Abstract
Background:
Peripheral Blood (PB) or Bone Marrow (BM) from unrelated donors can serve as a graft source for hematopoietic cell transplantation. Currently, PB is most commonly used in roughly 80% of adult recipients. Determining the long-term impact of graft source on outcomes would inform this decision.
Methods:
Data collected by the Center for International Blood and Marrow Transplant Research (CIBMTR) from 5,200 adult recipients of a first transplant from a 8/8 or 7/8- human-leukocyte antigen matched unrelated donor for treatment of acute leukemia, chronic myeloid leukemia or myelodysplastic syndrome between the years of 2001–2011 were analyzed to determine the impact of graft source on Graft-versus-Host Disease (GVHD), Relapse-Free Survival (GRFS) defined as freedom from grade III/IV acute GVHD, chronic GVHD requiring immunosuppressive therapy, relapse, or death and overall survival.
Results:
GRFS at 2-years was superior in recipients of BM 16% (95% confidence interval [CI] 14–18) when compared to PB 10% (95% CI, 8–11) (p-value <0·0001)in the 8/8 human-leukocyte antigen -matched cohort and 7/8-matched cohort 11% (95% CI, 8–14) vs. 5% (95% CI, 4–7) (p- value 0·001). With 8/8 matched unrelated donors, overall survival at 5 years was superior in recipients of BM (43%; 95% CI, 40–46) vs. PB (38%; 95% CI 36–40), p =0·014. Inferior survival at 5 years in the PB cohort was attributable to a higher frequency of deaths while in remission when compared to BM recipients. For recipients of 7/8-matched grafts, survival at 5 years were similar BM and PB recipients (32 vs. 29%, p=0·329).
Conclusion:
BM grafts results in improved long-term GRFS and overall survival in recipients of matched unrelated donor transplants and should be considered the unrelated allograft of choice, when available, for adults with acute leukemia, chronic myeloid leukemia and myelodysplastic syndrome.
INTRODUCTION
The majority of allogeneic hematopoietic cell transplants (HCT) performed utilize unrelated donors (URD) with peripheral blood (PB) selected over bone marrow (BM) as the graft source in over 80% of adult recipients.1 In 2012, the results of a randomized trial of BM vs. PB in URD recipients conducted by the Blood and Marrow Clinical Trials Network (BMT CTN 0201 trial) were reported.2 Overall survival (OS) at 2 years was similar in both groups as were the rates of acute GVHD, relapse and non-relapse mortality (NRM). However, PB recipients had higher rates of chronic GVHD and were more likely to require immunosuppressive therapy (IST) at 2 years. In contrast, BM recipients had worse engraftment kinetics including higher rates of primary graft failure. Long-term, patient-reported outcomes from this trial found receipt of BM was associated with higher psychological well-being scores, less chronic GVHD symptom burden and a 50% higher likelihood of returning to work at 5 years when compared to PB.3
Recently the BMT CTN has incorporated a novel endpoint, GVHD and Relapse-Free Survival (GRFS), into the network’s prospective trials4 defined as freedom from grade III-IV acute GVHD, chronic GVHD requiring IST, relapse or death. This endpoint accounts for competing risks (relapse and NRM) while capturing the clinically most significant GVHD events.5 The Center for International Blood and Marrow Transplant Research (CIBMTR) has estimated GRFS to be 23% at 1-year.4 A recent analysis determined that adult recipients of BM from matched siblings had the highest GRFS with PB stem cells from matched sibling donors resulting in a 50% increased risk of events contributing to GRFS at 1- and 2-years than BM.6 Accordingly, the aim of this study is to determine long-term GRFS and OS in adult recipients of HLA-A, B, C and DR matched or 1-locus mismatched URD’s who received either a PB or BM graft. Our hypothesis is that the higher rates of chronic GVHD associated with PB allografts will result in worse long-term survival which may be detected through analysis of a larger data set with sufficient long-term outcomes. If confirmed, these findings would support a change from the current practice favoring PB over BM.
PATIENTS AND METHODS
Patients
The CIBMTR is a voluntary working group of more than 450 centers that contribute detailed data on consecutive allogeneic and autologous HCT to a Statistical Center at the Medical College of Wisconsin in Milwaukee, Wisconsin and the National Marrow Donor Program (NMDP) Coordinating Center in Minneapolis, Minnesota. Participating centers are required to report all transplants consecutively and compliance is monitored by on-site audits. All patients provided written informed consent in accordance with the Declaration of Helsinki for data submission and research participation. The Institutional Review Board of the NMDP approved this study.
Inclusion Criteria
Included patients were ≥ 18 years and underwent a first HCT from an URD for acute leukemia (AL), chronic myeloid leukemia (CML) or myelodysplastic syndrome (MDS) between the years 2001– 2011. All donor-recipient pairs were fully matched or had a single mismatch at HLA-A, - B, -C, and -DRB1 at the allele-level (8/8 or 7/8 HLA-match), and 89% were also matched at HLA-DQ. Of the total 5200 donor-recipient pairs, 81% had HLA typing at HLA-A, -B, -C, and - DRB1 performed through the NMDP retrospective high-resolution typing project as previously described7 and the remaining were reported to the CIBMTR by the transplant centers and validated by the NMDP. Ex vivo T-cell depleted graft recipients were excluded.
Patient and Treatment Characteristics
Graft source was the primary variable for this study. Additional factors examined included patient and donor age, Karnofsky performance score (KPS), the use of T-cell antibodies [anti-thymocyte globulin (ATG) or alemtuzumab], GVHD prophylaxis regimen, diagnosis, disease status (early, intermediate or advanced), time from diagnosis to transplant (< 6, 6–12 and > 12 months), year of transplant (2001–2004, 2005–2007 and 2008–2011), conditioning regimen intensity and CMV serologic status, ABO blood group and sex match between the donor - recipient pairs.
