Bone Marrow Transplantation (2006) 37, 745–749 & 2006 Nature Publishing Group All rights reserved 0268-3369/06 $30.00 www.nature.com/bmt

ORIGINAL ARTICLE Fludarabine, cyclophosphamide and horse antithymocyte globulin conditioning regimen for allogeneic peripheral blood stem cell transplantation performed in non-HEPA filter rooms for multiply transfused patients with severe aplastic anemia

R Kumar1, S Prem1, M Mahapatra1, T Seth1, DR Chowdhary1, P Mishra1, L Pillai1, AMVR Narendra1, NK Mehra2, R Saxena1 and VP Choudhry1

1Department of Hematology, All India Institute of Medical Sciences, New Delhi, India and 2Department of Transplant Immunology and Immunogenetics, All India Institute of Medical Sciences, New Delhi, India

Multiply transfused patients of severe aplastic anemia are Introduction at increased risk of graft rejection. Five such patients underwent peripheral blood stem cell transplantation from Bone marrow transplantation (BMT) is a curative therapy HLA-identical siblings with a fludarabine-based protocol. for aplastic anemia, but the incidence of graft rejection in The conditioning consisted of fludarabine 30 mg/m2/ older literature is reported as between 30 and 60%.1 With day  6 days, cyclophosphamide 60 mg/kg/day  2 days improvement in transplant practices, the graft failure rates and horse antithymocyte globulin (ATG)  4 days. Two have fallen to about 10%.2 In developing countries, most different ATG preparations were used: ATGAM (dose patients are multiply transfused before transplant. Owing 30 mg/kg/day  4 days) or Thymogam (dose 40 mg/kg/ to economic reasons, leukocyte filters are rarely used and day  4 days). Engraftment: median time to absolute occasionally blood from family donors is transfused neutrophil count (ANC) 40.5  109/l was 11 days (range: because of blood shortage. Therefore, the risk of rejection 8–17) and median time to platelet count 420  109/l was is much higher. Even with current conditioning protocols 11 days (range: 9–17). At a median follow-up of 171 days and immunosuppression, the graft failure rates in HLA- (range: 47–389), there has been no graft rejection and all identical sibling transplants have been reported as 31% patients are in complete remission. Acute GVHD (grade from India.3 We present our experience of a fludarabine, 1) occurred in one patient only. Chronic GVHD developed cyclophosphamide and antithymocyte globulin (ATG) in two patients (extensive in one and limited in another). conditioning regimen for hematopoietic stem cell trans- The transplants were performed in non-HEPA filter plantation (HSCT) from HLA-identical siblings. rooms. In only one patient, systemic antifungal therapy There is no standard ATG preparation recommended for (voriconazole) was used. The use of Thymogam brand HSCT. Some studies have preferred rabbit ATG over horse of ATG for conditioning is being reported for the first ATG4,5 for conditioning in aplastic anemia. We have used time. Our experience suggests that this fludarabine-based horse ATG in our patients, from two different sources – protocol allows rapid sustained engraftment in high-risk Pfizer (USA) and Bharat Serums and Vaccines (India). The patients without significant immediate toxicity. Indian preparation has been used for HSCT for the first Bone Marrow Transplantation (2006) 37, 745–749. time. doi:10.1038/sj.bmt.1705321; published online 6 March 2006 Keywords: aplastic anemia; fludarabine; antithymocyte globulin Patients and methods

