SUMMARY
We sought to determine whether differences in chronic graft versus host disease (GVHD) rates would lead to survival differences comparing 2463 peripheral blood (PB) and 1713 bone marrow (BM) hematopoietic cell transplant recipients. Patients had acute leukemia, chronic myeloid leukemia (CML) or myelodysplastic syndrome, received myeloablative conditioning regimens and calcineurin-inhibitor GVHD prophylaxis. There were no significant differences in long-term survival after transplantation of PB and BM except first chronic phase CML. For these patients, 5-year rate of survival was lower after transplantation of PB compared to BM (35 % vs. 56%, p=0.001). Although mortality risks were higher in patients with chronic GVHD after both PB (HR 1.58; p<0.001) and BM (HR 1.73; p<0.001) transplants, its effect on mortality did not differ by graft (p=0.42). BM is the preferred graft for first chronic phase CML where as either graft is suitable for other leukemias.
INTRODUCTION
Over the past decade transplantation of peripheral blood hematopoietic cells (PB) has increased and now accounts for 75% of unrelated adult donor transplants. The results of a phase III clinical trial that randomized 550 donors and their recipients to receive PB or bone marrow (BM) from unrelated adult donors did not record significant two-year survival differences between PB and BM transplantation.1 However, the incidence of chronic graft-versus-host disease (GVHD) was higher after PB transplantation and more severe requiring longer duration of therapy compared with BM. The effect of long-term outcomes was not determined.
In HLA-matched sibling transplantation, long-term follow-up did not demonstrate significant survival differences between PB and BM transplantation for acute leukemia.2,3 However, there were differences in long-term survival for those with chronic myeloid leukemia (CML).2 In that report, compared with transplantation of BM, survival rates were lower after transplantation of PB for patients transplanted in first chronic phase, but higher for those transplanted with more advanced disease.2
In unrelated adult donor transplants, it is uncertain whether with longer follow-up, the recorded higher incidence of chronic GVHD after PB transplantation will reduce survival. Financial constraints limit the follow-up of clinical trial recipients beyond the trial period, which is usually on the order of two years and insufficient to measure longer-term outcomes. Therefore, using data reported to the Center for International Blood and Marrow Transplant Research, we asked whether PB or BM results in better long-term survival in patients with hematologic malignancy.
PATIENTS AND METHODS
Patients
Included are patients aged 18 years and older with acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), myelodyspalstic syndrome (MDS) and CML transplanted in the United States between 2000 and 2008. Patients received PB or BM from adult unrelated donors who were HLA-matched at the allele-level at HLA-A, -B, -C and –DRB1 (n=3174) or mismatched at a single HLA-locus (n=1002). The transplant-conditioning regimens were myeloablative and GVHD prophylaxis included either cyclosporine or tacrolimus. Myeloablative regimens were defined as containing a total body irradiation dose of 1000cGy or higher, a busulfan dose >8 mg/kg orally or >6.5 mg/kg intravenously, or a melphalan dose >150 mg/m2.4 Transplantations with ex vivo T-cell depleted or CD34 selected grafts were excluded. The median follow-up of PB recipients was 5 years and that of BM recipients, 6 years.
Endpoints
The primary outcome was overall survival. Incidences of chronic GVHD were based on reports from each transplant center using standard criteria.5 Non-relapse mortality was defined as death not related to disease recurrence, and relapse was defined as disease recurrence based on morphological, cytogenetic or molecular evaluation. Death from any cause was considered overall mortality and surviving patients were censored at last follow-up.
Statistical analysis
Patient, disease and transplantation characteristics were compared using chi-square statistics for categorical variables. The probability of chronic GVHD was calculated using the cumulative incidence estimator to accommodate competing risks.6 Probabilities of overall survival, non-relapse mortality and relapse were computed by disease and disease status and, graft type from Cox proportional hazard regression models,7 adjusted for patient age, performance score, donor-recipient cytomegalovirus serostatus, HLA-match, ABO match, donor age, conditioning regimen, GVHD prophylaxis and transplant year. All analyses were performed using SAS, version 9.3 (Cary, NC).
