Scholarly article on topic 'Long-term follow-up after allogeneic granulocyte colony-stimulating factor--primed bone marrow transplantation'

Long-term follow-up after allogeneic granulocyte colony-stimulating factor--primed bone marrow transplantation Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{}

Abstract of research paper on Clinical medicine, author of scientific article — Luis Isola, Eileen Scigliano, Steven Fruchtman

Abstract Granulocyte colony-stimulating factor (G-CSF) priming increases the number of progenitor cells in harvested bone marrow (BM) and has been used for allogeneic transplantation. Primed bone marrow (pBM) seems to offer faster engraftment than steady-state BM, but the stability of such engraftment has been questioned. The incidence of graft-versus-host disease (GVHD) and disease relapse after pBM, compared with such incidence after BM or peripheral blood progenitor allotransplantation, has not been established. We studied the long-term outcome (median follow-up, 24 months) of sibling matched allografting with G-CSF pBM. Seventeen patients received pBM from matched sibling donors primed with G-CSF 10 microg/kg per day for 2 days before BM harvest. Conditioning consisted of total body irradiation and cyclophosphamide (CY); busulfan and CY; or total lymphoid irradiation, CY, and antithymocyte globulin. All infused grafts contained > or = 3.5 to 4 x 10(8) mononuclear cells per kilogram. Ten of 17 patients received methotrexate as part of their GVHD prophylaxis. International Bone Marrow Transplant Registry definitions for engraftment were used. Control subjects consisted of 112 consecutive patients who received allogeneic transplantation at our institution with steady-state BM; control subjects for length of hospitalization consisted of the subset of patients who underwent transplantation during 1996. Neutrophil engraftment occurred a median of 7 days earlier in primed bone marrow transplantation (pBMT) patients when compared with steady-state BMT patients; this shortened hospitalization by a median of 11 days. The peritransplant mortality rate was 18% in pBMT patients and 25% in BMT patients (not significant). The rate of GVHD of grade > II and the rate of relapse were almost identical in pBMT and BMT patients (GVHD: 18% and 19%, respectively; relapse: 14% and 13%, respectively). There were 4 transplant-related deaths within the first 100 days; 1 patient died of disease relapse on day 470. Twelve patients remained alive on days 430 through 1522 after BMT. Results showed that pBM allografts resulted in more rapid engraftment and shorter hospitalization. All patients maintained stable donor engraftment. In this cohort of patients, G-CSF pBMT resulted in rates of GVHD, disease relapse, and peritransplant mortality that were similar to those produced by conventional BMT. Biol Blood Marrow Transplant 2000;6(4A):428-33.

Academic research paper on topic "Long-term follow-up after allogeneic granulocyte colony-stimulating factor--primed bone marrow transplantation"

Biology of Blood and Marrow Transplantation 6:428-433 (2000) © 2000 American Society for Blood and Marrow Transplantation

Long-Term Follow-Up After Allogeneic Granulocyte Colony-Stimulating Factor-Primed Bone Marrow Transplantation

Luis Isola, Eileen Scigliano, Steven Fruchtman

Bone Marrow Transplantation Service, Division of Hematology, Department of Medicine, Mount Sinai Medical Center, New York, New York

Correspondence and reprint requests: Luis Isola, Box 1079, Mount Sinai Medical Center, 1 Gustave L. Levy Pl., New York, NY 10029 (e-mail: L_Isola@smtplink.mssm.edu)

(Received December 17, 1999; accepted March 28, 2000)