Patient and donor selection From November 2004 to October 2005, five patients with severe aplastic anemia underwent allogeneic peripheral blood stem cell transplant from six-antigen HLA-matched sibling donors at our center. All had undergone a bone marrow aspirate and biopsy to confirm bone marrow Correspondence: Professor R Kumar, Department of Hematology, All aplasia, and peripheral blood counts showed pancytopenia India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, consistent with severe aplastic anemia. Fanconi’s anemia India. E-mail: [email protected] was excluded by stress cytogenetics using mitomycin C. Received 8 December 2005; revised 26 January 2006; accepted 27 Donors were selected based on HLA molecular typing for January 2006; published online 6 March 2006 class I and II antigens. Fludarabine for transplant in aplastic anemia R Kumar et al 746 Conditioning regimen and stem cell source were added if fever persisted for more than 3 days while All patients received the same conditioning regimen. This on antibiotic cover. comprised of fludarabine 30 mg/m2/day (days À10 to À5), cyclophosphamide 60 mg/kg/day for 2 days (day À6 and Engraftment À5) along with hydration and intravenous mesna and Day of myeloid engraftment was defined as the first of 3 ATG. Two different ATG preparations were used, based consecutive days with ANC more than 0.5 Â 109/l. Platelet on economic criteria. The first preference was ATG engraftment was defined as unsupported platelet count (ATGAM) from Pfizer (Pharmacia & Upjohn Company, maintained above 20 Â 109/l. Kalamazoo, Michigan 49001, USA) in a dose of 30 mg/kg/ day. If this was not affordable, the alternative was ATG (Thymogam) from Bharat Serums and Vaccines (Mumbai, Results India) and, on recommendations of the company, a higher dose of 40 mg/kg/day was used. Antithymocyte globulin Patient characteristics was administered for 4 days (days À5toÀ2). The median age of the patient group was 18 years (range The stem cell source was G-CSF-mobilized peripheral 17–36 years). All the patients were multiply transfused with blood stem cells, which were infused on day 0 of the non-leukocyte-depleted blood products; the median num- conditioning regimen. The minimum target cell dose was ber of transfused packed red cells/platelet concentrates was 3 Â 106 cells/kg CD34 þ cells. If the dose collected was less 6 30 (range 23–47). Before the transplant, all the patients than 3 Â 10 cells/kg, the harvest was repeated the next day were on androgens (stanozolol 2 mg/kg/day) and three to increase the dose and both days were taken as day 0. The þ patients were also receiving cyclosporine 5 mg/kg/day. CD34 cell estimation was made on the final collection of Patient number 4 had a history of seizure disorder for apheresis product, and the report was available within 2 h. þ which she had been receiving carbamazepine, where as The CD34 cell estimation was made by flow cytometry patient number 1 had cyanotic heart disease, which had 6 using the Milan Mulhouse Protocol using a FACS Calibur been operated in the first year of life. The median duration (Becton Dickinson). For each estimation, 50 000 cells were þ of illness before transplant was 8 months (range 1.5–108 studied, with CD34 cells expressed as a percent of the months) (Table 1). total population of leukocytes. Patient number 1 was admitted and treated 1 month before transplant for pneumonia and a week before GVHD prophylaxis transplant he had fever with a furuncle on the face. Patient Patients received cyclosporine 5 mg/kg/day intravenously in number 3 had developed appendicitis 3 months before two divided doses from day À1 and the dose was adjusted transplant and was treated conservatively resulting in to keep trough levels between 200 and 300 ng/ml. This was formation of an appendicular lump. There was recurrence changed to appropriate oral dose by days 10–14, and was of local pain with fever 1 month before transplant, and continued at full doses for 6 months, then gradually tapered cellulitis of face 10 days before the transplant. Patient off over 6 months. Methotrexate was given intravenously number 4 had a history of fever and respiratory infection at a dose of 15 mg/m2 on day 1 and 10 mg/m2 on days 3, 6 2 weeks before the transplant. and 11. Cell dose, engraftment and toxicity þ 6 Supportive care The median dose of CD34 cells infused was 4.81 Â 10 /kg As HEPA filtered rooms are not available in our center, (range 1.9–9.15). In patient 3, the dose of stem cells patients were nursed in non-HEPA filtered air conditioned collected was below the desired target. The median day of private rooms with barrier nursing; wet mopping of the myeloid engraftment was 11 days (range 8–17 days) and all 9 room was carried out twice daily, and entry was restricted patients also attained an ANC of more than 1.0 Â 10 /l on to two attendants. Cooked food and mineral or boiled the same day. Unsupported platelet count of more than 9 water was allowed. Patients received anti-microbial pro- 20 Â 10 /l was attained at a median of 11 days (range 9–17 9 phylaxis: oral ciprofloxacin, fluconazole and acyclovir as days) and 450 Â 10 /l at a median of 13 days (range 9–17 antibacterial, antifungal and antiviral prophylaxis, respec- days). Patient number 2 underwent ABO mismatched tively. Oral Co-trimoxazole was started on day À6, transplant (donor B þ ; recipient A þ ) and had delayed discontinued 1 day before stem cell infusion and restarted erythroid engraftment, for which erythropoeitin was after neutrophil recovery. Hemoglobin was maintained administered for 2 months post transplant and the above 9 g/dl and platelet counts above 20 Â 109/l by blood hemoglobin normalized. There was minimal conditioning and platelet transfusions (irradiated and filtered). G-CSF induced toxicity. All the patients had mild infusion-related was started 24 h after stem cell infusion and discontinued fever during infusion of ATG. There was no serum sickness when the ANC remained above 0.5 Â 109/l on 2 consecutive or anaphylaxis. There was no veno-occlusive disease. days. Fever was defined as a temperature of more than 1011F, Infections or more than 100.41F for more than 1 h. Any febrile During conditioning, three patients developed fever for episode was treated with broad-spectrum antibiotics. First which they received broad-spectrum antibiotics (Table 1). line cover consisted of piperacillin-tazobactam (Zosyn), After day 0, all patients had episodes of febrile neutropenia, amikacin7teicoplanin (Targocid). Systemic antifungals but focus of infection could be documented only in patient

