¹Research Center for Hematology, The General Hospital of Air Force, PLA, Beijing, PR China

²Division of Hematology/Oncology and Stem Cell Transplant Program, Department of Medicine, University of Kentucky, Lexington, KT, USA

Correspondence to: Dr S-Q Ji, Research Center for Hematology, The General Hospital of Air Force, PLA, Beijing100036, PR China or Dr C-Q Xun, CC301 Markey Cancer Research Center, 800 Rose Street, Lexington, Kentucky 40536, USA

Based on our encouraging results of G-CSF-primed HLA-matched related marrow transplants for high-risk leukemia, we extended the study from matched related to haploidentical transplants using G-CSF primed marrow and sequential immunosuppressants to prevent both graft-versus-host disease (GVHD) and host-versus-graft rejection (HVGR). Fifteen high-risk leukemia patients, who needed urgent transplantation but lacked an HLA-matched donor, underwent G-CSF-primed haploidentical marrow transplantation without ex vivo T cell depletion. Donors were given G-CSF (Lenograstim) at 3-4 g/kg/day for 7 days prior to marrow harvest. GVHD and HVGR prophylaxis were combined in the sequential usage of cyclosporin A, methotrexate, anti-thymocyte globulin and mycophenolate mofetil. All patients established sustained trilineage engraftment at a median of 19 days and 21 days for neutrophil and platelets respectively. G-CSF priming significantly increased CD34⁺ and CFU-GM cells, reduced total lymphocytes and reversed the CD4⁺/CD8⁺ ratio in the donor marrow. The incidence of grade II-IV acute GVHD was 33.3%. Nine patients survived more than a year with a Karnofsky performance status of 100%. Estimated overall disease-free survival at 2 years was 60 ± 7%. In conclusion, using G-CSF priming marrow grafts along with sequential immunosuppressants provided an excellent alternative for the treatment of high-risk hematological malignancy in patients who lack matched donors.

Bone Marrow Transplantation (2002) 30, 861-866. doi:10.1038/sj.bmt.1703769

G-CSF; GVHD; engraftment; haploidentical bone marrow transplantation

Transplantation of unmodified bone marrow (BM) from HLA-haploidentical two- or three-loci incompatible donors has been associated with an unsuccessful outcome caused by the high incidence of severe graft-versus-host disease (GVHD) and graft failure.^1,2 While vigorous pan-T cell depleted haploidentical grafting is the most effective means of preventing severe GVHD, it has been associated with delayed recovery of T cell function leading to fatal infection, graft rejection, and leukemia relapse.^3,4,5 It is clear that one of the most difficult issues in the clinical performance of haploidentical transplantation has been attempts to balance the necessary degree of T cell depletion and immunosuppression afforded by the conditioning in order to prevent both graft-versus-host disease (GVHD) and host-versus-graft rejection (HVGR).

Based on the fact that granulocyte colony-stimulating factor (G-CSF) facilitates engraftment, G-CSF-primed marrow grafts have 10 times fewer lymphocytes compared to peripheral blood stem cell grafts, and both GVHD and HGVR are immune-mediated reactions, we undertook HLA match-related G-CSF primed marrow transplants for high-risk leukemia.^6,7 Our transplant results were very encouraging, with a surprisingly low incidence of acute GVHD at only 6.3%. All patients achieved sustained trilineage engraftment. We sought to extend the study to the haploidentical transplant setting with the addition of immunosuppressants to prevent both GVHD and HVGR. Our strategies were to use G-SCF-primed marrow to facilitate engraftment without increasing the T cells in the grafts, different from G-CSF-mobilized peripheral blood stem cells (PBSC). We added anti-thymocyte globulin (ATG) and mycophenolate mofetil to our baseline cyclosporin A (CsA)/methotrexate (MTX) GVHD prophylaxis. Here, we report our transplant protocol and the sequential results in 15 patients with high-risk leukemia without matched donors, who underwent the HLA-haploidentical G-CSF-primed marrow transplantation without ex vivo T cell depletion.