ABSTRACT

Granulocyte colony-stimulating factor (G-CSF) priming increases the number of progenitor cells in harvested bone marrow (BM) and has been used for allogeneic transplantation. Primed bone marrow (pBM) seems to offer faster engraftment than steady-state BM, but the stability of such engraftment has been questioned. The incidence of graft-versus-host disease (GVHD) and disease relapse after pBM, compared with such incidence after BM or peripheral blood progenitor allotransplantation, has not been established. We studied the long-term outcome (median follow-up, 24 months) of sibling matched allografting with G-CSF pBM. Seventeen patients received pBM from matched sibling donors primed with G-CSF 10 pg/kg per day for 2 days before BM harvest. Conditioning consisted of total body irradiation and cyclophosphamide (CY); busulfan and CY; or total lymphoid irradiation, CY, and antithymocyte globulin. All infused grafts contained >3.5 to 4 X 108 mononuclear cells per kilogram. Ten of 17 patients received methotrexate as part of their GVHD prophylaxis. International Bone Marrow Transplant Registry definitions for engraftment were used. Control subjects consisted of 112 consecutive patients who received allogeneic transplantation at our institution with steady-state BM; control subjects for length of hospitalization consisted of the subset of patients who underwent transplantation during 1996. Neutrophil engraftment occurred a median of 7 days earlier in primed bone marrow transplantation (pBMT) patients when compared with steady-state BMT patients; this shortened hospitalization by a median of 11 days. The peritransplant mortality rate was 18% in pBMT patients and 25% in BMT patients (not significant). The rate of GVHD of grade >II and the rate of relapse were almost identical in pBMT and BMT patients (GVHD: 18% and 19%, respectively; relapse: 14% and 13%, respectively). There were 4 transplant-related deaths within the first 100 days; 1 patient died of disease relapse on day 470. Twelve patients remained alive on days 430 through 1522 after BMT. Results showed that pBM allografts resulted in more rapid engraftment and shorter hospitalization. All patients maintained stable donor engraftment. In this cohort of patients, G-CSF pBMT resulted in rates of GVHD, disease relapse, and peritransplant mortality that were similar to those produced by conventional BMT.

KEY WORDS

Cytokine • Stem cells • Transplantation • Mobilization • Harvest

INTRODUCTION

Hastened engraftment after allogeneic transplantation is a highly desirable objective. It decreases the risk of infectious and bleeding complications, lessens the use of antibiotics and blood products, and reduces hospitalization and the procedure costs. Peripheral blood progenitor allo-transplants produce more rapid hematologic reconstitution than related [1,2] and unrelated bone marrow transplants [3]. This results in higher overall survival, especially in

high-risk patients. In standard-risk patients, survival appears similar to that after bone marrow transplantation (BMT). A concern exists that peripheral blood stem cell (PBSC) allotransplants may be fraught with a higher incidence of chronic graft-versus-host disease (GVHD) [4,5]. Thus, for patients with standard risk, PBSC transplantations may offer inferior quality of life when compared with BMT and may not have any clear advantage aside from more rapid hematologic reconstitution.

We demonstrated in a pilot study that granulocyte colony-stimulating factor (G-CSF) priming before marrow harvesting resulted in a graft with a higher progenitor cell content, which produced rapid engraftment [6]. Two randomized studies compared cytokine-primed bone marrow (BM) and cytokine-mobilized PBSCs in the setting of autologous transplants for non-Hodgkin's lymphoma (NHL) and found no significant differences in rate of engraftment [7,8]. In mice BM, G-CSF priming markedly increased repopulating capacity [9]. However, some investigators have expressed serious concerns regarding the stability of engraftment provided by primed bone marrow transplantation (pBMT). In a study of 12 patients who underwent allogeneic pBMT, 4 experienced delayed graft failure (ie, between days 55 and 130), and 2 required a stem cell (SC) boost. This was in contrast to an appropriate control group in which graft failure was not seen [10].

There are no studies addressing the incidence of chronic GVHD and disease relapse after pBMT. Our previous study had a median follow-up of 265 days after transplantation (range, 214-314 days) [6]. In this article we report on long-term follow-up (ie, 1-5 years) in 17 patients who underwent G-CSF pBMT at Mount Sinai Medical Center, New York, NY.