Bone Marrow Transplantation Fludarabine for transplant in aplastic anemia R Kumar et al 747 Table 1 Patient characteristics, engraftment data and follow-up

Patient 1 Patient 2 Patient 3 Patient 4 Patient 5

Sex/age (years) M/17 F/24 M/18 F/17 M/36 Symptom duration (months) 12 6 108 1.5 8 before transplant Prior transfusions 30 26 47 23 33 Prior therapy Sta, CsA Sta, CsA Sta Sta Sta, CsA Coexisting morbidity Operated cyanotic Seizures heart disease Donor sex/age F/18 M/29 M/20 F/11 M/32 TNC ( Â 108/kg) 12.54 15.88 3.66 13.38 7.86 MNC ( Â 108/kg) 5.19 6.35 1.77 6.02 3.06 CD34+ ( Â 106/kg) 5.94 9.15 1.936 4.81 3.16 ANC40.5 Â 109/l day+11 day+17 day+13 day+10 day+8 ANC41.0 Â 109/l day+11 day+17 day+13 day+10 day+8 Platelet420 Â 109/l day+11 day+17 day+13 day+10 day+9 Platelet450 Â 109/l day+11 day+17 day+13 day+13 day+9 Fever during From day — — From day From day conditioning À5 Â 1 day À7 Â 2 days –8 Â 2 days Localization Nil Nil Nil Nil Perianal Fever after From day From day From day From day From day stem cell infusion +4 Â 3 days 0 Â 3 days 0 Â 3 days +3 Â 2 days +5 Â 3 days Localization — — — Pneumonia, Hickman — tunnel infection Antimicrobials used Z, A, T, I Z, A, T I, A, levofloxacin Z, A, T, M I, A, metronidazole, M, T, A during transplant Antifungals used — — — Voriconazole —