Patients and methods

Patients with advanced disease or leukemia with poor prognostic features that needed urgent transplantation were eligible for this trial. Additional criteria for eligibility included the presence of a suitable related haploidentical donor, absence of life-threatening infection, and near normal cardiac, hepatic, pulmonary and renal function. From February 1999 to November 2000, 15 consecutive patients with high-risk leukemia were enrolled into the study. None of these patients had an HLA-matched sibling or single-antigen mismatched related donor. The protocol used for this study was approved by our institutional review board. Written informed consent was obtained from all the patients, donors and/or their guardians. Clinical characteristics of the 15 transplant patients and donor-host relationships are summarized in Table 1.

HLA typing and donor selection

Patients and donors were HLA-typed according to standard techniques. Typing for class I HLA-A and B antigens was performed using serologic methods. HLA class II typing for DRB1 was carried out by PCR-SSP high resolution methods. All pairs of donors and recipients were identical for one HLA haplotype and incompatible at two or three loci (HLA-A, B and DR) of the unshared haplotype. Donor-host relationship and HLA typing are listed in Table 1. The donors were parents in 11 cases, siblings in three cases and a cousin in one case. Donors were prioritized on the basis of the greatest HLA matching A, B or DR loci, cytomegalovirus (CMV) screen negativity, younger age, same sex, and better health.

Conditioning regimen

All patients received the same conditioning regimens consisting of cytarabine (Cytosar; Pharmacia and Upjohn, Puurs, Belgium) 3.0 g/m², twice a day for 3 consecutive days, on days -7, -6 and -5; cyclophosphamide (CY) 45 mg/kg/day for 2 consecutive days on days -5 and -4; total body irradiation (TBI) at 1000 cGy in two fractions on days -2 and -1, using a high energy linear accelerator, the midline dose rate being 5-6 cGy/min; and rabbit antithymocyte globulin (ATG) (Fresenius AG, Oberursel, Germany) at 5 mg/kg on day -4, -3, -2 and -1.

Donor priming regimen and bone marrow harvest

Donors received G-CSF (Lenograstim; Chugai Pharmaceutical, Tokyo, Japan) at 3-4 g/kg/day daily for 7 consecutive days as described before.^6,7 Marrow cells were collected on the 8th day under epidural anesthesia, from the posterior iliac crests with a target marrow volume of 18-20 ml/kg recipient body weight. Total nucleated cell counts (TNC) were obtained using an automated counter. The number of CD34⁺ and T subsets (CD3⁺, CD4⁺ and CD8⁺ cells) was assessed by flow cytometry. Fresh and unmanipulated marrow was infused on the same day (day 0). With ABO major blood group incompatibility, red cells in the marrow were removed by sediment manipulation.

GVHD prophylaxis

A combination of CsA, MTX and MMF was used for GVHD prophylaxis at different stages of the transplant course. CsA was given at 1.5 mg/kg/day i.v. on day -7 to -1, then 3 mg/kg/day continuous 24 h infusion from day -1 until gut function recovered, then changed to 6 mg/kg/day orally in three divided doses. CsA levels were measured weekly by a fluorescence polarization immunoassay and the dose adjusted accordingly to maintain serum levels at approximately 200 ± 50 ng/ml. CsA dose was reduced if the trough level of CsA was over 300 ng/ml or the serum creatinine level exceeded 2 mg/dl. MTX was given at 15 mg/m² on day +1 and 10 mg/m² on day +3, +6 and +11. MMF, 1.0 g/day was administered from day +7 to +100. Acute and chronic GVHD were graded and staged according to the consensus conference on GVHD grading.⁸ Acute grade I GVHD or more than grade I was treated with methylprednisolone 1-2 mg/kg/day for 2 weeks, tapered as soon as possible.

