Cyclophosphamide and antithymocyte globulin ... - Wiley Online Library

short report

Cyclophosphamide and antithymocyte globulin as a conditioning regimen for allogeneic marrow transplantation in patients with aplastic anaemia: a long-term follow-up

Christoph Kahl,1 Wendy Leisenring,1,2 H. Joachim Deeg,1,3 Thomas R. Chauncey,1,3 Mary E. D. Flowers,1,3 Paul J. Martin,1,3 Jean E. Sanders1,3 and Rainer Storb1,3 1

Clinical Research Division, Fred Hutchinson

Cancer Research Center, and Departments of 2

Biostatistics and 3Medicine, University of

Washington School of Medicine, Seattle, WA, USA

Received 11 May 2005; accepted for publication 14 June 2005 Correspondence: Rainer Storb, MD, Clinical Research Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave N, Seattle, WA 98109, USA. E-mail: [email protected]

Summary A total of 81 severe aplastic anaemia patients, aged 2–63 years, received human leucocyte antigen-matched related marrow grafts after cyclophosphamide + antithymocyte globulin followed by postgrafting methotrexate + ciclosporin. Median follow-up was 9Æ2 years. Ninety-six per cent of patients had sustained engraftment, 24% developed acute graft-versushost disease (GVHD), grade II in all but two patients, and 26% developed chronic GVHD; all surviving patients eventually responded to immunosuppressive therapy. Six patients developed cancer: one fatal lymphoma and five carcinomas (all five patients are now free of cancer). Survival was 88%. The regimen appeared well tolerated and effective in heavily pretreated patients with aplastic anaemia. Keywords: aplastic anaemia, human leucocyte antigen-matched marrow grafts, cyclophosphamide/antithymocyte globulin conditioning.

Based on canine studies, cyclophosphamide (CY) and antithymocyte globulin (ATG) were first explored to condition patients with aplastic anaemia for second allogeneic marrow transplantation after rejection of their first grafts. Given encouraging results, the regimen has been used to condition patients for first grafts from human leucocyte antigen (HLA)matched related donors since 1988. In 1994, we reported outcomes in 39 patients given CY/ATG and treated with methotrexate (MTX) and ciclosporin (CSP) for the prevention of graft-versus-host disease (GVHD) (Storb et al, 1994). Only 5% of patients rejected their grafts, and 36 (92%) survived. Storb et al (2001) described the combined experience in Seattle, Stanford and City of Hope National Medical Centre with the regimen, which yielded results similar to those originally reported (Storb et al, 2001). The present report updates the early Seattle results and describes 42 additional Seattle patients given HLA-matched related marrow grafts with this regimen. Apart from early events, such as acute GVHD and graft rejection, we describe late effects, including chronic GVHD, secondary malignancies and Karnofsky performance scores.

Materials and methods Between 13 July 1988 and 1 June 2004, 81 patients received marrow transplants from HLA-matched family members after

CY/ATG. Table Ia lists the patient characteristics. CY, 50 mg/ kg/d i.v. was given for 4 d. Approximately 12 h after each of the first three CY doses, horse anti-human ATG, (ATGAM; Pharmacia-Upjohn, Kalamazoo, MI, USA) 30 mg/kg/d i.v., was administered. Donor marrow was infused i.v. 36 h after the last CY infusion, ranging from 0Æ94 to 10Æ75 (median 2Æ79) · 108 nucleated cells/kg recipient body weight (corrected for peripheral white blood cell counts). Dose of CD34 cells in 17 patients ranged from 1Æ1 to 9Æ6 (median 3Æ7) · 106 cells/kg. The patients received MTX, 15 mg/m2 on day 1 and 10 mg/ m2/day on days 3, 6 and 11, and CSP, beginning a day before transplant at 1Æ5 mg/kg i.v. every 12 h until recovery from gastrointestinal toxicity, when it was given at 6Æ25 mg/kg orally every 12 h until day 50, and then tapered until day 180 (Storb et al, 1994).

Results and discussion Three patients (3Æ7%) rejected their grafts on days 21, 39, and 198 respectively (Fig 1A). Two of the three patients are alive following reconditioning and marrow grafting from original donors, while one patient died from infection 2 days after the second transplantation. All other patients had sustained first grafts. Donor chimaerism was assessed in 35 gender mismatched recipients by fluorescent in situ hybridisation, using

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 130, 747–751

doi:10.1111/j.1365-2141.2005.05667.x

Short Report

Patients (n) Median age, year (range) Sex (f/m), n Median granulocyte count before transplant, cells · 109/l (range)* Median platelet count before transplant, cells · 109/l (range)* Possible causes of aplastic anaemia (n) Unknown Hepatitis Pregnancy Drugs or chemicals Dyskeratosis congenita Fanconi anaemia Paroxysmal nocturnal hemoglobinuria Transfused/untransfused patients (n) Panel reactive (anti-human leucocyte antigen) antibodies (n) Positive Negative Unknown Prior treatment (n) Antithymocyte globulin Ciclosporin Steroids Androgens Granulocyte or granulocyte macrophage colony-stimulating factors Immunoglobulin Vincristine Median duration of aplastic anaemia, months (range)

81 25Æ1 (2–63) 34/47 390 (7–9345) 23 000 (1000–150 000)

Table Ia. Patient characteristics.