Abbreviations: A ¼ amikacin; CsA ¼ cyclosporin A; I ¼ Imipenem; M ¼ Meronem (meropenem); Sta ¼ stanazolol; T ¼ Targocid (teicoplanin); Z ¼ Zosyn (piperacillin-tazobactam). number 4. Systemic antifungal therapy with voriconazole have evolved over time and are identified as a low marrow was started for patient 4 for pneumonia, which occurred cell dose,7,8 pretransplant blood transfusions (20 or more), while on systemic antibiotics. pretransplant infections, interval between diagnosis and transplant more than 1 month,9 and any therapy consisting GVHD of immunosuppression or androgens, before the trans- Patient number 1 had grade 2 acute GVHD (hepatic), plant.10 The outcome in patients with graft failure is poor, which developed on day 38 post transplant. He was treated especially in those with nonengraftment.11 In those patients with oral steroids, which were subsequently tapered. At day who have initial engraftment but subsequent rejection, 184 post transplant, he developed extensive skin GVHD about 60% can be salvaged with a second bone marrow (biopsy proven) with subsequent eye, oral mucosa and liver transplant.11 Most patients in developing countries have involvement. Initially, corticosteroids were added to multiple poor risk factors, and owing to economic and cyclosporine, but subsequently these were replaced by resource constraints, a second transplant is not a feasible mycophenolate mofetil from the 10th month post trans- option. It is therefore critical to minimize the chances of plant. No other patient has developed acute GVHD. graft failure in the first transplant. Patient number 3 developed limited chronic GVHD Increasing intensity of conditioning is one strategy to involving the liver on day 114 post transplant, which reduce graft rejection. Addition of radiation to cyclopho- resolved on adding oral steroids to cyclosporine. sphamide has reduced graft failure, but not improved survival owing to higher risk of GVHD, interstitial Survival and outcome pneumonia and late complications of radiation.2,10 Addi- At a median follow-up of 171 days (range 47–389), all tion of ATG to cyclophosphamide has led to superior patients are alive and in complete remission. One patient survival12 and is considered standard therapy for multiply has extensive chronic GVHD controlled on immuno- transfused patients.13,14 However, the high graft rejections suppression. No late infections or other complications seen in India with these protocols3 and subsequent poor have occurred. No molecular chimerism studies were outcome made us search for more intense conditioning performed. In patients number 1, 2 and 5, where there regimens. was ABO blood group incompatibility, a change of blood Addition of fludarabine to cyclophosphamide and ATG group was documented. has been recommended for unrelated transplant in severe aplastic anemia13 and has been reported in limited series for Discussion severe aplastic anemia4,15 and in Fanconi’s anemia.16 We chose to use this option in related transplants, because of The problems of bone marrow transplant in developing the multiple risk factors for graft failure in our patients. countries are similar to those seen two decades earlier in the To reduce chances of graft failure, supplementation with West. The risk factors for graft failure or poor survival donor buffy coat cells has been used, but this resulted in

