or saline „. The of GVHD, and weight loss was not observed in the surviv- Figure represents the results of one out of three similar experiments.

Bone Marrow Transplantation Prophylactic fludarabine in GVHD after BMT in mice ROret al 265 trol CBA mice were the same. Similar results were obtained peutic efforts have attempted to establish a potential state using BALB/c mice transplanted with B10.D2 spleen cells of host-versus-graft tolerance as a result of induction of (data not shown). transient mixed chimerism, which may protect the host from severe GVHD, while at the same time endeavoring to 22 Effect of fludarabine on chimerism in the MHC model preserve the GVL effect. The immunocompetence which is affected by fludara- Engraftment was achieved in all experimental mice with no bine, a process occurring also with other purine ana- significant difference in the percentage of donor-type cells logues,25 causes prolonged T cell dysfunction, particularly in the blood of control and fludarabine-treated animals, both in CLL patients receiving the drug repeatedly over rela- groups displaying 70–80% degree of chimerism. tively long periods of time. Furthermore, the immunomodu- latory effect of fludarabine is also evident in CLL patients who may develop auto-immune hemolytic anemia, suggest- Discussion ing imbalance between B and T cell subsets arising during treatment.26 Very rarely other patients may acquire trans- GVHD and the sequelae of its various treatments remain a fusion-associated GVHD,27–29 indicating purine analogue- major complication of allogeneic BMT.2,3,15–17 The research induced elimination of the recipient’s GVHD-abortive performed here is an attempt to assess the effect of low- capacity of allogeneic cells.30 It is indeed feasible that flu- dose fludarabine on GVHD and its possible future role in darabine may affect antigen-presenting cells or monocytes, the prevention of this disorder. The experimental data or even have a polarizing influence on Th1 and Th2 helper obtained in this study suggest that fludarabine may decrease lymphocyte subsets in the host,14,18 all of which may be the incidence and degree of GVHD after allogeneic BMT implicated in the pathogenesis of auto- or alloimmune in a murine model. These results lend support to recent phenomena. observations on the potent immunosuppressive activity of The finding that fludarabine altered the incidence and this purine analogue observed after skin allografting in extent of GVHD in a specific allogeneic BMT setting, with- monkeys.18 If and when fludarabine is established as an out affecting the degree of chimerism, underscores the effective mode of prevention and treatment of GVHD, the immunomodulatory properties of this drug. The above drug’s possible interference with GVL will have to be experimental data and the early experience gained in allo- examined. In this context it should be mentioned that the geneic BMT after the use of fludarabine in CLL patients widely used GVHD prophylactic drug cyclosporin might suggest that it may be justified to explore use of the drug impair the anti-leukemic effect of allogeneic immune cells during the course of BMT in the hope that hard evidence against residual clonogenic leukemia cells.19 for its modulatory effect could be found. A further aspect Fludarabine is an effective lymphotoxic agent that abro- then is that relatively small doses of the drug, as used in gates T and B cell function, both in vitro and in vivo.13,18 the present study, may well prove to decrease GVHD while Its mode of action involves the inhibition of DNA synthesis preserving, or even enhancing, the effect of GVL post and the subsequent fragmentation of nuclear DNA, thereby marrow transplantation. The results of additional experi- possibly blocking cellular repair, and culminating in the ence with fludarabine in the setting of allogeneic BMT will induction of cellular apoptosis through an as yet unknown be awaited with some expectation. intracellular pathway. Normal CD4ϩ and CD8ϩ T cells are more sensitive to the effect of the drug than CD20ϩ B cells. Indeed, in patients with lympho-proliferative disorders Acknowledgements receiving fludarabine, both normal T and B cell levels have been shown to drop by 90% and 50%, respectively, even This work was supported in part by an educational grant from after a single cycle of therapy.13,14,20 Schering, Berlin, Germany. The drug’s selective lympholytic activity holds promise for effective immunosuppression in some disorders, and may ultimately have an impact on the clinical approach References to GVHD. The prevention of GVHD by fludarabine in the present 1 Thomas ED, Storb R, Clift RA et al. Bone marrow transplan- study was particularly effective after major mismatch trans- tation. New Engl J Med 1975; 292: 832–843. plantation but not really significant after minor mismatched 2 Nash RA, Storb R. Graft-versus-host effect after allogeneic transplantation (B10.BR → CBA mice). The paradoxical hematopoietic stem cell transplantation: GVHD and GVL. effect of fludarabine after BMT in the MiHC and MHC Curr Opin Immunol 1996; 8: 674–680. models may reflect immunomodulation rather than frank 3 Ferrara JLM. Cytokine dysregulation as a mechanism of graft- immunosuppression. In this context, it is of interest that versus-host disease. Curr Opin Immunol 1993; 5: 794–799. mention has been made of clinical impressions regarding 4 Horowitz MM, Gale RP, Sondel PM et al. Graft-versus- patients with CLL who received fludarabine as therapy leukemia reactions after bone marrow transplantation. Blood 1990; 75: 555–562. prior to undergoing allogeneic BMT in whom the GVHD 21 5 Slavin S, Ackerstein A, Naparstek E et al. The graft-versus- manifestations were less grave. More recently the leukemia (GVL) phenomenon: is GVL separable from immunomodulatory effects of fludarabine have also been GVHD? Bone Marrow Transplant 1990; 6: 155–161. utilized in different combination regimens as part of non- 6 Truitt RL, Atasoylu AA. Contribution of CD4ϩ and CD8ϩ myeloablative conditioning prior to BMT.22–24 These thera- T cells to graft-versus-host disease and graft-versus-leukemia