RESULTS
Patient, disease and transplant characteristics
The characteristics of the study population are shown in Table 1. There were differences in the characteristics of patients who received BM and PB. Recipients of PB were slightly older (median age 42 versus 39 years), more likely to report performance scores less than 90, slightly more often HLA-matched to their donor but less likely to be blood group ABO matched, more often received non-irradiation containing regimens, and more often received tacrolimus-containing GVHD prophylaxis. PB recipients were less likely to have CML. PB transplants were more recent, with most transplants occurring after 2004.
Table 1.
Characteristics of Patients, Disease and Transplantation
| Characteristics | Bone marrow | Peripheral blood | p-value |
|---|---|---|---|
| Number | 1713 | 2463 | |
| Age, years* | <0.0001 | ||
| 18 – 29 | 501 (29%) | 566 (23%) | |
| 30 – 39 | 393 (23%) | 515 (21%) | |
| 40 – 49 | 473 (28%) | 677 (27%) | |
| 50 – 59 | 295 (17%) | 573 (23%) | |
| 60 – 70 | 51 ( 3%) | 132 ( 5%) | |
| Sex, male | 940 (55%) | 1366 (55%) | 0.73 |
| Performance score | <0.0001 | ||
| 90, 100 | 1075 (63%) | 1498 (61%) | |
| < 90 | 440 (26%) | 770 (31%) | |
| Not reported | 198 (12%) | 195 ( 8%) | |
| Recipient CMV serostatus | 0.11 | ||
| Positive | 964 (56%) | 1406 (57%) | |
| Negative | 738 (43%) | 1026 (42%) | |
| Not reported | 11 ( 1%) | 31 ( 1%) | |
| Disease and disease status | <0.0001 | ||
| Acute myeloid leukemia | |||
| 1st complete remission | 258 (15%) | 539 (22%) | |
| 2nd complete remission | 225 (13%) | 329 (13%) | |
| Relapse | 299 (17%) | 461 (19%) | |
| Acute lymphoblastic leukemia | |||
| 1st complete remission | 144 ( 8%) | 256 (10%) | |
| 2nd complete remission | 149 ( 9%) | 194 ( 8%) | |
| Relapse | 101 ( 6%) | 151 ( 6%) | |
| Myelodysplastic syndrome | |||
| Refractory anemia | 54 ( 3%) | 92 ( 4%) | |
| Refractory anemia with excess blasts (>5% blasts in bone marrow) | 91 ( 5%) | 146 ( 6%) | |
| Chronic myeloid leukemia | |||
| 1st chronic phase | 238 (14%) | 116 ( 5%) | |
| 2nd chronic/accelerated phase | 126 ( 7%) | 143 ( 6%) | |
| Blast phase | 28 ( 2%) | 36 ( 1%) | |
| Donor-recipient HLA match | 0.04 | ||
| 8/8 HLA-matched | 1273 (74%) | 1901 (77%) | |
| 7/8 HLA-matched | 440 (26%) | 562 (23%) | |
| Donor-recipient ABO match | <0.0001 | ||
| Matched | 697 (41%) | 927 (38%) | |
| Minor mismatch | 411 (24%) | 499 (20%) | |
| Major mismatch | 491 (29%) | 711 (29%) | |
| Not reported | 114 ( 7%) | 326 (13%) | |
| Donor-recipient sex match | 0.14 | ||
| Female donor, male recipient | 282 (16%) | 418 (17%) | |
| Other | 1414 (83%) | 2003 (81%) | |
| Not reported | 17 ( 1%) | 42 ( 2%) | |
| Donor age, years | 0.73 | ||
| 18 – 32 | 706 (41%) | 941 (38%) | |