Bone Marrow Transplantation Fludarabine for transplant in aplastic anemia R Kumar et al 748 more chronic GVHD.1,2 The use of G-CSF-mobilized Systemic antifungal agents were not required in four of peripheral blood stem cells (PBSC) for transplants results our patients. In patient 4, voriconazole was added in a larger collection of stem cells, and presumably less empirically on development of pneumonia, at an early graft failure, but with an increased risk of chronic stage. The possible reasons for the low rate of infective GVHD.2,17 The other advantage of PBSC transplant is complications are the use of prophylactic ciprofloxacin/ more rapid neutrophil and platelet engraftment,18 which fluconazole and early empirical use of effective antimicro- would reduce the risk of early infections and cost of bials. An early neutrophil recovery owing to use of PBSC supportive therapy. As we do not have facilities of HEPA supplemented by G-CSF is likely to have contributed to the filter rooms, we hypothesized that early engraftment would favorable outcome. minimize the risk of increased infections. We therefore This experience shows that in patients of severe aplastic chose to use PBSC as the stem cell source, although we do anemia at high risk of graft rejection, a conditioning not know if the potential advantages would be neutralized regimen of fludarabine, cyclophosphamide and ATG by chronic GVHD. resulted in successful engraftment. A longer follow-up will Fludarabine-based regimens for HSCT in aplastic show if use of ATG in conditioning can reduce the anemia have been described, but there has been consider- incidence and severity of chronic GVHD. In countries with able variation in doses and preparation.4,13,15 Antithymo- similar patient profiles and shortage of conventional BMT cyte globulin given during conditioning is known to reduce units, such a strategy appears promising. both acute and chronic GVHD, especially if it is given nearer the day of transplant.19 The circulating ATG produces an in vivo T-cell depletion of the infused donor Acknowledgements T cells, as serum levels may be quite high at the time of transplant. As we were using PBSC, we used a high dose of We thank Dr A Srivastava and Dr M Chandy (Christian ATG to reduce the chances of chronic GVHD. Medical College, Vellore, India) for sharing their experience in We have been using horse ATG (ATGAM) for using different protocols for BMT in aplastic anemia. immunosuppressive therapy for our aplastic anemia pa- tients and preferred to continue with the same preparation. Unfortunately, the preparation is expensive and was References unaffordable in two cases, patients 1 and 5. In these patients, economic considerations made us use the Indian 1 Deeg HJ, Self S, Storb R, Doney K, Appelbaum FR, preparation, as it less expensive. However, there is no Witherspoon RP et al. Decreased incidence of marrow graft literature on use of this preparation for HSCT. As our rejection in patients with severe aplastic anemia: changing preliminary experience with Thymogam as an immuno- impact of risk factors. Blood 1986; 68: 1363–1368. suppressant in aplastic anemia was encouraging, we decided 2 Horowitz MM. Current status of allogeneic bone marrow to use it with patients’ informed consent. On advice of the transplantation in acquired aplastic anemia. Semin Hematol company, we chose a higher dose than with ATGAM. In 2000; 37: 30–42. patient 1, the estimated cost of ATGAM at a dose of 3 Chandy M, Srivastava A, Dennison D, Mathews V, George B. Allogeneic bone marrow transplantation in the developing 30 mg/kg/day  4 days was US $ 6500, where as the cost of world: experience from a center in India. Bone Marrow Thymogam at a dose of 40 mg/kg/day was US $ 3000 only. Transplant 2001; 27: 785–790. Apart from this variation, the protocols were identical. 4 Kang HJ, Shin HY, Choi HS, Ahn HS. Fludarabine, Both neutrophil and platelet engraftments were rapid and cyclophosphamide plus thymoglobulin conditioning regimen sustained in all cases. This is especially significant, as all the for unrelated bone marrow transplantation in severe aplastic patients had multiple risk factors for graft rejection. Acute anemia. Bone Marrow Transplant 2004; 34: 939–943. GVHD did not occur in four patients and was mild in one. 5 Bacigalupo A, Locatelli F, Lanino E, Marsh J, Socie G, Maury Extensive chronic GVHD has occurred in patient 1, where S et al. Fludarabine, cyclophosphamide and anti-thymocyte as limited chronic GVHD has occurred in another. With globulin for alternative donor transplants in acquired severe the limited follow-up, it is not possible to draw any aplastic anemia: a report from the EBMT-SAA Working Party. Bone Marrow Transplant 2005; 36: 947–950. conclusions on occurrence and severity of chronic GVHD 6 Johnsen HE, Baech J, Nikolajsen K. Validation of the Nordic and any beneficial role of high doses of ATG that we flow cytometry standard for CD34+ cell enumeration in blood have used. and autografts: report from the third workshop. Nordic Stem We were concerned about a higher infection risk, Cell Laboratory Group. J Hematother 1999; 8: 15–28. especially of fungal infections, in our patients as we do 7 Storb R, Prentice RL, Thomas ED, Appelbaum FR, Deeg HJ, not have HEPA filter rooms in our center. Recently, the Doney K et al. Factors associated with graft rejection after protective value of specialized airway ventilation systems HLA-identical marrow transplantation for aplastic anaemia. and HEPA filter rooms has been questioned and in some Br J Haematol 1983; 55: 573–585. centers even allogeneic transplants have been made 8 Storb R, Prentice RL, Thomas ED. Marrow transplantation for treatment of aplastic anemia. An analysis of factors successfully without protective isolation.20,21 There are no associated with graft rejection. N Engl J Med 1977; 296: 61–66. such data from developing countries where environmental 9 Gluckman E, Horowitz MM, Champlin RE, Hows JM, pollution is higher, and HEPA filter rooms are considered Bacigalupo A, Biggs JC et al. Bone marrow transplantation 3 advisable. Despite a very immunosuppressive conditioning for severe aplastic anemia: influence of conditioning and graft- in our high-risk patients, and routine protective isolation, versus-host disease prophylaxis regimens on outcome. Blood we had no refractory or life-threatening infections. 1992; 79: 269–275.