Supportive care

Transplanted patients were hospitalized in a room with a high-efficiency particulate air filter until the neutrophil count recovered to 1 ´ 10⁹/l. All patients received prophylactic trimethoprim-sulfamethoxazole for Pneumocystis carinii, ciprofloxacin for selective gut decontamination and fluconazole for fungal prophylaxis. Herpes simplex prophylaxis consisted of acyclovir. Fever during the period of neutropenia was treated with broad-spectrum antibiotics. Intravenous immunoglobulins were given at a dose of 500 mg/kg weekly starting at day 1 post transplantation. G-CSF 3-4 g/kg/day was administered from the second day of transplant until the neutrophil count reached 0.5 ´ 10⁹/l for 3 consecutive days. Patients were transfused if their hemoglobin or their platelets dropped below 80 g/l or 20 ´ 10⁹/l, respectively. All blood products were irradiated.

Evaluation of engraftment

Daily complete blood counts (CBC) were taken. Neutrophil engraftment was defined as the first day of ANC >0.5 ´ 10⁹/l after the neutrophil nadir. Platelet engraftment was defined as the day when the platelet count exceeded 20 ´ 10⁹/l without previous platelet transfusions for at least 7 days. A bone marrow aspirate and biopsy was done routinely at 1 month post-transplant to assess engraftment. Donor engraftment and chimerism were evaluated by either cytogenetic (sex mismatch) or DNA analysis (same sex) of one or more genetic differences between the donor and recipient.

Flow cytometry and progenitor cell analysis

Cell surface markers were determined by dual or three-color staining with monoclonal antibody directly conjugated to FITC, PE and CY5 against CD34⁺, CD3⁺, CD4⁺, CD8⁺ (Coulter Immunology, Hialeah, FL, USA). In this study, 100 l of cell solution aliquots were incubated at room temperature with a saturating concentration of two different monoclonal antibodies for 15 min. After incubation, the red blood cells were lysed and the stained cells were fixed and analyzed by an EPICS XL flow cytometer (Coulter Immunology). The number of CD34⁺ cells and T cell subsets was determined by multiplying the numbers of lymphocytes per microliter by the percentage of nucleated cells with that phenotype. CFU-GM was determined by semi-solid agar culture.⁹

Statistical analysis

Median values and ranges for each parameter were determined. The Wilcoxon rank sum test was performed to compare results obtained before and after administration of G-CSF. Cumulative incidence estimates were used for GVHD. Survival was estimated according to the method of Kaplan-Meier. The date of the final analysis was 15 December, 2001.

Results

Effect of G-CSF priming on marrow contents and graft characteristics

Table 2 summarizes graft contents that were collected from the donors primed with G-CSF at the target volume of 18-20 ml/kg recipient weight. The median volume was 1.1 liters (range 0.7-1.5). Table 3 compares the CD34⁺, CD3⁺, CD4⁺ and CD8⁺, and the proliferation of CFU-GM in the bone marrow before and after administration of G-CSF. G-CSF priming significantly increased TNCs, CD34⁺ cells (P = 0.02) and CFU-GM cells (P = 0.001) and decreased total lymphocyte percentage in the donor marrow. Although the percentage of the CD3⁺ subset was not changed significantly, the ratio of CD4⁺/CD8⁺ was decreased (P = 0.04) after G-CSF priming.

Engraftment

All patients experienced successful trilineage engraftment. The median time to a neutrophil engraftment was 19 days (range 13-22 days). The median time to platelet engraftment was 21 days (range 16-32 days) (Table 2). Chimerism analyses demonstrated complete donor chimerism (>95% donor cells) in all patients and sustained engraftment at a median follow-up of 22 months (13-35 months).