| 33 – 50 | 806 (47%) | 1051 (43%) | |
| > 50 | 79 ( 5%) | 116 ( 5%) | |
| Not reported | 122 ( 7%) | 355 (14%) | |
| Conditioning regimen | <0.0001 | ||
| TBI + cyclophosphamide | 1016 (59%) | 1098 (44%) | |
| TBI + other agents | 55 ( 4%) | 192 ( 8%) | |
| Busulfan + cyclophosphamide | 482 (28%) | 793 (32%) | |
| Busulfan + fludarabine | 160 ( 9%) | 380 (15%) | |
| GVHD prophylaxis | <0.0001 | ||
| Tacrolimus-containing | 1054 (62%) | 1849 (75%) | |
| Cyclosporine-containing | 659 (38%) | 614 (25%) | |
| Cell dose per kilogram body weight | <0.0001 | ||
| TNC <3 × 108 / CD34 < 4.5 × 106 | 934 (55%) | 350 (14%) | |
| TNC ≥3 × 108 / CD34 ≥ 4.5 × 106 | 635 (37%) | 1526 (62%) | |
| Not reported | 144 ( 8%) | 587 (24%) | |
| Transplant period | <0.0001 | ||
| 2000 – 2004 | 1167(68%) | 907 (37%) | |
| 2005 – 2008 | 546 (32%) | 1556 (63%) | |
| Follow-up, median (range), months | 73 (3 – 137) | 61 (3 – 136) |
Chronic GVHD
Chronic GVHD was significantly higher after PB transplantation compared to BM (HR 1.45, 95% CI 1.32 – 1.59, p<0.0001). Chronic GVHD was graded as extensive in 85% of PB compared with 76% of BM recipients (p<0.0001). Although mortality risks were higher in patients with chronic GVHD after PB (HR 1.58; p<0.001) and BM (HR 1.73; p<0.001) transplants, its effect on mortality did not differ by graft despite the fact that chronic GVHD after PB transplants were more severe (p=0.42).
Non-relapse mortality, relapse and overall survival
The probabilities of overall survival, non-relapse mortality, and relapse by disease and disease status, and adjusted for patient age, performance score, donor-recipient cytomegalovirus serostatus, HLA-match, ABO match, donor age, conditioning regimen, GVHD prophylaxis and transplant year, factors associated with transplantation outcomes in multivariate analysis are shown in Table 2, Figures 1–3. For CML in first chronic phase, outcomes were also adjusted for tyrosine kinase inhibitor use prior to transplantation. CD34 cell dose or total nucleated cell dose was not associated with outcomes. For patients with acute leukemia, there were no significant differences in 7-year rates of overall survival, non-relapse mortality, and relapse. Similarly, for patients with early stage MDS there were no significant differences in rates of overall survival, non-relapse mortality, and relapse. However, for those transplanted with advanced stage MDS (refractory anemia with excess blasts), relapse rates were lower after PB transplants. There were significant differences in long-term survival by graft type for CML. Transplantation of PB for those in first chronic phase was associated with higher rates of non-relapse mortality and consequently, lower rates of survival.
Table 2.