Bone Marrow Transplantation Fludarabine for transplant in aplastic anemia R Kumar et al 749 10 Ades L, Mary JY, Robin M, Ferry C, Porcher R, Esperou H 16 George B, Mathews V, Shaji RV, Srivastava V, Srivastava A, et al. Long-term outcome after bone marrow transplantation Chandy M. Fludarabine-based conditioning for allogeneic for severe aplastic anemia. Blood 2004; 103: 2490–2497. stem cell transplantation for multiply transfused patients 11 Champlin RE, Horowitz MM, van Bekkum DW, Camitta with Fanconi’s anemia. Bone Marrow Transplant 2005; 35: BM, Elfenbein GE, Gale RP et al. Graft failure following bone 341–343. marrow transplantation for severe aplastic anemia: risk factors 17 Stem Cell Trialists’ Collaborative Group. Allogeneic peri- and treatment results. Blood 1989; 73: 606–613. pheral blood stem-cell compared with bone marrow trans- 12 Storb R, Etzioni R, Anasetti C, Appelbaum FR, Buckner CD, plantation in the management of hematologic malignancies: Bensinger W et al. Cyclophosphamide combined with anti- an individual patient data meta-analysis of nine randomized thymocyte globulin in preparation for allogeneic marrow trials. J Clin Oncol 2005; 23: 5074–5087. transplants in patients with aplastic anemia. Blood 1994; 84: 18 Korbling M, Anderlini P. Peripheral blood stem cell versus 941–949. bone marrow allotransplantation: does the source of hemato- 13 Marsh JC, Ball SE, Darbyshire P, Gordon-Smith EC, Keidan poietic stem cells matter? Blood 2001; 98: 2900–2908. AJ, Martin A et al. Guidelines for the diagnosis and 19 Bacigalupo A. Antilymphocyte/thymocyte globulin for graft management of acquired aplastic anaemia. Br J Haematol versus host disease prophylaxis: efficacy and side effects. 2003; 123: 782–801. Bone Marrow Transplant 2005; 35: 225–231. 14 Kahl C, Leisenring W, Deeg HJ, Chauncey TR, Flowers ME, 20 Hayes-Lattin B, Leis JF, Maziarz RT. Isolation in the Martin PJ et al. Cyclophosphamide and antithymocyte allogeneic transplant environment: how protective is it? globulin as a conditioning regimen for allogeneic marrow Bone Marrow Transplant 2005; 36: 373–381. transplantation in patients with aplastic anaemia: a long-term 21 Russell JA, Chaudhry A, Booth K, Brown C, Woodman RC, follow-up. Br J Haematol 2005; 130: 747–751. Valentine K et al. Early outcomes after allogeneic stem cell 15 Chan KW, Li CK, Worth LL, Chik KW, Jeha S, Shing MK transplantation for leukemia and myelodysplasia without et al. A fludarabine-based conditioning regimen for severe protective isolation: a 10-year experience. Biol Blood Marrow aplastic anemia. Bone Marrow Transplant 2001; 27: 125–128. Transplant 2000; 6: 109–114.

Bone Marrow Transplantation