Endpoints
The primary endpoint was GRFS at 2 years post-HCT. Patients event-free for GRFS were those who did not experience grade III-IV acute GVHD, chronic GVHD requiring IST, disease relapse or death from any cause. Kaplan-Meier estimates were used to determine the unadjusted probability of GRFS with differences between the curves determined using log-rank tests. Coxregression was used to determine the independent factors impacting GRFS. Outcomes for the 8/8 HLA- and 7/8-HLA matched cohorts were performed separately as patient characteristics were strongly correlated by HLA-match. Incidences for primary graft failure, acute GVHD and chronic GVHD requiring IST were based on reports from each center using standard criteria. Acute GVHD was graded according to standard criteria while chronic GVHD was diagnosed according to CIBMTR criteria, which includes all patients with clinical criteria of chronic GVHD, regardless of the time of onset of symptoms.8–10 NRM was defined as death not related to disease recurrence and relapse was defined as disease recurrence based on morphologic evaluation.
Statistical Analysis
Multivariate analysis was performed using the Cox proportional hazards model.11 With respect to the primary analysis, all endpoints were examined within 2 years of transplant and patients who survived after 2 years were censored at 2 years. All variables were tested for the affirmation of the proportional hazards assumption. In the 8/8-HLA matched cohort, proportional hazards assumption (which assumes the relative risk remains constant over time) was violated when examining the impact of graft source for several endpoints (except relapse and acute GVHD III- IV) indicating that the impact of graft source varied over time. To account for this, we determined optimal cut points of ≤5 and > 5 months for the GRFS endpoint and chronic GVHD and ≤2 and >2 months for OS, disease-free survival (DFS) and NRM based on the maximum likelihood approach. For these endpoints, separate hazard rates (HR’s) for graft source within these two time intervals were estimated via a time-dependent manner in fitting the Cox models. Additional factors (apart from graft course) that did not conform to the proportional hazards assumption were adjusted through stratification. A stepwise procedure was used to develop models for each outcome with a threshold of 0·05 for both entry and to stay in the model. Interactions between the graft source and adjusted covariates were tested at a significance level of 0·01. No significant interactions were detected. Similarly, as a secondary analysis, we also examined the late effect of outcomes at specific time points. The ‘center’ effect was adjusted in all the multivariable models. All p-values are two-sided. Data analyses were performed using SAS version 9.4 (SAS institute, Cary, NC).
RESULTS
Patient Characteristics
3,901 patients received an 8/8-HLA matched URD HCT between the years of 2001– 2011 of which 72% were PB and 28% were BM. PB was increasingly used over time (from 2001–2004 PB grafts were 56% vs. 81% between the years 2008–2011). Patient characteristics are detailed in Table 1a. The median (range) follow-up for survivors was 60 months (1–124) and 48 months (1–122) for BM and PB (p<0·001), respectively.
Table 1a.
Characteristics of adult patients undergoing first allogeneic MA/RIC transplant for AML, ALL, CML, MDS from an 8/8-matched unrelated donor between 2001–2011, as reported to the CIBMTR.
| Bone | Peripheral | ||
|---|---|---|---|
| Variables | marrow | blood | P-value |
| Number of patients | 1087 | 2814 | |
| Number of centers | 113 | 138 | |
| Patient-related | |||
| Age at transplant, years, median (range) | 44 (18–71) | 48 (18–76) | <0·001 |
| Age at transplant, years | <0·001 | ||
| 18–29 | 272 (25) | 455 (16) | |
| 30–39 | 189 (17) | 459 (16) | |
| 40–49 | 255 (23) | 608 (22) | |
| 50+ | 371 (34) | 1292 (46) | |
| Disease | <0·001 | ||
| AML | 552 (51) | 1631 (58) | |
| ALL | 201 (18) | 474 (17) | |
| CML | 192 (18) | 288 (10) | |
| MDS | 142 (13) | 421 (15) | |
| Disease status * | 0·07 | ||
| Early | 542 (50) | 1417 (50) | |
| Intermediate | 276 (25) | 627 (22) | |
| Advanced | 266 (24) | 767 (27) | |
| Missing | 3 (<1) | 3 (<1) | |
| Time from diagnosis to transplant, months, median (range) | 8(<1–238) | 7 (<1–607) | <0·001 |
| Time from diagnosis to transplant, months | 0·005 | ||
| < 6 | 379 (35) | 1139 (40) | |
| 6≤12 | 293 (27) | 744 (26) | |
| >12 | 413 (38) | 924 (33) | |
| Missing | 2 (<1) | 7 (<1) | |
| Karnofsky performance score at transplant | <0 001 | ||
| < 90 | 278 (26) | 953 (34) | |
| ≥ 90 | 691 (64) | 1695 (60) | |
| Missing | 118 (11) | 166 ( 6) | |
| Donor-related | |||
| Matched unrelated donor age, years, median (range) | 33 (19–60) | 32 (18–61) | 0·03 |
| Matched unrelated donor age at transplant, years | 0·43 | ||
| 18–32 | 538 (49) | 1458 (52) | |
| 33–49 | 486 (45) | 1200 (43) | |
| 50+ | 63 ( 6) | 156 ( 6) | |
| Donor-recipient gender match | 0.08 | ||
| Male / male | 403 (37) | 1153 (41) | |
| Male / female | 314 (29) | 811 (29) | |
| Female / male | 175 (16) | 393 (14) | |
| Female / female | 195 (18) | 457 (16) | |
| Donor-recipient CMV status | 0.14 | ||
| Negative / negative | 303 (28) | 843 (30) | |
| Negative / positive | 384 (35) | 1049 (37) | |
| Positive / negative | 129 (12) | 286 (10) | |
| Positive / positive | 259 (24) | 576 (20) | |
| Missing | 12 ( 1) | 60 ( 2) | |
| Donor-recipient ABO mismatch | 0.87 | ||
| Matched | 485 (45) | 1235 (44) | |
| Minor mismatch | 276 (25) | 699 (25) | |
| Major mismatch | 248 (23) | 662 (24) | |
| Bidirectional mismatch | 78 ( 7) | 218 ( 8) | |
| Transplant-related | |||
| Year of transplant | <0.001 | ||
| 2001–2004 | 424 (39) | 547 (19) | |
| 2005–2007 | 424 (39) | 1242 (44) | |
| 2008–2011 | 239 (22) | 1025 (36) | |
| Conditioning regimen intensity & TBI usage | <0.001 | ||
| MA - TBI | 477 (44) | 905 (32) | |
| MA - no TBI | 452 (42) | 1143 (41) | |
| RIC - TBI | 17 ( 2) | 88 ( 3) | |
| RIC - no TBI | 141 (13) | 678 (24) | |
| ATG / Campath used in conditioning regimen or GVHD | 0.90 | ||
| prophylaxis? | |||
| Yes | 338 (31) | 881 (31) | |
| No | 749 (69) | 1933(69) | |
| GVHD prophylaxis | <0.001 | ||
| Tac-based | 721 (66) | 2177 (77) | |
| CsA-based | 327 (30) | 537 (19) | |
| Others | 39 ( 4) | 100 ( 4) | |
| Follow-up of survivors, median (range) | 60 (1–124) | 48 (1–122) | <0·001 |
Abbreviations: TNC = total nucleated cell, RIC = reduced intensity conditioning, Tac = Tacrolimus, CsA = Cyclosporine.