MATERIALS AND METHODS Patients and Conditioning

Seventeen patients underwent sibling matched BMT between 1995 and 1998. Depending on the disease and protocol, conditioning consisted of either (1) total body irradiation 1500 cGy or busulfan (Bu) 16 mg/kg plus cyclophosphamide (CY) 120 mg/kg or (2) total lymphoid irradiation 900 cGy plus CY 200 mg/kg and antithymocyte globulin (Atgam; Upjohn, Kalamazoo, MI) 80 mg/kg.

Priming and Transplantation

After signing an informed consent, all donors received 2 daily doses of 10 pg/kg G-CSF SCs on the 2 days before harvest. For harvest, the target collection was 4 X 108 mononuclear cells (MNCs) per kilogram. The cell content to estimate the volume of harvest was 2.2 X 108 MNCs per mL bone marrow. The maximum volume collected was limited to 20 mL per kg of donor body weight. The number of cells transplanted was adjusted to 3.5 X 108 MNCs/kg. No manipulation was performed on the product other than red cell or plasma depletion as required.

GVHD Prophylaxis and Supportive Therapy

GVHD prophylaxis included cyclosporin A (CsA 12 mg/kg/day) (tapered after day 120) and methotrexate (MTX) in 10 patients (15 mg/m2 on day 1 and 10 mg/m2 on days 3, 6, and 11). MTX was withheld if the total serum bilirubin concentration was >2 mg/dL. Seven patients with elevated liver function test results received corticosteroids in addition to CsA. Patients who did not receive MTX were given intravenous (IV) methylprednisolone sodium succinate (Solu-Medrol; Pharmacia & Upjohn, Kalamazoo, MI) 0.5 mg/kg per day on days 7 through 14 and 0.5 mg/kg twice per day on days 14 through 28. On discharge, patients treated with methylprednisolone sodium succinate received prednisone

(PRD), which was tapered according to the following schedule: 0.8 mg/kg per day on days 29 through 43, 0.5 mg/kg per day on days 44 through 56, 0.2 mg/kg per day on days 57 through 119, and 0.1 mg/kg per day on days 120 through 179. Intravenous 7-globulin 0.5 g/kg was given every

2 weeks starting on day 1. Supportive care included gut decontamination, sterile laminar airflow (LAF) isolation, fluconazole, high-dose IV acyclovir, and Pneumocystis carinii pneumonia prophylaxis. Patients with diagnoses other than acute myelocytic leukemia (AML) or chronic myelocytic leukemia (CML) received G-CSF or granulocyte-macro-phage colony-stimulating factor starting on day 0. Prophylactic transfusions of platelets were given for a platelet count of <10 X 109/L if no bleeding occurred and <50 X 109/L if bleeding occurred.

Engraftment, GVHD, Length of Hospitalization, and Disease Relapse

Definitions for engraftment were as follows: the first of

3 consecutive days with absolute neutrophil count (ANC) >0.5 X 109/L and ANC >1 X 109/L and the first of 3 consecutive days with platelet count >20 X 109/L after 1 week without platelet transfusions. GVHD was graded as described previously [11]. Donor chimerism was determined by conventional cytogenetics, fluorescence in situ hybridization, variable number of tandem repeats analysis, or any combination of the three. Length of hospitalization was calculated from day of admission to day of discharge.

Control patients for engraftment, GVHD, and peri-transplant mortality consisted of 112 consecutive patients who received allogeneic steady-state BMT at our institution; control patients for length of stay consisted of the subset of patients who underwent transplantation in 1996.

Statistics

For engraftment, Kaplan-Meier plots were compared using the log rank test. For GVHD, regimen-related mortality, and disease relapse, the Mann-Whitney test was used.

RESULTS

Patients

Patient characteristics are summarized in Table 1. Sixteen of the 17 patients had a Karnofsky performance score (KPS) of 100%; 1 had a KPS of 80%. Twelve of the patients were standard risk (ie, had chronic-phase CML, severe aplastic anemia (SAA), AML, or NHL in first complete remission), and 5 were high risk (ie, had refractory lymphoma/leukemia or accelerated-phase CML, or had undergone a previous transplant).