GVHD

Eight of 15 patients (60%) developed grade I acute GVHD (aGVHD). Five patients developed grade II-IV aGVHD at a cumulative incidence of 33 ± 8%. The onset of aGVHD occurred at a median of day 21 (range, 10-36). Three patients with grade III-IV aGVHD failed the treatment and died as a direct consequence of aGVHD (Table 4). The other two patients responded to steroid treatment. Nine patients who survived more than 6 months had limited chronic GVHD (cGVHD) with oral dryness and skin involvement. None of them developed extensive cGVHD and there were no cGVHD related deaths.

Outcome

Nine of 15 transplanted patients are currently alive at a median follow-up of 22 months (range 13-35 months) with 100% Karnofsky performance. The estimated overall disease-free survival rate at 2 years was 60 ± 7% (Figure 1). Of the six deaths in this series, three patients (patients 2, 8 and 12) died of severe aGVHD and two (patients 3 and 10) died of infection. One patient who was transplanted in relapse died of leukemia relapse 3 months after the transplant (Table 4).

Discussion

Haploidentical hematopoietic stem cell transplantation (HSCT) is a good solution for patients who need allogeneic HSCT but do not have a matched donor.¹⁰ However, problems with the haploidentical HSCT have included a high incidence of severe GVHD and graft rejection/failure.^1,2 Ex vivo T cell depletion reduced the risk of GVHD but was associated with delayed recovery of T cell function leading to fatal infection, graft rejection, and leukemia relapse.^3,4,5,11 In our study, 15 high-risk leukemia patients who did not have an HLA-matched donor received haploidentical marrow transplants with G-CSF donor priming, without ex vivo T cell depletion. All recipients achieved sustained trilineage engraftment. G-CSF significantly increased CD34⁺ and CFU-GM cells in the marrow grafts and facilitated engraftment. We used 7-day low-dose G-CSF priming based on our previous experience in the match related marrow transplants.^6,7 We compared G-CSF-primed marrow with the steady state marrow transplant in HLA-matched related transplants in leukemia patients and CML patients. Both studies demonstrated that G-CSF-primed marrow transplants had significantly faster engraftment and a lower incidence of acute GVHD. The incidence of acute GVHD was surprisingly low; only 6.3% in G-CSF-primed marrow transplants compared to 33% in steady state marrow transplants. Three other groups also demonstrated that G-CSF donor marrow priming facilitated marrow engraftment.^12,13,14 G-CSF doses used for donor-marrow priming in the three groups were different. The Isola group used filgrastim 10 g/kg/day for 2 days. The Serody group used filgrastim 10 g/kg/day for 4 days. The Couban group used 10-15 g/kg for 4 days. Neutrophil and platelet engraftment were very similar among the four groups. We used post-graft G-CSF treatment until the ANC >0.5 ´ 10⁹/l and the other study groups did not use post-graft G-CSF. It appeared that post-grafting G-CSF did not further facilitate engraftment if the donor marrow had already been exposed to G-CSF. The question that remains to be answered is what is the optimal dose of G-CSF for donor marrow priming?