: Disease status is defined as: Early = AML/ALL (CR1); CML (CP1); MDS (RA/RARS/pre-HCT marrow blasts <5%) Intermediate = AML/ALL (>=CR2); CML (AP or >=CP2) Advanced = AML/ALL (REL/PIF); CML (BP); MDS (RAEB/RAEB-t/CMML or marrow blasts >=5%)
1299 patients received a 7/8-HLA matched URD between the years of 2001–2011 of which 72% received PB and 28% BM. PB was increasingly used over time (2001–2004 PB grafts represented 55% vs. 81% between the years 2008–2011). Patient characteristics for the 7/8- HLA cohort is shown in Table 1b. The median (range) follow-up for survivors was 97 months (4–156) and 73 months (6–168) for BM and PB (p<0·001), respectively.
Table 1b.
Characteristics of adult patients undergoing first allogeneic MA/RIC transplant for AML, ALL, CML, MDS from a 7/8-matched unrelated donor between 2001–2011, as reported to the CIBMTR.
| Bone | Peripheral | ||
|---|---|---|---|
| Variables | marrow | blood | P-value |
| Number of patients | 370 | 929 | |
| Number of centers | 103 | 113 | |
| Patient-related | |||
| Age at transplant, years, median (range) | 41 (18–72) | 46 (18–71) | <0·001 |
| Age at transplant, years | <0·001 | ||
| 18–29 | 103 (28) | 168 (18) | |
| 30–39 | 74 (20) | 169 (18) | |
| 40–49 | 109 (29) | 220 (24) | |
| 50+ | 84 (23) | 372 (40) | |
| Disease | <0·001 | ||
| AML | 183(49) | 546 (59) | |
| ALL | 95(26) | 170 (18) | |
| CML | 67 (18) | 102 (11) | |
| MDS | 25 ( 7) | 111(12) | |
| Disease status * | 0·006 | ||
| Early | 152 (41) | 413 (44) | |
| Intermediate | 126 (34) | 229 (25) | |
| Advanced | 91(25) | 283 (30) | |
| Missing | 1 (<1) | 4 (<1) | |
| Time from diagnosis to transplant, months, median (range) | 10 (<1–269) | 8(<1–229) | 0·005 |
| Time from diagnosis to transplant, months | 0·04 | ||
| < 6 | 98 (26) | 304 (33) | |
| 6≤12 | 106 (29) | 284 (31) | |
| >12 | 165 (45) | 339 (36) | |
| Missing | 1 (<1) | 2 (<1) | |
| Karnofsky performance score at transplant | 0·01 | ||
| < 90 | 231(62) | 542 (58) | |
| ≥ 90 | 105 (28) | 331 (36) | |
| Missing | 34 ( 9) | 56 ( 6) | |
| Donor-related | |||
| Matched unrelated donor age, years, median (range) | 36 (19–60) | 36 (18–59) | 0·57 |
| Matched unrelated donor age at transplant, years | 0·70 | ||
| 18–32 | 138 (37) | 370 (40) | |
| 33–49 | 196 (53) | 472 (51) | |
| 50+ | 36 (10) | 87 ( 9) | |
| Number of patients | 370 | 929 | |
| Donor-recipient gender match | 0·29 | ||
| Male / male | 136 (37) | 313 (34) | |
| Male / female | 98 (26) | 251 (27) | |
| Female / male | 60 (16) | 191 (21) | |
| Female / female | 76 (21) | 174 (19) | |
| Donor-recipient CMV status | 0·50 | ||
| Negative / negative | 83 (22) | 228 (25) | |
| Negative / positive | 38 (10) | 97 (10) | |
| Positive / negative | 131 (35) | 317 (34) | |
| Positive / positive | 116 (31) | 272 (29) | |
| Missing | 2 (<1) | 15 ( 2) | |
| Donor-recipient ABO mismatch | 0·94 | ||
| Matched | 155 (42) | 381 (41) | |
| Minor mismatch | 90 (24) | 226 (24) | |
| Major mismatch | 91(25) | 242 (26) | |
| Bidirectional mismatch | 34 ( 9) | 79 ( 9) | |
| Missing | 0 | 1 (<1) | |
| Transplant-related | |||
| Year of transplant | <0·001 | ||
| 2001–2004 | 184 (50) | 228 (25) | |
| 2005–2007 | 116 (31) | 407 (44) | |
| 2008–2011 | 70 (19) | 294 (32) | |
| Conditioning intensity * TBI usage | <0·001 | ||
| MA - TBI | 203 (55) | 346 (37) | |
| MA - No TBI | 129 (35) | 349 (38) | |
| RIC - TBI | 2 (<1) | 34 ( 4) | |
| RIC - No TBI | 36 (10) | 200 (22) | |
| ATG / Campath used in conditioning regimen or GVHD | 0·83 | ||
| prophylaxis? | |||
| Yes | 139 (38) | 343 (37) | |
| No | 231(62) | 586 (63) | |
| Number of patients | 370 | 929 | |
| GVHD prophylaxis | <0·001 | ||
| Tac + MMF ± others | 18 ( 5) | 139 (15) | |
| Tac + MTX ± others | 184 (50) | 428 (46) | |
| Tac ± others | 35 ( 9) | 115(12) | |
| CsA + MMF ± others | 4 ( 1) | 35 ( 4) | |
| CsA + MTX ± others | 99 (27) | 138 (15) | |
| CsA ± others | 6 (2) | 32 (3) | |
| Others | 24 (6) | 42 (5) | |
| Follow-up of survivors, months, median (range) | 97 (4–156) | 73 (6–168) | <0·001 |
Abbreviations: TNC = total nucleated cell, CY = cyclophosphamide, BU = busulfan, FLUD = fludarabine, LPAM = melphalan, MA = myeloablative, RIC = reduced intensity conditioning, Tac = Tacrolimus, CsA = Cyclosporine, MMF = mycophenolate mofetil, MTX = methotrexate.