Follow-up data for analysis of GVHD were available for 16 patients; data regarding disease relapse were available for 13 patients. Ten of the 17 patients were previously described after a median follow-up of 144 days (range, 48-585 days) post BMT [6]. Fourteen patients were male, and 3 were female. The median age was 38 years (range, 15-59 years). All patients were genotypically identical with their sibling donors at the HLA class I and II loci as assigned by family studies. Nine had sex-matched and 8 had sex-mismatched donors (5 female donor/male recipients, 2 with previous pregnancies, and 3 male donor/female recipients). GVHD

Table 1. Characteristics of Primed Bone Marrow Transplantation Patients (n = 17)*

UPN MTX Age, y Sex Diagnosis Donor Sex/Parity Conditioning Risk Day Post BMT Current Status

214 _ 44 M NHL F TBI/CY H 1527 A

229 + 38 M CML/CP M TBI/CY L 76 D

231 + 35 M CML/CP M TBI/CY L 1422 A

254 - 24 M NHL F Bu/CY H 43 D

265 - 28 M SAA M TLI/CY/ATG L 1210 A

284 - 36 F AML M TBI/CY L 1100 At

293 + 36 M CML/CP M TBI/CY L 1030 A

304 + 59 M CML/AP M TBI/CY H 91 D

307 + 15 F PNH M TBI/CY L 967 A

314 - 46 M HL/NHL M TBI/CY H 916 A

323 + 24 F SAA M TLI/CY/ATG L 867 A

350 - 54 M MM M TBI/CY H 470 Dt

362 + 49 M MDS F TBI/CY H 24 D

366 - 38 M CML/CP M TBI/CY L 617 A

399 + 41 M SAA F/P TLI/CY/ATG L 454 A

400 + 43 M CML/CP M TBI/CY L 442 A

401 + 45 M CML/CP F/P TBI/CY L 435 A

*UPN indicates unique patient number; MTX, methotrexate; BMT, bone marrow transplantation; NHL, non-Hodgkin's lymphoma; TBI, total body irradiation; CY, cyclophosphamide; H, high; A, alive; CML/CP, chronic myelocytic leukemia, chronic phase; L, low; D, died; Bu, busulfan; SAA, severe aplastic anemia; TLI, total lymphoid irradiation; ATG, antithymocyte globulin; AML, acute myelocytic leukemia; CML/AP, chronic myelocytic leukemia, acute phase; PNH, paroxysmal nocturnal hemoglobinuria; HL, Hodgkin's lymphoma; MM, multiple myeloma; MDS, myelodysplastic syndrome; F/P, parous female. tRelapsed.

prophylaxis consisted of CsA/MTX in 10 patients and CsA/PRD in 7 patients. Seven patients received growth factor support after transplantation.

Characteristics of the control (BMT) group are summarized in Table 2. This group was comparable to the study group in median age, proportion of patients with CML, proportion of patients receiving growth factor support, and ratio of high-risk to standard-risk patients. The proportion of patients receiving CsA/PRD was higher in the pBMT group.

Engraftment

The cell content and characteristics of stimulated and unstimulated grafts have been described [6]. The means for total nucleated count harvested and infused, and CD3+ and CD34+ cells per kilogram recipient body weight were comparable between the pBM and BM grafts. The number of gran-ulocyte-macrophage colony-forming units per kg recipient body weight was higher in pBM grafts (data not shown).

The kinetics of engraftment and length of hospital stay are illustrated in Figure 1. Neutrophil engraftment occurred more rapidly in patients receiving pBM than in those receiving steady-state BM. The median time to an ANC of 500 was 17 and 24 days for pBMT and BMT patients, respectively (P = .0003); median time to an ANC of 1000 was 19 and 26 days, respectively (P = .007). Faster engraftment resulted in shortened median hospitalization (34 days) for pBMT patients compared with BMT patients (45 days) (P = .0005). Median time to a platelet count >20,000 was 21 days for pBMT patients and 25 days for BMT patients; the difference, however, was not statistically significant.