Bone Marrow Transplantation Prophylactic fludarabine in GVHD after BMT in mice ROret al 266 reactivity after transplantation of NHC-compatible bone methylprednisolone on the graft-versus-leukemia effects marrow. Bone Marrow Transplant 1991; 8: 51–58. across major histocompatability barriers in mice following 7 Weiden PL, Flournoy N, Thomas ED et al. Antileukemic allogeneic bone marrow transplantation. Bone Marrow Trans- effect of graft-versus-host disease in human recipients of allo- plant 1990; 6: 229–233. geneic-marrow grafts. New Engl J Med 1979; 300: 1068– 20 Chun HG, Leyland-Jones B, Cheson BD. Fludarabine phos- 1073. phate: a synthetic purine with significant 8 Sullivan KM, Storb R, Buckner CD et al. Graft-versus-host activity against lymphoid malignancies. J Clin Oncol 1991; 9: disease as adoptive immunotherapy in patients with advanced 175–188. hematologic neoplasms. New Engl J Med 1989; 320: 828–834. 21 Mehta J, Powles R, Singhal S et al. T cell-depleted allogeneic 9 Pigaditou A, Rohatiner AZS, Whelan JS et al. Fludarabine bone marrow transplantation from a partially HLA-mis- in low-grade lymphoma. Semin Oncol 1993; 20 (Suppl. 7): matched unrelated donor for progressive chronic lymphocytic 24–27. leukemia and fludarabine-induced bone marrow failure. Bone 10 Johnson S, Smith AG, Loffler H et al. Multicentre prospective Marrow Transplant 1996; 17: 881–883. randomised trial of fludarabine versus , 22 Slavin S, Nagler A, Naparstek E et al. Nonmyeloablative stem , and prednisone (CAP) for treatment of advanced- cell transplantation and cell therapy as an alternative to con- stage chronic lymphocyte leukaemia. The French Cooperative ventional bone marrow transplantation with lethal cytoreduc- Group on CLL. Lancet 1996; 347: 1432–1438. tion for the treatment of malignant and nonmalignant hematol- 11 O’Brien S, Kantarjian H, Beran M et al. Results of fludarabine ogic diseases. Blood 1998; 91: 756–763. and prednisone therapy in 264 patients with chronic lympho- 23 Giralt S, Estey E, Albitar N et al. Engraftment of allogeneic cytic leukemia with multivariate analysis-derived prognostic hematopoietic progenitor cells with purine analog-containing model for response to treatment. Blood 1993; 82: 1695–1700. chemotherapy: harnessing graft-versus-leukemia without mye- 12 Gandhi V, Estey E, Kentin MJ, Plunkett W. Fludarabine loablative therapy. Blood 1997; 89: 4531–4536. potentiates metabolism of in patients with acute 24 Kapelushnik J, Or R, Slavin S, Nagler A. A fludarabine-based protocol for bone marrow transplantation in Fanconi’s anemia. myelogenous leukemia during therapy. J Clin Oncol 1993; 11: Bone Marrow Transplant 1997; 20: 1109–1110. 116–124. 25 Priebe T, Platsoucas CD, Seki H et al. Purine nucleoside 13 Wijermans PW, Gerrits WB, Haak HL. Severe immunodefi- modulation of functions of human lymphocytes. Cell Immunol ciency in patients treated with fludarabine monophosphate. 1990; 129: 321–328. Eur J Haematol 1993; 50: 292–296. 26 Myint H, Coppelstone JA, Orchard J et al. Fludarabine-related 14 Dighiero G. Potential immunological action of purine nucleo- autoimmune haemolytic anaemia in patients with chronic lym- side analogues. Drugs 1994; 47 (Suppl. 6): 57–62. phocytic leukaemia. Br J Haematol 1995; 91: 341–344. 15 Gratwohl A, Hermans J, Apperley J et al. Acute graft-versus- 27 Briz M, Cabrera R, Sanjua´nIet al. Diagnosis of transfusion- host disease: grade and outcome in patients with chronic myel- associated graft-versus-host disease by polymerase chain reac- ogenous leukemia. Blood 1995; 86: 813–818. tion in fludarabine-treated B-chronic lymphocytic leukaemia. 16 Chao NJ, Schmidt GM, Niland JC et al. Cyclosporine, metho- Br J Haematol 1995; 91: 409–411. trexate, and prednisone compared with cyclosporine and 28 Zulian GB, Roux E, Tiercy JM et al. Transfusion-associated prednisone for prophylaxis of acute graft-versus-host disease. graft-versus-host disease in a patient treated with New Engl J Med 1993; 329: 1225–1230. (2-chlorodeoxyadenosine): demonstration of exogenous DNA 17 Nash RA, Etzioni R, Storb R et al. Tacrolimus (FK506) alone in various tissue extracts by PCR analysis. Br J Haematol or in combination with methotrexate or methylprednisolone 1995; 89: 83–89. for the prevention of acute graft-versus-host disease after mar- 29 Briones J, Pereira A, Alcorta I. Transfusion-associated graft- row transplantation from HLA-matched siblings: a single- versus-host disease (TA-GVHD) in fludarabine-treated center study. Blood 1995; 85: 3746–3753. patients: is it time to irradiate blood component? Br J Haema- 18 Goodman ER, Fiedor PS, Fein S et al. Fludarabine phosphate: tol 1996; 93: 737–741. a DNA synthesis inhibitor with potent immunosuppressive 30 Lee TH, Donegan E, Slichter S, Busch MP. Transient increase activity and minimal clinical toxicity. Am Surg 1996; 62: in circulating donor leukocytes after allogeneic transfusions in 435–442. immunocompetent recipients compatible with donor cell pro- 19 Weiss L, Reich S, Slavin S. Effect of cyclosporine A and liferation. Blood 1995; 85: 1207–1214.

Bone Marrow Transplantation