: Disease status is defined as: Early = AML/ALL (CR1); CML (CP1); MDS (RA/RARS/pre-HCT marrow blasts <5%) Intermediate = AML/ALL (>=CR2); CML (AP or >=CP2) Advanced = AML/ALL (REL/PIF); CML (BP); MDS (RAEB/RAEB-t/CMML or marrow blasts >=5%)
Multivariate Analysis for GRFS and OS
GRFS was superior in BM when compared to PB recipients of an 8/8-HLA URD when evaluated >5 months post-transplant with a HR of 1·62 (95% CI 1–38-1·89, p<0·0001) (Figure 1a, Table 2a). However, before 5-months (when chronic GVHD is less common), GRFS was similar in PB and BM recipients [hazard ration (HR) of 0·92; 95% confidence interval (CI) 0·83–1·02; p- value=0·0.10].In addition to graft source (beyond 5 months), patient age > 50 years, donor age >32 years, lack of T-cell antibody, a diagnosis of CML, being less than 1-year from diagnosis to transplant and transplant during the interval of 2001–2004 resulted in worse GRFS (Table 2a). In the 7/8-HLA cohort, PB grafts were associated with worse GRFS (HR= 1·15; 95% CI 1·00–1·33; p-value 0·0476) (Figure 1b, Table 2b). Additional factors for worse GRFS in the 7/8-HLA cohort included lack of T-cell antibody, a diagnosis of CML or MDS, having advanced disease and use of a mycophenolate mofetil-containing GVHD prophylaxis regimen (Table 2b).
Figures 1a-d:
Graft-versus-Host Disease, Relapse-Free Survival in (a) 8/8-HLA matched and (b) 7/8-HLA matched unrelated donor recipients and Overall Survival in (c) 8/8-HLA matched and (d) 7/8-HLA matched unrelated donor recipients
Table 2a.
Predictors for GVHD-free, Relapse-free Survival (GRFS) in recipients of 8/8 HLA-matched unrelated donor transplants*.
| Risk Factor | N (%) | HR (95% CI) | P-value |
|---|---|---|---|
| Graft Source (≤ 5 months) | 0·0977 | ||
| Bone Marrow | 1059 (28) | 1·00 | |
| Peripheral Blood | 2765 (72) | 092 (0·83–1·02) | 0·0977 |
| Graft Source (>5 months) | < 0001 | ||
| Bone Marrow | 418 (27) | 1·00 | |
| Peripheral Blood | 1122 (73) | 162 (1·38–1·89) | < 0001 |
| Patient age | 0·0074 | ||
| 18–29 | 712 (19) | 1·00 | |
| 30–39 | 634 (16) | 0·97 (0·86–1·09) | 0·6257 |
| 40–49 | 847 (22) | 1·06 (0·95–1·18) | 0·3299 |
| 50+ | 1631 (43) | 1·15 (1·03–1·27) | 0·0104 |
| ATG / Campath | < 0001 | ||
| Yes | 1191 (31) | 1·00 | |
| No | 2633 (69) | 1·35 (1·23–1·48) | < 0001 |
| Disease | 0·0760 | ||
| AML | 2155 (56) | 1·00 | |
| ALL | 658 (17) | 1·02 (0·92–1·13) | 0·7076 |
| CML | 467 (12) | 1·17 (1·04–1·32) | 0·0096 |
| MDS | 544 (15) | 1·02 (0·91–1·13) | 0·7572 |
| Donor age | 0·0375 | ||
| 18 – 32 | 1966 (51) | 1·00 | |
| 33 – 49 | 1644 (43) | 1·08 (1·01–1·16) | 0·0360 |
| 50 + | 214 ( 6) | 1·16 (0·99–1·35) | 0·0596 |
| Time from diagnosis to transplant (month) | 0·0449 | ||
| < 6 | 1495 (39) | 1·00 | |
| 6–12 | 1018 (27) | 1·01 (0·93–1·11) | 0·7502 |
| > 12 | 1312 (34) | 0·90 (0·81–0·99) | 0·0384 |
| Year of transplant | 0·0001 | ||
| 2001–2004 | 959 (25) | 1·00 | |
| 2005–2007 | 1628 (43) | 0·90 (0·82–0·98) | 0·0221 |
| 2008–2011 | 1237 (32) | 0·80 (0·72–0·88) | < 0001 |
ALL, Acute Lymphoid Leukemia; AML, Acute Myeloid Leukemia; CML, Chronic Myeloid Leukemia; MDS, Myelodysplastic Syndrome
: Stratified variables: disease stage, GVHD prophylaxis, Karnofsky score, Transplantation centers.