Among patients who received pBMT, there were differences in days to an ANC of 500 and an ANC of 1000 according to whether patients received MTX as part of their

GVHD prophylaxis regimen. Because of the small number of patients in each group, these differences were not significant but favored the group that did not receive MTX. Nevertheless, the cohort of patients receiving pBM and MTX fared

Table 2. Characteristics of Control (Bone Marrow Transplantation) Patients (n = 112)*

Characteristic

Median age (range), y 32 (1-63)

Risk, high/low 30/82

Diagnosis

ALL CRI and CR2 10

AML CRI and CR2 22

CML/CP 30

CML/AP and BC 11

Other 34

Conditioning

TBI/CY 86

Bu/CY 18

Other 8

Growth factor support 49

GVHD prophylaxis

CsA/MTX 97

CsA/PRD 15

*Data are n unless otherwise specified. ALL indicates acute lymphocytic leukemia; CR, complete remission; AML, acute myelocytic leukemia; CML/CP, chronic myelocytic leukemia, chronic phase; AP, acute phase; BC, blast crisis; SAA, severe aplastic anemia; TBI, total body irradiation; CY, cyclophosphamide; Bu, busulfan; GVHD, graft-versus-host disease; CsA, cyclosporine A; MTX, methotrexate; PRD, prednisone.

Figure 1. Kaplan-Meier plots of engraftment comparing the number of primed bone marrow transplantation (BMT) (-) and steady-state BMT

(---) patients with absolute neutrophil counts (ANCs) of 0.5 X 109/L and 1.0 X 109/L, platelets >20 X 109/L, and length of hospitalization in the 2

groups. SCT indicates stem cell transplantation.

better than our historical controls. Interestingly, no obvious differences were noted in median time to a platelet count >20,000 or in length of hospitalization between pBMT patients who did and did not receive MTX as part of their GVHD prophylaxis (Figure 2).

Toxicity, Chimerism, GVHD, and Disease Status

All 17 patients completed conditioning and underwent transplantation. Three patients in the high-risk group died from severe veno-occlusive disease (1 each on days 24, 43, and 94); full donor engraftment had occurred in the last 2 patients who succumbed. One patient in the low-risk group died of thrombotic thrombocytic purpura/hemolytic uremic syndrome on day 74; full donor engraftment had occurred, and the patient had grade I skin GVHD. The overall peritransplant mortality rate was 24% in pBMT patients and 25% in BMT patients (not significant). Twelve patients remained alive at 435 to 1527 days after transplantation. One patient in the high-risk group died from

myeloma relapse on day 470. One patient in the low-risk group experienced an isolated central nervous system relapse that was treated with craniospinal irradiation, systemic chemotherapy, and donor leukocyte infusion on day 325. That patient remained in complete unsustained remission on day 1100; results of cerebrospinal fluid tests at that time were negative.

The median follow-up time for pBMT patients was 862 days. Outcomes according to risk groups are summarized in Table 1.

Of the 12 long-term survivors, 1 has a PS of 50% as a result of chronic extensive GVHD; the remaining 11 have a PS of 100%. The Kaplan-Meier estimate for event-free survival at 3 years is 58% for the pBMT group versus 48% for control (BMT) patients (P = .08) (data not shown). Mortality at day 100, the rate of acute GVHD of grade II through IV, and the relapse rate were similar in pBMT patients and controls (Table 3). One patient in the pBMT group has chronic extensive GVHD.

Figure 2. Median days to absolute neutrophil count (ANC) of 0.5 X 109/L, ANC of 1.0 X 109/L, and platelet count of 20 X 109/L; and median length of hospitalization. Patients who underwent steady-state bone marrow transplantation (El); patients who underwent primed bone marrow transplantation and who did (■) and did not (■) receive methotrexate as part of graft-versus-host disease prophylaxis.