Probabilities of non-relapse mortality, relapse and overall survival after PB and BM transplantation, adjusted for patent age, performance score, cytomegalovirus serostatus, donor-recipient HLA-match, ABO- match, conditioning regimen, GVHD prophylaxis, donor age and year of transplant
| Probability estimate* (95% confidence interval) |
||||
|---|---|---|---|---|
| Number evaluable BM/PB |
BM | PB | P-value | |
| Non-relapse Mortality | ||||
| * AL CR1 | 402/795 | 35% (30 – 40) | 36% (32 – 40) | 0.71 |
| * AL CR2 | 374/523 | 41% (35 – 46) | 36% (31 – 41) | 0.22 |
| * AL Relapse | 400/612 | 37% (32 – 41) | 39% (35 – 43) | 0.49 |
| ** MDS RA | 54/92 | 36% (23 – 49) | 48% (37 – 58) | 0.21 |
| ** MDS RAEB-1, RAEB-2 | 91/146 | 31% (22 – 40) | 44% (36 – 52) | 0.04 |
| *** CML CP1 | 233/105 | 38% (32 – 45) | 59% (49 – 70) | 0.002 |
| *** CML CP2, AP, BP | 154/179 | 40% (33 – 48) | 37% (29 – 45) | 0.57 |
| Relapse | ||||
| * AL CR1 | 402/795 | 30% (25 – 35) | 29% (26 – 32) | 0.75 |
| * AL CR2 | 374/523 | 29% (25 – 34) | 36% (31 – 40) | 0.07 |
| * AL Relapse | 400/612 | 54% (49 – 59) | 51% (48 – 56) | 0.52 |
| ** MDS RA | 54/92 | 13% (4 – 23) | 18% (10 – 26) | 0.45 |
| ** MDS RAEB-1, RAEB-2 | 91/146 | 43% (33 – 54) | 23% (16 – 30) | 0.003 |
| *** CML CP1 | 230/104 | 9% (5 – 13) | 6% (1 – 11) | 0.38 |
| *** CML CP2, AP, BP | 154/179 | 27% (20 – 35) | 34% (26 – 41) | 0.23 |
| Overall survival | ||||
| * AL CR1 | 402/795 | 41% (36 – 47) | 41% (37 – 45) | 0.89 |
| * AL CR2 | 374/523 | 32% (27 – 37) | 31% (26 – 36) | 0.79 |
| * AL Relapse | 400/612 | 13% (9 – 16) | 12% (9 – 15) | 0.74 |
| ** MDS RA | 54/92 | 53% (39 – 67) | 38% (28 – 48) | 0.10 |
| ** MDS RAEB-1, RAEB 2 | 91/146 | 31% (22 – 41) | 31% (23 – 39) | 0.99 |
| *** CML CP1 | 238/109 | 56% (50 – 62) | 35% (26 – 45) | 0.001 |
| *** CML CP2, AP, BP | 154/179 | 42% (35 – 50) | 33% (25 – 40) | 0.07 |
BM = bone marrow; PB = peripheral blood; AL = acute leukemia; CR = complete remission; MDS = myelodysplastic syndrome, RA = refractory anemia; RAEB = refractory anemia with excess blasts; CML = chronic myeloid leukemia; CP = chronic phase; AP = accelerated phase; BP = blast phase
Shown are the 7-year rates of non-relapse mortality, relapse and overall survival for AL CR1, CR2 and relapse. Causes of death did not differ by graft type; approximately 40% of patients in both groups died from recurrent leukemia.
Shown are the 5-year rates of non-relapse mortality, relapse and overall survival for MDS RA, RAEB-1 and RAEB-2. BM recipients with RAEB-1 and RAEB-2 were more likely to report death from recurrent disease compared to PB recipients (37% vs. 23%). Other causes of death did not differ by graft type.
Shown are the 5-year rates of non-relapse mortality, relapse and overall survival for CML CP1, CP2, AP and BP. Death from recurrent leukemia for patients with CML CP1 was infrequent accounting for 5% of deaths in BM recipients and 3% of deaths in PB recipients. Death from chronic GVHD varied by graft type for patients with CML CP1; 26% in BM recipients compared to 36% in PB recipients.
Figure 1.
The 7-year probability of overall survival by disease status after PB and BM transplantation for acute leukemia, adjusted for patent age, performance score, cytomegalovirus serostatus, donor-recipient HLA-match, blood group ABO-match, conditioning regimen, GVHD prophylaxis, donor age and year of transplant
Figure 3.
The 5-year probability of overall survival by disease status after PB and BM transplantation for chronic myeloid leukemia, adjusted for patent age, performance score, cytomegalovirus serostatus, donor-recipient HLA-match, blood group ABO- match, conditioning regimen, GVHD prophylaxis, donor age and year of transplant. Additionally, survival rates for chronic myeloid leukemia in first chronic phase, is adjusted for tyrosine kinase inhibitor use prior to transplantation and the interval from diagnosis to transplantation.