62 patients have GRFS endpoint missing; 6 patients have ‘disease stage’ missing; 9 patients have ‘Time from diagnosis to transplant’ missing.
Table 2b.
Predictors for GVHD-free, Relapse-free Survival (GRFS) in recipients of 7/8 HLA-matched unrelated donor transplants*.
| Risk Factor | N (%) | HR (95% CI) | p-value |
|---|---|---|---|
| Graft Source | 0·0476 | ||
| Bone marrow | 361 (28) | 1·00 | |
| Peripheral Blood | 915(72) | 1·15 (1·00–1·33) | 0·0476 |
| ATG / Campath | < 0001 | ||
| Yes | 469 (37) | 1·00 | |
| No | 807(63) | 1·36 (1·19–1·56) | < 0001 |
| Disease | 0·0378 | ||
| AML | 718(56) | 1·00 | |
| ALL | 263 (21) | 1·09 (0·93–1·27) | 0·2915 |
| CML | 168 (13) | 1·21 (1·00–1·46) | 0·0496 |
| MDS | 127 (10) | 1 27 (1·04–1·56) | 0·0203 |
| Disease Status | < 0001 | ||
| Early | 556 (44) | 1·00 | |
| Intermediate | 350 (27) | 1·21 (1·04–1·40) | 0·0120 |
| Advanced | 370 (29) | 1·54 (1·34–1·78) | < 0001 |
| GVHD Prophylaxis | 0·0015 | ||
| CNI + MMF ± others | 190 (15) | 1·00 | |
| CNI + MTX ± others | 834 (65) | 0·72 (0·60–0·87) | 0·0005 |
| Others | 252 (20) | 0·80 (0·65–0·99) | 0·0413 |
ALL, Acute Lymphoid Leukemia; AML, Acute Myeloid Leukemia; CML, Chronic Myeloid Leukemia; CNI, Calcineurin Inhibitor; MDS, Myelodysplastic Syndrome; MMF, Mycophenolate Mofetil; MTX, Methotrexate *Transplantation centers.
: 19 patients have GRFS endpoint missing; 5 patients have ‘disease stage’ missing
For patients receiving an 8/8-HLA matched URD, within the first 2-months, OS was significantly better in those who received PB vs. BM with a HR of 0·63 (95% CI 0·50–0·79, p- value of <0·0001). Over time, the survival advantage of PB disappeared such that at 2-years OS became statistically comparable between graft sources (p=0·27) (Table 4a and Figure 1c). Proportionality assumptions were maintained for the 7/8-HLA cohort with OS at 2-years not differing for PB or BM recipients (HR 1·01; 95% CI of 0·86–1·19, p= 0·91) (Table 4b, Figure 1d).
Table 4a.
Predictors for Overall Survival (OS) in recipients of 8/8 HLA-matched unrelated donor transplants*.
| Risk Factor | N (%) | HR (95% CI) | P-value |
|---|---|---|---|
| Graft Source (≤ 2 months) | < 0001 | ||
| Bone Marrow | 1082 (28) | 1·00 | |
| Peripheral Blood | 2804(72) | 0·63 (0·50–0·79) | < 0001 |
| Graft Source (>2 months) | |||
| Bone Marrow | 939 (27) | 1·00 | |
| Peripheral Blood | 2546 (73) | 1·10 (0·97–1·24) | 0·1427 |
| Donor-Recipient ABO type match | 0·0018 | ||
| Matched | 1713 (44) | 1·00 | |
| Minor mismatch | 971 (25) | 1·11 (0–99-1–24) | 0·0763 |
| Major mismatch | 907 (23) | 1·24 (1·11–1·39) | 0·0001 |
| Bidirectional mismatch | 295 (8) | 1·05 (0·88–1·24) | 0·5999 |
| Patient age | < 0001 | ||
| 18–29 | 724 (19) | 1·00 | |
| 30–39 | 646 (17) | 1·02 (0·87–1·20) | 0·7766 |
| 40–49 | 860 (22) | 1·27 (1·09–1·47) | 0·0015 |
| 50+ | 1656 (43) | 1·64 (1·43–1·89) | < 0001 |
| Donor-Recipient CMV match | 0·0233 | ||
| +/+ | 1141 (29) | 1·00 | |
| +/− | 1428 (37) | 1·20 (1·08–1·34) | 0·0012 |
| −/+ | 413 (11) | 1·18 (1·01–1·38) | 0·0410 |
| −/− | 832 (21) | 1·14 (1·00–1·30) | 0·0442 |
| Missing | 72 (2) | 1·15 (0·83–1·59) | 0·4149 |
| Conditioning regimen intensity & TBI | < 0001 | ||
| usage | |||
| MA - TBI | 1379 (35) | 1·00 | |
| MA - no TBI | 1591 (41) | 0·78 (0·70–0·87) | < 0001 |
| RIC - TBI | 105 (3) | 0·92 (0·68–1·23) | 0·5595 |
| RIC - no TBI | 811 (21) | 0·76 (0·66–0·88) | 0·0002 |
| Time from diagnosis to transplant | 0·0193 | ||
| (month) | |||
| < 6 | 1515 (39) | 1·00 | |
| 6–12 | 1035 (27) | 1·11 (0·99–1·24) | 0·0746 |
| > 12 | 1336 (34) | 0·93 (0·82–1·05) | 0·2482 |
| Karnofsky score | < 0001 | ||
| < 90 | 1226 (32) | 1·00 | |
| ≥ 90 | 2377 (61) | 0·74 (0·67–0·82) | < 0001 |
| Missing | 283 (7) | 0·80 (0·66–0·97) | 0·0218 |
ALL, Acute Lymphoid Leukemia; AML, Acute Myeloid Leukemia; CML, Chronic Myeloid Leukemia; MDS, Myelodysplastic Syndrome
: Stratified variables: disease stage, GVHD prophylaxis, year of transplant, donor-recipient gender match, Transplantation centers.