DISCUSSION

In a previous report we described early outcome in 10 patients who underwent pBMT. This pilot study had a relatively short follow-up (median, 144 days; range, 48-585 days) after BMT [6], and therefore, late graft failure, which has been reported by other investigators [10], remained a valid concern. Furthermore, chronic GVHD often occurs a year or more after transplantation, and the data in our previous report were not adequate to estimate its prevalence. The current study, which has a longer follow-up time and incorporates a larger group of patients, confirms that pBM hastens neu-trophil recovery, and the difference in such recovery between pBMT and conventional BMT is statistically significant.

A caveat of our study is the fact that a higher proportion of patients in the control group received MTX for GVHD prophylaxis. As demonstrated in Figure 2, MTX partially blunts the advantage of pBM in terms of neutrophil recovery. There were not enough control patients receiving CsA/PRD or enough patients in the two pBM subgroups to allow for meaningful subset comparisons. However, such comparisons are a highly desirable goal and should be part of a future randomized study.

At 1 to 5 years of follow-up, full donor engraftment is seen in all patients without indication of primary or secondary graft failure, as described by Mavroudis et al. [10]. There are two possible reasons for this discrepancy. First, in the Mavroudis study, the grafts were subjected to T-cell depletion by positive CD34+ selection, which resulted in substantial loss of progenitor cells. The dose infused was 2 to 3 times lower than that given to our patients (0.82 to 3.1 X 106/kg versus 3 to 20 X 106/kg) [6,10]. In addition, T-cell depletion is known to result in occasional graft failure. Second, our protocol was designed to use a very short course (2

days) of G-CSF before harvesting, whereas the study by Mavroudis et al. used a standard (5-day) regimen of G-CSF peripheral blood progenitor mobilization. Our rationale for using a short course of G-CSF was based on the fact that the number of CD34+ cells in peripheral blood peaks after 4 to 6 days of cytokine treatment. This probably coincides with an efflux of progenitor cells from the marrow. It was our hope that using a short course of G-CSF would maximize proliferation and minimize mobilization.

As expected, hospital discharge was more rapid in the group of patients receiving pBM. This could not be attributed to changes in transplant care, as the control group used was chosen because those patients' admission time overlapped with that of the cohort studied. Primed BM involves a modest added cost for the cytokine used in the donor, but this cost is clearly offset by the shortened hospitalization.

An additional concern involves the administration of G-CSF to healthy individuals, especially siblings of myeloid leukemia patients. Our protocol called for follow-up blood cell counts at 1, 6, and 12 months after harvest in donors. More revealing will be the results of a current National Marrow Donor Program study that involves long-term follow-up of donors after a 5-day G-CSF mobilization regimen.

The current study does not allow firm conclusions in terms of disease relapse because of the heterogeneity of patients and controls. However, there have been no relapses in the limited number of patients with chronic-phase CML who received pBM. Overall survival after pBMT compared with survival after BMT or PBSC allotransplants requires further examination in randomized studies using donor/recipient age-, sex-, and disease-matched case-control subjects.

Our patient cohort exhibited an incidence of GVHD that was strikingly similar to that seen in our historical controls who underwent transplantation with sibling matched BM. Because the follow-up time in both groups is relatively long, the respective incidences are not expected to vary significantly with additional follow-up. Although it is conceivable that with larger accrual, pBMT may show a different pattern of acute or chronic GVHD than BMT, currently both procedures appear similar in producing minor antigen-related GVHD.

The change from BM to PBSCs as a source of reconstitution for autologous SC transplantation resulted in dramatic improvement in transplantation-related mortality and had a major impact on the outcome of such transplants. However, outcomes after allogeneic transplantation are also affected by other factors such as GVHD and graft-versus-

Table 3. Acute GVHD, Mortality at Day 100, and Disease Relapse in

pBMT Patients and Control Subjects*

pBMT Steady-State BMT

% n % n

Acute GVHD > grade II 19 3/16 19 19/101

Mortality at day 100 24 4/17 25 28/112

Disease relapse 15 2/13 13 11/84

Data are median (range) or n.