Besinger et al¹⁵ compared G-CSF-mobilized PBSCs to historical steady-state marrow transplants and reported a similar incidence of GVHD although T cell numbers in G-CSF mobilized PBPCs were often 10 times higher than in marrow. The Serody and Darrant groups compared the G-CSF-mobilized PBSCs to G-CSF-primed marrow for incidences of acute and chronic GVHD.^12,16 They both found that G-CSF-primed marrow grafts had lower incidences of both acute and chronic GVHD, especially chronic GVHD. One of the explanations was that the number of T cells in the G-CSF-primed marrow was much lower than that in G-CSF-mobilized PBSCs. The incidence of grade II-IV aGVHD in our previous study was surprisingly low (6.3%) in the patients who received G-CSF-primed marrow grafts compared to the steady state marrow transplants, suggesting that other factors besides the number of T cells were also important in controlling aGVHD. Our results demonstrated that G-CSF did not change the total number of CD3⁺ T cells, but altered subsets of T cells and significantly lowered the CD4/CD8 ratio. It is known that cytokines produced by CD4⁺ and CD8⁺ T cells can be characterized into two patterns, Th1 (IL-2 and interferon-gamma) and Th2 responses (IL4 and IL10).¹⁷ Th-1 type responses are critical for acute GVHD.¹⁸ Treatment that induces Th2 response reduces GVHD.¹⁹ G-CSF polarizes T cell differentiation from Th1 to Th2 type cells and induces Th2 responses, with the production of IL-4 and IL-10.²⁰ Tayebi et al²¹ compared phenotypic and functional properties of lymphocytes from bone marrow or PBSC donors after G-CSF treatment in a randomized study. They found that not only were lymphocyte counts in the marrow grafts 10-fold lower than those in the G-CSF-mobilized PBSC grafts, but the production of Th1 cytokines (IL-2, IFN-gamma, tumor necrosis factor) after G-CSF treatment was also severely impaired.²¹ Thus, modulation of cytokine production by G-CSF may help to explain the surprisingly low incidence of clinical acute GVHD observed in our study group with the combination of donor G-CSF marrow priming and post-grafting G-CSF treatment.

In general, high-dose TBI, 1200 cGy or more, in attempting to decrease host resistance to donor grafts and eradicate leukemia/tumor, is an integral part of conditioning for haploidentical transplants. A large body of evidence has demonstrated that intensified chemo-radiation conditioning increased both regimen-related toxicity and more importantly, aGVHD.^22,23,24 Many cures of hematological malignancies have been attributed to immunological anti-leukemia reactions effected by allografting.^25,26 Mineishi et al²⁶ reported that addition of high-dose Ara-C to a CY/TBI conditioning regimen improved survival without increasing regimen-related toxicities. We added Ara-C to the conditioning regimen and lowered TBI to 1000 cGy. All patients in our study experienced successful trilineage engraftment after transplantation, taking a median of 19 days to neutrophil engraftment and there were no regimen-related deaths. Because the toxicity curve of TBI is not linear and accumulated toxicity increases dramatically at higher levels, a small reduction in TBI dose at a higher level markedly decreases accumulated toxicity. In our study, the lower dose 1000 cGy TBI reduced tissue injury and the risk of acute GVHD.

We used sequential immunosuppression to prevent both GVHD and HVGR in haploidentical BMT, without ex vivo T cell depletion. This is the first study to evaluate ATG and MMF in combination with standard CsA + MTX for GVHD prophylaxis and graft rejection. We used CsA and MTX as our baseline and added different immunosuppressants at different stages of the transplant course. We used ATG (half-life 2 weeks) on days -4 to -1 to suppress and/or deplete host lymphocytes to prevent host-versus-graft rejection, as well as to suppress/deplete the infused donor lymphocytes in vivo to prevent GVHD. We added MMF on day +7 to prevent rapid proliferation of donor lymphocytes at this stage. MMF, as a purine analogue, is a powerful inhibitor of lymphocyte proliferation. A combination of CsA + MMF showed a more potent immunosuppressive effect in preventing graft rejection than that of CsA + MTX.²⁸ Basara et al²⁹ demonstrated that MMF could be used safely and effectively for the prophylaxis of aGVHD in HLA-mismatched BMT patients. The combination of sequential immunosuppression allowed us to perform haploidentical BMT without ex vivo T cells. There was no graft rejection and the incidence of GVHD was comparable to the HLA-matched related steady state BMT.

CMV infection is often a major problem in allogeneic BMT, especially in patients who have received T cell depleted grafts or CD34⁺ selected grafts.¹⁵ In our study, we had no CMV infections. We used a CMV-PP56 antigen assay for both donor and recipient screening. If positive, donor or recipient was treated with a standard prophylactic dose of gancyclovior for 4 weeks. We think the low incidence of CMV infection in our study was more likely due to the marrow transplants being conducted without vigorous ex vivo T cell depletion. However, this will need more study, especially concerning immune reconstitution of T, B and NK cells after the transplantation. Our study involved mainly young patients, median age 15 years, and the number was relatively small. Nevertheless, the results with minimal regimen-related toxicity, no graft rejection and a low incidence of GVHD are very encouraging and certainly warrant a larger study with more, and older patients.