9 patients have ‘Time from diagnosis to transplant’ missing.
Table 4b.
Predictors for Overall Survival (OS) in recipients of 7/8 HLA-matched unrelated donor transplants.
| Risk Factor | N (%) | HR (95% CI) | p-value |
|---|---|---|---|
| Graft Source | 0·9070 | ||
| Bone marrow | 369 (29) | 1·00 | |
| Peripheral Blood | 925 (7l) | 1·01 (0·86–1·19) | 0·9070 |
| Patient Age | < 0001 | ||
| 18–29 | 269 (21) | 1·00 | |
| 30–39 | 242 (l9) | 1·41 (1·12–1·78) | 0 0038 |
| 40–49 | 328 (25) | 1·58 (1·27–1·97) | < 0001 |
| 50+ | 455 (35) | 1·68(1·35–2·08) | < 0001 |
| Disease | 0·0003 | ||
| AML | 730 (56) | 1·00 | |
| ALL | 265 (20) | 1·44 (1·19–1·73) | 0·0002 |
| CML | 169(l3) | 0·99 (0·78–1·25) | 0·9117 |
| MDS | 130 (10) | 1·34 (1·06–1·69) | 0·0158 |
| Disease Stage | < 0001 | ||
| Early | 565 (44) | 1·00 | |
| Intermediate | 355 (27) | 1·42 (1·19–1·70) | 0·0001 |
| Advanced | 374 (29) | 1·94 (1·64–2·30) | < 0001 |
| Donor Age | 0·0199 | ||
| 18–32 | 506 (39) | 1·00 | |
| 33–49 | 665 (51) | 1·10 (0·95–1·28) | 0·2009 |
| 50+ | 123 (10) | 1·41 (1·10–1·79) | 0·0057 |
| Karnofsky score | 0·0035 | ||
| < 90 | 435 (34) | 1·00 | |
| ≥ 90 | 770 (60) | 0·77 (0·66–0·90) | 0·0009 |
| Missing | 89 (7) | 0·82 (0·61–1·11) | 0·1996 |
ALL, Acute Lymphoid Leukemia; AML, Acute Myeloid Leukemia; CML, Chronic Myeloid Leukemia; CNI, Calcineurin Inhibitor; MDS, Myelodysplastic Syndrome; MMF, Mycophenolate Mofetil; MTX, Methotrexate, Transplantation centers. 5 patients have ‘Disease Stage’ missing
Composite Factors: Multivariate Analysis for Cumulative Incidences for Acute GVHD Grades III-IV, Chronic GVHD requiring IST, Relapse and NRM within 2 years post-transplant in Recipients of BM or PB grafts.
Acute GVHD grades III-IV did not statistically differ in PB and BM recipients (HR 1·08, 95% CI 0·89–1·30, p-value 0·44). In contrast, the cumulative incidence of chronic GVHD requiring IST was worse with receipt of PB (> 5 months HR of 1·80; 95% CI of 1·50–2·15, p-value of <0·0001). Additional factors associated with increased chronic GVHD requiring IST included the lack of T-cell antibody, diagnosis of CML, mycophenolate mofetil-containing GVHD prophylaxis and female donors in male recipients. Within the first 2 months of HCT, PB was protective for NRM (HR 0·49; 95% CI 0·39–0·63, p-value of <0·0001); however at 2-years post-transplant NRM was similar in BM and PB recipients (27 vs. 24%, p-value 014) (Table 3) while at 5-years post-transplant, NRM was worse in recipients of PB (30 vs. 34%, p-value 0·0087).
Table 3.
Adjusted Cumulative Incidence and Survival Rates in Recipients of 8/8 HLA-matched unrelated donor transplants.*
| Outcomes | Bone marrow Probability (95% CI) |
Peripheral blood Probability (95% CI) |
Pointwise p-value |
|---|---|---|---|
| Relapse @ 1 year | 29 (26–31) | 28 (27–30) | 0·91 |
| Relapse @ 2 years | 33 (30–36) | 33 (31–35) | 0·97 |
| Relapse @ 5 years | 37 (34–40) | 37 (35–39) | 0·96 |
| NRM @ 1 year | 24 (21–27) | 20 (18–21) | 0·0047 |
| NRM @ 2 years | 27 (24–30) | 24 (23–26) | 0·14 |
| NRM @ 5 years | 30 (27–33) | 35 (33–37) | 0·016 |
| GRFS @ 1 year | 21 (19–23) | 15 (14–16) | < 0001 |
| GRFS @ 2 years | 16 (14–19) | 10 (8–11) | <0·0001 |
| GRFS @ 5 years | 14 (11–16) | 7(5–8) | <0·0001 |
| OS @ 1 year | 56 (53–59) | 61 (59–63) | 0·0030 |
| OS @ 2 years | 48 (45–51) | 51 (50–53) | 0·11 |
| OS @ 5 years | 42 (39–46) | 38 (36–40) | 0·022 |
Adjusted covariates for (1) Relapse: donor-recipient ABO type match, conditioning regimen intensity & TBI usage, GVHD prophylaxis, time from diagnosis to transplant, Karnofsky score, disease type, year of transplant; (2) NRM: patient age, donor age, donor-recipient ABO type match, donor-recipient CMV match, ATG/Campath use, conditioning regimen intensity & TBI usage, GVHD prophylaxis, Karnofsky score, disease type, disease stage, year of transplant; (3) GRFS: patient age, donor age, ATG/Campath use, time from diagnosis to transplant, GVHD prophylaxis, Karnofsky score, disease type, disease stage, year of transplant; (4) OS: patient age, donor-recipient ABO type match, donor-recipient CMV match, conditioning regimen intensity & TBI usage, time from diagnosis to transplant, Karnofsky score, disease stage, GVHD prophylaxis, year of transplant, donor- recipient gender match
The presence of chronic GVHD resulted in a higher risk for NRM in recipients of PB (HR 1·58; 95% CI: 1·20–2·08, P=0·0012). The impact of chronic GVHD on NRM in BM recipients was not significant (HR 1·29, 95% CI: 0·84–1·99, P=0·25). Relapse did not differ in BM or PB recipients (HR 0·93; 95% CI: 0·81–1·07, p-value 0·29).