*GVHD indicates graft-versus-host disease; pBMT, primed bone marrow transplantation.

leukemia effects. Therefore, the success rate with the use of PBSCs will be a result of the interaction of these variables with hematologic and immune reconstitution. At this point there is no convincing evidence that, for patients with SAA, early chronic-phase CML, or AML in first remission and without comorbidities, PBSC allotransplants are superior to BM transplants, with the exception of faster engraftment. Primed BM may offer the engraftment advantages of PBSCs and the lower incidence of GVHD seen with BM. Manipulation of PBSCs by, for instance, positive CD34+ selection, may lower the incidence of GVHD. However, carefully designed studies are needed to determine whether such an approach results in a higher incidence of disease relapse.

BM harvesting has the distinct disadvantages of requiring anesthesia and operating room time. In donors with poor venous access, BM harvest circumvents the risks of a central line. Our priming regimen involves a shorter exposure to cytokines for healthy donors than that used for PBSC mobilization. The safety record of G-CSF in healthy donors is quite remarkable, but minimal exposure to cytokines may be desirable in relatives of patients with myeloid disorders. This study supports the need for a randomized study comparing G-CSF-primed BM versus G-CSF-mobilized PBSCs in patients with low transplantation risk. BM priming with other cytokines appears to be a reasonable avenue of investigation.

ACKNOWLEDGMENTS

The authors acknowledge Nigel Henry for expert assistance with data management and Christine Keaveny for protocol management.

REFERENCES

1. Bensinger WI, Weaver CH, Appelbaum FR, et al. Transplantation of allogeneic peripheral blood stem cells mobilized by recombinant human granulocyte colony-stimulating factor. Blood. 1995;85:1655-1658.

2. Korbling M, Przepiorka D, Huh YO, et al. Allogeneic blood stem cell transplantation for refractory leukemia and lymphoma: potential advantage of blood over marrow allografts. Blood. 1995;85: 1659-1665.

3. Ringden O, Remberger M, Runde V, et al. Peripheral blood stem cell transplantation from unrelated donors: a comparison with marrow transplantation. Blood. 1999;94:455-464.

4. Korbling M, Huh YO, Durett A, et al. Allogeneic blood stem cell transplantation: peripheralization and yield of donor-derived primitive hematopoietic progenitor cells (CD34+ Thy-1dlm) and lymphoid subsets, and possible predictors of engraftment and graft-versus-host disease. Blood. 1995;86:2842-2848.

5. Brown RA, Adkins D, Khoury H, et al. Long-term follow-up of high-risk allogeneic peripheral-blood stem-cell transplant recipients: graft-versus-host disease and transplant-related mortality. J Clin Oncol. 1999;17:806-812.

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

7. Damiani D, Fanin R, Silvestri F, et al. Randomized trial of autol-ogous filgrastim-primed bone marrow transplantation versus fil-grastim-mobilized peripheral blood stem cell transplantation in lymphoma patients. Blood. 1997;90:36-42.

8. Weisdorf D, Miller J, Verfaillie C, et al. Cytokine-primed bone marrow stem cells vs. peripheral blood stem cells for autologous transplantation: a randomized comparison of GM-CSF vs. G- CSF. Biol Blood Marrow Transplant. 1997;3:217-223.

9. Bodine DM, Seidel NE, Orlic D. Bone marrow collected 14 days after in vivo administration of granulocyte colony-stimulating factor and stem cell factor to mice has 10-fold more repopulating ability than untreated bone marrow. Blood. 1996;88:89-97.

10. Mavroudis DA, Read EJ, Molldrem J, et al. T cell-depleted gran-ulocyte colony-stimulating factor (G-CSF) modified allogeneic bone marrow transplantation for hematologic malignancy improves CD34+ cell content but is associated with delayed pancytopenia. Bone Marrow Transplant. 1998;21:431-440.

11. Vogelsang GV. Acute graft-versus-host disease following marrow transplantation. Marrow Transplant Reviews. Winter 1992/1993; 2:150.