63 8 4 3 1 1 1 76/5 15 58 8 14 14 16 11 12 6 1 2Æ3 (0Æ2–146Æ6)

*Exact numbers available in 76 of the 81 patients. Many patients had received transfusions before arriving at the transplant centre.

Table Ib. Multivariate analysis* of factors predicting for chronic GVHD. Covariate

Hazard ratio

Age (years) £18 1 19–37 1Æ6 ‡38 3Æ9 Marrow cell dose (· 108/kg) £2Æ3 1 2Æ4–3Æ3 3Æ8 ‡3Æ4 7Æ7 Sex Female 1 Male 1Æ5

P-value

95% CI

0Æ40 0Æ04

0Æ5, 4Æ8 1Æ1, 14Æ1

0Æ06 0Æ004

1Æ0, 14Æ8 2Æ0, 30Æ6

0Æ37

0Æ6, 3Æ8

*Cox proportional hazard model. Tertiles. Baseline category.

probes for X and Y chromosomes, and in the remaining recipients by informative probes for variable number tandem repeat polymorphisms. Chimaerism studies were performed in peripheral blood granulocytes and T cells and marrow. Studies were conducted at around day 84, 1 year, and irregular intervals thereafter. Upon latest testing, donor contributions in 748

patients with sustained first or second grafts ranged from 95% to 100% (median 99). The probability of developing acute GVHD was 24% (Fig 1B), which was grade II for 21Æ6% and grade IV for 2Æ4% of patients. The cumulative incidence of chronic GVHD, requiring extended immunosuppression, was 26% (not shown); the prevalence curve in Fig 1C describes the times of chronic GVHD onset, its resolution, and eventual discontinuation of all immunosuppressive therapy, using previously described methodology (Pepe et al, 1991). Risk factor analyses for acute and chronic GVHD used Cox proportional hazard models to evaluate the impact of patient age, numbers of nucleated marrow cells infused, and patient gender. None of these factors were significant when acute GVHD was the outcome, consistent with previous findings (Storb et al, 1983a). However, both age at time of haemopoietic cell transplantation and marrow cell doses appeared to have significant impacts on chronic GVHD (Table Ib). Being in the upper tertile for age (‡38 years) significantly increased the risk of chronic GVHD compared with patients in the lowest tertile (£18 years). This finding has also been described before (Storb et al, 1983b). An unexpected result was the increased risk of chronic GVHD with increasing numbers of nucleated marrow cells transplanted – a correlation between CD34 cell dose and outcomes could not be established, as


Short Report

Fig 1. Human leucocyte antigen-matched related marrow grafts for treatment of 81 patients with aplastic anaemia conditioned by cyclophosphamide and antithymocyte globulin and given postgrafting immunosuppression by methotrexate and ciclosporin. (A) Probability of graft rejection. (B) Probability of acute graft-versus-host disease (GVHD) grades II–IV. (C) Prevalence of chronic GVHD. (D) Probability of cancer. (E) Probability of survival.

CD34 cell counts were only available for a minority of patients. Specifically, the hazard ratio for developing chronic GVHD was 3Æ8 when 2Æ4–3Æ3 · 108 cells/kg were infused compared with £2Æ3 · 108 cells/kg; a further increase in the hazard ratio to 7Æ7 occurred with marrow doses ‡3Æ4 · 108 cells/kg. This correlation had not been previously reported (Storb et al, 1983b; Bittencourt et al, 2002), and it suggested that the current recommendation of collecting as many marrow cells as possible might need revision. That recommendation was based on a previously observed correlation between marrow cell doses and graft rejection (Storb et al, 1977). Given that graft rejection has become the exception (current incidence 3Æ7%) owing to a more effective conditioning regimen and the use of leucocyte-poor and irradiated blood products before transplantation, and given the current findings, transplanting lower marrow cell numbers might be desirable in order to reduce morbidity from chronic GVHD and its treatment. Six patients developed cancer. One was fatal Epstein–Barr virus-associated lymphoproliferative syndrome in donor cells (d 113) after steroid-resistant acute GVHD and treatment with an anti-CD3 monoclonal antibody and rabbit ATG (Sangstat Medical Corp., Menlo Park, CA, USA). The syndrome has been previously described in patients given T-depleted grafts or anti-T-cell therapy for acute GVHD (Bhatia & Bhatia, 2004; Curtis et al, 2005). At last follow-up, the remaining five patients were alive and free of cancer; they had adenocarcinoma of the sigmoid (4Æ9 years after transplant; resection and chemotherapy with 5-fluorouracil), basal cell skin carcinomas (5Æ4, 9Æ1 and 14Æ5 years after transplant respectively; resection)