In the 7/8-HLA matched cohort, acute GVHD III-IV did not differ for PB vs. BM (HR 1·14; 95% CI 0·87–1·49, p-value 0·32); whereas chronic GVHD requiring IST was more common in recipients of PB (HR 1·39; 95% CI of 1·12–1·73, p-value of 0·0031). Relapse rates did not differ (HR 0·92; 95% CI of 0·73–1·17, p-value 0·49) nor did NRM (HR 1·11; 95% CI of 0·89–1·39, p=0·3404).
Contributors to treatment failure (Grade III/IV acute GVHD, Chronic GVHD requiring IST, Relapse, Death in Remission) in Recipients of BM versus PB grafts
Figure 2 is a stacked plot analysis of causes of treatment failure (with respect to GRFS) in 8/8- HLA matched BM (a) and PB (b) and 7/8-HLA matched BM (c) and PB (d) recipients. Chronic GVHD requiring IST was the main cause of treatment failure for all cohorts however it was a more common cause of failure in recipients of PB vs. BM grafts among the 8/8-HLA (42% versus 31%, p-value<0·001) and 7/8-HLA cohort (35% versus 29% at 5-years post-HCT, p- value=0·02). In addition, death in remission contributed to failure more commonly in PB recipients when compared to BM in the 8/8-cohort (16% versus 11% at 5-years post-HCT, p- value<0·001). The contribution of relapse and acute GVHD III/IV as a cause of treatment failure did not differ based on graft source for either 8/8 or 7/8-HLA cohorts.
Figure 2a-d:
Causes of Treatment Failure with respect to Graft-versus-Host Disease, Relapse- Free Survival in recipients of an 8/8-HLA matched (a) bone marrow and (b) peripheral blood and 7/8-HLA matched (c) bone marrow and (d) peripheral blood allograft from unrelated donors
Long-term Outcomes: OS and GRFS at 5 years
In order to determine the impact that late complications (namely chronic GVHD) played with respect to long-term survival OS, the adjusted survival rates were estimated at 5-years and found to be significantly higher in recipients of BM (43%, 95% CI: 40–46) when compared to PB (38%, 95% CI: 36–40), p-value 0·014 in the 8/8-HLA cohort (Table 3). Additional factors impacting OS included donor-recipient ABO match, CMV- status, conditioning regimen, time from diagnosis to transplant and KPS (Table 4a, with predictors of DFS shown in supplemental Table 1a). Note that the multivariable Cox models in Table 4a also showed that there were different hazards between PB and BM for OS within the first 2 months following transplant (PB superior to BM) but no significant difference in hazards between PB and BM for OS beyond 2 months following transplant. The probabilities of OS for PB and BM at 5 years depend on the hazard functions of PB and BM over the entire time following transplant up until 5 years. Similar trends were seen for OS at 5 years in the 7/8-HLA cohort which did not reach statistical significance (p=0·33) (Table 4b, DFS shown in supplemental Table 1b). The cumulative incidence for GRFS at 5 years was almost twice as high for BM recipients as PB in both the 8/8 (14 vs. 7%, p<0 0001) and 7/8-HLA cohorts (9 vs. 4%, p<0·0001) (Table 3 and Figures 1a-d).
DISCUSSION
The first attempt to definitively address the question of graft source was the BMT CTN 0201 trial which enrolled its first patient in 2004, at which time PB had already surpassed BM as the predominant source in patients undergoing URD-HCT. In 2013, the primary results from BMT CTN0201 were published demonstrating comparable survival at 2 years for those randomized to PB and BM2. The failure to show a survival benefit in this trial as well as prior retrospective analyses13–19 may have contributed to PB remaining the predominant graft source in roughly 80% of all URD recipients despite significantly higher rates of chronic GVHD and the need for prolonged immunosuppressants.1 More recently, long-term patient-reported quality of life (QOL) data from the BMT CTN 0201trial were published. These results showed that at 5-years posttransplant, BM recipients reported significantly better QOL and were more likely to have returned to work when compared to PB; however, survival remained equivalent at 5-years.3 In contrast, our results demonstrate at 5-years post-transplant OS is superior in recipients of BM following 8/8-HLA MUD transplant. It should be noted that the 0201 trial was powered to detect a 12.5% absolute survival difference at 2 -years. In contrast, our sample size was roughly 10 times larger and allowed for detection of a smaller absolute difference which only became apparent with sufficient follow-up to allow for the impact of late mortality in patients with chronic GVHD to be appreciated. The demonstration of improved survival and GRFS at 5-years post- transplant with the use of BM grafts calls into question current practice favoring PB in recipients 8/8-HLA URD’s transplant.
A clear survival advantage for BM was not demonstrated in the 7/8-HLA cohort in our analysis. It is important to note that graft failure rates were comparable in the 7/8 and 8/8 cohorts. As such, theoretical concerns for higher rejection when using a 7/8 HLA-matched donor justifying selection of PB over BM is not supported by our data (data not shown). Despite the lack of survival advantage for BM in the 7/8-HLA cohort, GRFS was superior due to a lower risk for chronic GVHD requiring IST. With comparable survival but worse GRFS, the routine use of PB grafts in recipients of 7/8 HLA-matched URD’s should be reconsidered as well.