DISCUSSION
We report on transplantation-outcomes in a large cohort of patients with acute leukemia, MDS and CML who received PB or BM transplantation in 2000 to 2008. Consistent with the findings of others, chronic GVHD rates were higher and more severe after transplantation of PB compared to BM.1,9 Anti-thymocyte globulin with standard GVHD prophylaxis, which has been shown to be effective in lowering GVHD rates as well as its severity, was used for 20% of PB and 24% of BM transplants.10–12 Trials aimed at better GVHD prophylaxis that is either drug-mediated or through graft manipulation are needed to lower chronic GVHD rates after PB transplantation. For patients with acute leukemia, long-term survival was comparable after transplantation of PB and BM. Our findings are consistent with that reported after HLA-matched sibling transplantation for acute leukemia2,3 except for one meta-analysis8, which suggested higher survival after transplantation of PB from HLA-matched siblings for those with advanced acute leukemia. Long-term survival was comparable after transplantation of PB and BM for MDS but for more advanced MDS (RAEB), the pattern of treatment failure differed by graft type. Any advantage from fewer relapses after transplantation of PB for MDS-RAEB was negated by higher non-relapse mortality. On the other hand for good risk CML (first chronic phase), higher non-relapse mortality after transplantation of PB compared to BM resulted in lower survival, long-term after transplantation of PB. Our findings are consistent with long-term survival observed after HLA-matched sibling transplantation.2 In early CML, relapse rates are low with both graft types (9% and 6%; Table 2) leaving little to gain by higher chronic GVHD after PB transplantation, which resulted in higher mortality, long-term. In contrast, for patients with CML transplanted in second chronic, accelerated or blast phase there were no significant differences in rates of overall survival, non-relapse mortality or relapse. This is different from HLA-matched sibling transplantation where mortality risks were lower using PB in those with advanced CML. Non-relapse mortality rates are high with both graft types after unrelated donor transplantation and in the absence of lower rates of relapse after PB transplantation it is not surprising we did not observe differences in rates of survival.
Review of numbers of unrelated donor transplantation for CML between 2000 and 2013 in the United States reported to the Center for International Blood and Marrow Transplant Research revealed the following: 1) there was a marked decline in numbers of unrelated donor transplants after 2000 and explained by the availability of tyrosine kinase inhibitors and 2) the predominance of PB transplants. In the period between 2000 and 2002, BM was the predominant graft accounting for 81% of transplants. There was an increase in the use of PB grafts beginning in 2003 and has continued to date. Between 2003 and 2008, 41% of transplants used BM grafts and between, 2009 and 2013 only 29% used BM grafts. When limited to CML first CP, between 2003 and 2008, 48% of transplants used BM grafts and between 2009 and 2013, 34% of transplants used BM grafts. The adverse effect on survival after transplantation of PB from HLA-matched siblings for CML in first CP was published in 2006.2 Although it can be argued the data reported were for HLA-matched sibling transplantation the switch to PB as the predominant graft for unrelated donor transplantation occurred in the absence of data to support the switch for patients with good risk disease.
This was not a randomized trial and therefore subject to bias owing to the complex selection criteria that underlie the choice of graft for unrelated donor transplantation. Nevertheless, this is the first report describing long-term outcomes comparing unrelated donor PB and BM transplantation for hematologic malignancy with myeloablative transplant conditioning regimens. With the exception of patients with CML transplanted in first chronic phase, long-term survival is similar to that after transplantation of PB and BM. For CML in first chronic phase, lower rates of long-term survival after PB transplantation imply BM is the preferred graft.
Figure 2.
The 5-year probability of overall survival by disease status after PB and BM transplantation for myelodysplastic syndrome, adjusted for patent age, performance score, cytomegalovirus serostatus, donor-recipient HLA-match, blood group ABO-match, conditioning regimen, GVHD prophylaxis, donor age and year of transplant
Highlights.