In summary, we describe 15 patients with high-risk leukemia who received HLA haploidentical G-CSF-primed marrow transplants without ex vivo T cell depletion. The high engraftment rate, low incidence of grade II-IV acute GVHD, minimal regimen-related toxicity and infection suggest that G-CSF-primed marrow grafting along with sequential immunosuppression could provide an excellent alternative for patients who need allogeneic marrow transplantation, but who do not have an HLA-matched donor.

1 Beatty PG, Glift RA, Mickelson EM et al. Marrow transplantation from related donor other than HLA-identical siblings. New Engl J Med 1985; 313: 765-771. MEDLINE

2 Anasetti C, Amos D, Beatty PG et al. Effect of HLA compatibility on engraftment of bone marrow transplants in patients with leukemia or lymphoma. New Engl J Med 1989; 320: 197-204. MEDLINE

3 Ringden O, Remberger M, Aschan J et al. Long-term follow-up of a randomized trial comparing T cell depletion with a combination of methotrexate and cyclosporine in adult leukemic marrow transplant recipients. Transplantation 1994; 58: 887-891.

4 Goldman JM, Gale PR, Horowitz MM et al. Bone marrow transplantation for chronic myelogenous leukemia in chronic phase: increased risk for relapse associated with T cell depletion. Ann Intern Med 1988; 108: 806-814. MEDLINE

5 Marks DI, Hughes TP, Szydlo R et al. HLA identical sibling donor bone marrow transplantation for chronic myeloid leukemia in the first chronic phase: influence of GVHD prophylaxis on outcome. Br J Haematol 1992; 81: 383-390. MEDLINE

6 Ji SQ, Chen HR, Xun CQ et al. The effect of G-CSF-stimulated donor marrow on engraftment and incidence of graft-versus-host disease in allogeneic bone marrow transplantation. Clin Transplant 2001; 15: 317-323.

7 Ji SQ, Chen HR, Wang HX et al. Comparison outcome of allogeneic bone marrow transplantation with and without granulocyte colony stimulating factor (G-CSF) (lenograstim) donor marrow priming in patients with chronic myelogenous leukemia (CML). Biol Blood Marrow Transpl 2002; 8: 261-267.

8 Przepiorka D, Weisdorf D, Martin P et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant 1995; 15: 825-828. MEDLINE

9 Pike BL, Robinson WA. Human bone marrow colony growth in agar gel. J Cell Physiol 1970; 76: 77-81. MEDLINE

10 Henslee-Downey PJ, Abhyankar SH, Parrish RS et al. Use of partially mismatched related donors extends access to allogeneic marrow transplant. Blood 1997; 89: 3864-3872. MEDLINE

11 O'Reilly RJ, Kernan NA, Canningham I et al. Allogeneic transplants depleted of T cell by soybean lectin agglutination and E-rosette depletion. Bone Marrow Transplant 1988; 3: 3-6.

12 Serody JS, Sparks CD, Lim S et al. Comparison of granulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood progenitor cells and G-CSF-stimulated bone marrow as a source of stem cells in HLA-matched sibling transplantation. Biol Blood Marrow Transplant 2000; 6: 434-440.

13 Isola LM, Scigliano E, Skerrett D et al. A pilot study of allogeneic bone marrow transplantation using related donor stimulated with G-CSF. Bone Marrow Transplant 1997; 20: 1033-1037.

14 Couban S, Messner HA, Andreou P et al. Bone marrow mobilized with granulocyte colony-stimulating factor in related allogeneic transplant recipients: a study of 29 patients. Biol Blood Marrow Transplant 2000; 6: 422-427.