and verrucous oral carcinoma (14Æ9 years; resection). The cancer rate was comparable with that reported after CY alone (Socie´ et al, 1993) and was higher among patients with chronic GVHD (n ¼ 3; 17%), even though immunosuppression had been discontinued, than among those without chronic GVHD (n ¼ 2; 3%). This association was known from previous reports (Bhatia & Bhatia, 2004). At the time of writing, 71 of the 81 patients were alive with a median follow-up of 9Æ2 (range, 0Æ5–16Æ4) years (Fig 1E). The survivors included 36 of the initially reported 39 patients. Most survivors (78%) had Karnofsky performance scores of 100, 13% had 90, and 9% had £80. Seven patients, all of whom had chronic GVHD, experienced avascular joint necroses; of the seven, four had received steroid treatment for their aplastic anaemia before transplantation. Nine female patients and seven partners of male patients reported 17 successful pregnancies, and one patient is currently pregnant. Additionally, five elective abortions and seven miscarriages were reported. All but one of the latter patients eventually had successful childbirths. These observations were consistent with previous reports in patients with aplastic anaemia conditioned with 200 mg CY/kg (Sanders et al, 1996). Ten of the 81 patients died, three from pre-existing complications, one on day 2 (Aspergillus pneumoniae), another on day 6 (generalised candidiasis), and another on day 70 (interstitial pulmonary fibrosis resulting from dyskeratosis congenita). Three patients died from infections (n ¼ 2) or lymphoproliferative syndrome (n ¼ 1) related to acute or


749

Short Report chronic GVHD and their treatment (days 90, 113 and 179 respectively), one died from Escherichia coli septicaemia following graft rejection (day 199), two from infections unrelated to GVHD (day 83, respiratory syncytial virus pneumonia; day 223, bacterial septicaemia), and another from idiopathic interstitial pneumonia (day 210). A multivariate analysis with death as the outcome showed increasing age as a risk factor (P ¼ 0Æ03), and there was a suggestion that higher marrow cell doses also increased the risk of death, although this was not significant (data not shown). Overall, CY/ATG was well tolerated in patients with an age range extending to 63 years. It provided effective immunosuppression, as indicated by the low incidence of graft rejection, even though most patients had multiple preceding transfusions. While adding ATG to CY was likely important in lessening rejection over historical controls, prevention of sensitisation to non-HLA antigens by use of leucocyte-poor and irradiated blood products might also have contributed. While the survival of 88% is encouraging, there is room for improvement. Excluding the three early deaths from preexisting infectious or genetic (dyskeratosis congenita) causes, approximately 9% of patients died from transplant-related causes. Two patients died as a direct consequence of acute GVHD, emphasising a need for better GVHD prevention; the 24% cumulative incidence of acute GVHD observed in the present patients given ATG before and MTX/CSP after haemopoietic cell transplantation was comparable with the 20% incidence seen with MTX/CSP and no pretransplant ATG (Storb et al, 1994). Therefore, if in vivo donor T-cell depletion through horse ATG occurred, it is doubtful that it had much benefit on acute GVHD. Granulocyte colony-stimulating factor-mobilised peripheral blood mononuclear cells (G-PBMC) have been increasingly used for allografts in patients with hematological malignancies [reviewed in (Schmitz, 2004)]. However, consistent with previous observations in patients with aplastic anaemia given combined marrow and PBMC grafts (Storb et al, 1994), many reports described higher incidences of chronic GVHD among G-PBMC compared with marrow recipients reviewed in Schmitz (2004). Current results suggest marrow is the preferred source of HLA-matched related stem cells in aplastic anaemia patients, given the low incidence of chronic GVHD and its almost uniform eventual response to therapy. As already discussed, lowering the marrow cell dose to approximately 2Æ5 · 108 cells/kg might further reduce the risk of chronic GVHD.

Acknowledgements The authors wish to thank the transplant and long-term follow-up teams, the referring physicians, and the database managers, particularly Gary Schoch, for their invaluable help with the study, and Bonnie Larson, Helen Crawford and Sue Carbonneau for help with manuscript preparation. 750

This work was supported by grants HL36444 and CA15704, National Institutes of Health, Bethesda, MD, USA. C.K. is a recipient of a fellowship from Deutsche Krebshilfe, Dr. Mildred-Scheel-Stiftung fu¨r Krebsforschung.