Comparable long-term survival after PB and BM transplants for acute leukemia
Lower long-term survival after PB transplants for CML in first chronic phase
Effect of chronic GVHD on survival did not differ by graft type
Acknowledgement
Supported by a Public Health Service grant (U24-CA76518) from the National Cancer Institute, the National Heart Lung and Blood Institute and the National Institute of Allergy and Infectious Diseases, and Health Resources and Services Administration (HHSH234200637015C) for the CW Bill Young Stem Cell Transplantation program’s contract for the Stem Cell Therapeutic Outcomes Database. ME is a Clinical Scholar of the Leukemia and Lymphoma Society.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of interest: the authors declare none
Reference
- 1.Anasetti C, Logan BR, Lee SJ, et al. Peripheral blood stem cells versus bone marrow from unrelated donors. N Engl J Med. 2012;367(16):1487–1496. doi: 10.1056/NEJMoa1203517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Schmitz N, Eapen M, Horowitz MM, et al. Long-term outcome of patients given transplants of mobilized blood or marrow: a report from the International Bone Marrow Transplant Registry and the European Group for Blood and Marrow Transplantation. Blood. 2006;108:4288–4290. doi: 10.1182/blood-2006-05-024042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Friedrichs B, Tichelli A, Bacigalupo A, et al. Long-term outcomes and late effects in patients transplanted with mobilized blood or bone marrow: a randomised trial. Lancet Oncol. 2010;11:331–338. doi: 10.1016/S1470-2045(09)70352-3. [DOI] [PubMed] [Google Scholar]
- 4.Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning regimens: working definitions. Biol Blood Marrow Transplant. 2009;15(12):1628–1633. doi: 10.1016/j.bbmt.2009.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host disease syndrome in man: a long-term clinicopathophysiologic study 20 Seattle patient. Am J Med. 1980;69:204–217. doi: 10.1016/0002-9343(80)90380-0. [DOI] [PubMed] [Google Scholar]
- 6.Lin DY. Non-parametric inference for cumulative incidence function in competing risks. Stat Med. 1997;16:901–910. doi: 10.1002/(sici)1097-0258(19970430)16:8<901::aid-sim543>3.0.co;2-m. [DOI] [PubMed] [Google Scholar]
- 7.Cox DR. Regression models and life tables. Stat Soc B. 1972;4:187–200. [Google Scholar]
- 8.Stem Cell Trialists Collaborative Group. Allogeneic peripheral blood stem-cell compared with bone marrow transplantation in the management of hematologic malignancies: an individual patient data meta-analysis of nine randomized trials. J Clin Oncol. 2005;23:5074–5087. doi: 10.1200/JCO.2005.09.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Flowers ME, Parker PM, Johnston LJ, et al. Comparison of chronic graft-versus-host disease after transplantation of peripheral blood stem cells versus bone marrow in allogeneic recipients: long-term follow-up of a randomized trial. Blood. 2002;100:415–419. doi: 10.1182/blood-2002-01-0011. [DOI] [PubMed] [Google Scholar]
- 10.Bacigalupo A, Lamperelli T, Barisione G, et al. Thymoglobulin prevents chronic graft-versus-host disease, chronic lung dysfunction, and late transplant-related mortality: long-term follow-up of a randomized trial in patients undergoing unrelated donor transplantation. Biol Blood Marrow Transplant. 2006;12:560–565. doi: 10.1016/j.bbmt.2005.12.034. [DOI] [PubMed] [Google Scholar]
- 11.Finke J, Bethge WA, Schmoor C, et al. Standard graft-versus-host disease prophylaxis with or without anti-T-cell globulin in hematopoietic cell transplantation from matched unrelated donors: a randomized, open label, multicenter phase 3 trial. Lancet Oncol. 2009;10:855–864. doi: 10.1016/S1470-2045(09)70225-6. [DOI] [PubMed] [Google Scholar]
- 12.Socie G, Schmoor C, Bethge WA, et al. Chronic graft-versus-host disease: long-term results from a randomized trial on graft-versus-host disease prophylaxis with or without anti-T-cell globulin ATG-Fresenius. Blood. 2011;117(23):6375–6382. doi: 10.1182/blood-2011-01-329821. [DOI] [PubMed] [Google Scholar]