15 Bensinger WI, Clift R, Martin P et al. Allogeneic peripheral blood stem cell transplantation in patients with advanced hematological malignancies: a retrospective comparison with marrow transplantation. Blood 1996; 88: 2794-2800. MEDLINE

16 Darrant ST, Morton AJ. A randomized trial of filgrastim (G-CSF) stimulated donor marrow versus peripheral blood for allogeneic transplantation: increased extensive chronic graft-versus-host disease following PBSC transplantation. Blood 1999; 94 (Suppl. 1): 608a.

17 Salgame P, Abrams JS, Clayberger C et al. Differing lymphokine profiles of function subset of human CD4 and CD8 T cell clones. Science 1991; 254: 279-282. MEDLINE

18 Ferrara JL. Pathogenesis of acute graft-versus-host disease: cytokines and cellular effectors. J Hematother Stem Cell Res 2000; 9: 299-306.

19 Krenger W, Cooke KR, Crawford JM et al. Transplantation of polarized type 2 donor T cells reduces mortality caused by experimental graft-versus-host disease. Transplantation 1996; 62: 1278-1285. MEDLINE

20 Pan L, Delmonte J, Jalonen K et al. Pretreatment of donor mice with granulocyte colony-stimulating factor polarizes donor T lymphocytes toward type 2 cytokine production and reduces severity of experimental graft-versus-host disease. Blood 1995; 86: 4422-4429. MEDLINE

21 Tayebi H, Kuttler F, Saas P et al. Effect of granulocyte colony-stimulating factor mobilization on phenotypical and functional properties of immune cells. Exp Hematol 2001; 29: 458-470. MEDLINE

22 Deeg HJ, Spitzer TR, Cottler-Fax M et al. Conditioning-related toxicity and acute GVHD in patients given methotrexate/cyclosporine prophylaxis. Bone Marrow Transplant 1991; 7: 193-198. MEDLINE

23 Xun CQ, Thompson JS, Jennings CD et al. Effect of total body irradiation, busulfan-cyclophosphamide or cyclophosphamide conditioning on inflammatory cytokine release in H-2 incompatible transplanted SCID mice. Blood 1994; 83: 2360-2367. MEDLINE

24 Antin JH, Ferrara JLM. Cytokine dysregulation and acute GVHD. Blood 1992; 80: 2964-2969. MEDLINE

25 Kolb HJ, Mittermuller J, Clemm C et al. Donor leukocyte transfusion for treatment of recurrent CML in marrow transplant patients. Blood 1990; 76: 2462-2465. MEDLINE

26 Weiden PL, Sullivan KM, Flournoy N et al. Antileukemic effect of chronic GVHD. Contribution to improved survival after allogeneic marrow transplantation. New Engl J Med 1981; 304: 1529-1533. MEDLINE

27 Mineishi S, Longo WL, Atkinson ME et al. Addition of high-dose Ara-c to the BMT conditioning regimen reduces leukemia relapse without an increase in toxicity. Bone Marrow Transplant 1999; 23: 1217-1222.

28 Storb R, Cong Y, Wanger JL et al. Stable mixed hematopoietic chimerism in DLA-identical littermate dogs given sublethal TBI before and pharmacological immunosuppression after marrow transplantation. Blood 1997; 89: 3048-3054. MEDLINE

29 Basara N, Blau WI, Kiehl MG et al. Mycophenolate mofetil for the prophylaxis of acute GVHD in HLA-mismatched bone marrow transplant patients. Clin Transplant 2000; 14: 121-126. Article MEDLINE

Figure 1 Probability of disease-free survival.

Table 1 Patients' characteristics, donor-host relationship and HLA typing

Table 2 Characteristics of the marrow grafts and engraftment

Table 3 Bone marrow contents before and after G-CSF donor priming

Table 4 GVHD and clinical outcome