References Bhatia, S. & Bhatia, R. (2004) Secondary malignancies after hematopoietic cell transplantation. In: Thomas’ Hematopoietic Cell Transplantation (ed. by K.G. Blume, S.J. Forman & F.R. Appelbaum), p. 962–977. Blackwell Publishing Ltd., Malden, MA. Bittencourt, H., Rocha, V., Chevret, S., Socie, G., Esperou, H., Devergie, A., Dal Cortivo, L., Marolleau, J.P., Garnier, F., Ribaud, P. & Gluckman, E. (2002) Association of CD34 cell dose with hematopoietic recovery, infections, and other outcomes after HLAidentical sibling bone marrow transplantation. Blood, 99, 2726– 2733. Curtis, R.E., Metayer, C., Rizzo, J.D., Socie´, G., Sobocinski, K.A., Flowers, M.E.D., Travis, W.D., Travis, L.B., Horowitz, M.M. & Deeg, H.J. (2005) Impact of chronic GVHD therapy on the development of squamous-cell cancers after hematopoietic stem-cell transplantation: an international case-control study. Blood, 105, 3802–3811. Pepe, M.S., Longton, G. & Thornquist, M. (1991) A qualifier Q for the survival function to describe the prevalence of a transient condition. Statistics in Medicine, 10, 413–421. Sanders, J.E., Hawley, J., Levy, W., Gooley, T., Buckner, C.D., Deeg, H.J., Doney, K., Storb, R., Sullivan, K., Witherspoon, R. & Appelbaum, F.R. (1996) Pregnancies following high-dose cyclophosphamide with or without high-dose busulfan or total-body irradiation and bone marrow transplantation. Blood, 87, 3045–3052. Schmitz, N. (2004) Peripheral blood hematopoietic cells for allogeneic transplantation. In: Thomas’ Hematopoietic Cell Transplantation (ed. by K.G. Blume, S.J. Forman & F.R. Appelbaum), p. 588–598. Blackwell Publishing Ltd., Malden, MA. Socie´, G., Henry-Amar, M., Bacigalupo, A., Hows, J., Tichelli, A., Ljungman, P., McCann, S.R., Frickhofen, N., Van’t Veer-Korthof, E., Gluckman, E. & for the European Bone Marrow TransplantationSevere Aplastic Anaemia Working Party (1993) Malignant tumors occurring after treatment of aplastic anemia. New England Journal of Medicine, 329, 1152–1157. Storb, R., Prentice, R.L. & Thomas, E.D. (1977) Marrow transplantation for treatment of aplastic anemia. An analysis of factors associated with graft rejection. New England Journal of Medicine, 296, 61–66. Storb, R., Prentice, R.L., Buckner, C.D., Clift, R.A., Appelbaum, F., Deeg, J., Doney, K., Hansen, J.A., Mason, M., Sanders, J.E., Singer, J., Sullivan, K.M., Witherspoon, R.P. & Thomas, E.D. (1983a) Graftversus-host disease and survival in patients with aplastic anemia treated by marrow grafts from HLA-identical siblings. Beneficial effect of a protective environment. New England Journal of Medicine, 308, 302–307. Storb, R., Prentice, R.L., Sullivan, K.M., Shulman, H.M., Deeg, H.J., Doney, K.C., Buckner, C.D., Clift, R.A., Witherspoon, R.P., Appelbaum, F.R., Sanders, J.E., Stewart, P.S. & Thomas, E.D. (1983b) Predictive factors in chronic graft-versus-host disease in patients with aplastic anemia treated by marrow transplantation from HLAidentical siblings. Annals of Internal Medicine, 98, 461–466. Storb, R., Etzioni, R., Anasetti, C., Appelbaum, F.R., Buckner, C.D., Bensinger, W., Bryant, E., Clift, R., Deeg, H.J., Doney, K., Flowers, M., Hansen, J., Martin, P., Pepe, M., Sale, G., Sanders, J., Singer, J.,


Short Report Sullivan, K.M., Thomas, E.D. & Witherspoon, R.P. (1994) Cyclophosphamide combined with antithymocyte globulin in preparation for allogeneic marrow transplants in patients with aplastic anemia. Blood, 84, 941–949. Storb, R., Blume, K.G., O’Donnell, M.R., Chauncey, T., Forman, S.J., Deeg, H.J., Hu, W.W., Appelbaum, F.R., Doney, K., Flowers,

M.E.D., Sanders, J. & Leisenring, W. (2001) Cyclophosphamide and antithymocyte globulin to condition patients with aplastic anemia for allogeneic marrow transplantations: the experience in four centers. Biology of Blood and Marrow Transplantation, 7, 39–44.


751