Our report highlights the challenges when using composite endpoints. GRFS has been recently suggested as an important endpoint for trials evaluating GVHD prophylaxis strategies.4 The advantage of GRFS is it assesses the impact of clinically important GVHD events while capturing competing risks of relapse and NRM. The challenges with this endpoint are that when comparing treatment strategies, events can occur at different time periods resulting in major differences over time. This was seen here; patients who received BM had higher rates of early NRM (primarily in the first two months post-transplant) whereas patients who received PB suffered from higher rates of chronic GVHD (a late event) which ultimately contributed to higher long-term rates of NRM. Therefore, it is essential that GRFS is assessed at a time point sufficiently long to account for all of these competing events. In addition, GRFS fails to account for patients who develop acute GVHD Grade III/IV or chronic GVHD requiring IST but subsequently recover. Adoption of a “current” GRFS defined as survival without evidence of relapse, grade III-IV acute GVHD or chronic GVHD requiring IST at the point of last follow-up would be one method to reflect the long-term impact of these events.20
There are a number of limitations in our analysis. Importantly, 75% of the patients received myeloablative conditioning. While conditioning intensity was accounted for in the modeling of all outcomes, it is not clear whether graft source has the same impact in less intense conditioning regimens. The question of graft source in non-TBI based reduced-intensity conditioned unrelated allografts has been previously studied by the CIBMTR which concluded outcomes were similar with respect to OS, acute and chronic GVHD in recipients of BM and PB grafts; however, GRFS was not examined in this report.13 As in our analysis, this earlier CIBMTR study also found a strong interaction with GVHD prophylaxis regimens with worse outcomes when mycophenolate mofetil-based prophylaxis was used. In contrast, a separate analysis from the European Registry found the use of PB to result in higher rates of acute GVHD, chronic GVHD and NRM but lower relapse rates in recipients of URD’s following reduced intensity conditioning when compared to BM with no difference in survival.14 However, follow-up was considerably shorter in the European analysis and roughly half had missing data regarding GVHD prophylaxis which may account for differences in these two reports.
An additional limitation in our analysis was the absence of newer prophylaxis regimens using post-transplant cyclophosphamide based approaches which have demonstrated low rates of chronic GVHD21,22. Recently, a three-arm randomized trial of prophylaxis regimens in the setting of reduced-intensity conditioning followed by PB HCT was presented in abstract form demonstrating improved GRFS and lower rates of chronic GVHD requiring 1ST in recipients of post-transplant cyclophosphamide, tacrolimus and mycophenolate arm when compared to a contemporary CIBMTR matched cohort23. In this trial, participants could only receive PB grafts and whether further reductions in the rate of chronic GVHD would have been seen had the use of BM grafts been allowed remains untested.
Finally, our analysis is limited to recipients of URD HCT. The question of graft source in recipients of other donor sources has been an area of interest. An advantage of BM over PB in the setting of HLA-matched sibling HCT with respect to GRFS has recently been reported in a two-center retrospective analysis;6 while a three center prospective, randomized trial showed comparable OS but significant reduction in relapse in those randomized to PB HLA-matched sibling transplants.24 Graft source following haploidentical HCT using GVHD prophylaxis with post-transplant cyclophosphamide remains an area of debate with limited data suggesting comparable outcomes for PB and BM recipients in the setting of non-myeloablative conditioning.25
In conclusion, the use of 8/8 HLA-matched URD BM when compared to PB allografts results in improvement in long-term OS and GRFS in adults with AL, CML or MDS and should be considered as the graft source of choice if available. When PB is utilized, strategies to minimize the impact of chronic GVHD should be employed.
Supplementary Material
Highlights:
Peripheral Blood (PB) is the most common graft in adults undergoing an unrelated donor (MUD) allograft.
GVHD, Relapse Free Survival at 2 years is superior for bone marrow (BM) vs. PB for adult MUD patients.
Overall Survival at 5 years is superior in recipients of BM vs. PB from 8/8-HLA MUD.
Long-term survival advantage for BM is due to lower risk for death while in remission.
Acknowledgments
The CIBMTR is supported primarily by Public Health Service Grant/Cooperative Agreement 5U24CA076518 from the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI) and the National Institute of Allergy and Infectious Diseases (NIAID); a Grant/Cooperative Agreement 4U10HL069294 from NHLBI and NCI; a contract HHSH250201200016C with Health Resources and Services Administration (HRSA/DHHS); two Grants N00014–17-1–2388 and N0014–17-1–2850 from the Office of Naval Research; and grants from *Actinium Pharmaceuticals, Inc.; *Amgen, Inc.; *Amneal Biosciences; *Angiocrine Bioscience, Inc.; Anonymous donation to the Medical College of Wisconsin; Astellas Pharma US; Atara Biotherapeutics, Inc.; Be the Match Foundation; *bluebird bio, Inc.; *Bristol Myers Squibb Oncology; *Celgene Corporation; Cerus Corporation; *Chimerix, Inc.; Fred Hutchinson Cancer Research Center; Gamida Cell Ltd.; Gilead Sciences, Inc.; HistoGenetics, Inc.; Immucor; *Incyte Corporation; Janssen Scientific Affairs, LLC; *Jazz Pharmaceuticals, Inc.; Juno Therapeutics; Karyopharm Therapeutics, Inc.; Kite Pharma, Inc.; Medac, GmbH; MedImmune; The Medical College of Wisconsin; *Mediware; *Merck & Co, Inc.; *Mesoblast; MesoScale Diagnostics, Inc.; Millennium, the Takeda Oncology Co.; *Miltenyi Biotec, Inc.; National Marrow Donor Program; *Neovii Biotech NA, Inc.; Novartis Pharmaceuticals Corporation; Otsuka Pharmaceutical Co, Ltd. - Japan; PCORI; *Pfizer, Inc; *Pharmacyclics, LLC; PIRCHE AG; *Sanofi Genzyme; *Seattle Genetics; Shire; Spectrum Pharmaceuticals, Inc.; St. Baldrick’s Foundation; *Sunesis Pharmaceuticals, Inc.; Swedish Orphan Biovitrum, Inc.; Takeda Oncology; Telomere Diagnostics, Inc.; and University of Minnesota. The views expressed in this article do not reflect the official policy or position of the National Institute of Health, the Department of the Navy, the Department of Defense, Health Resources and Services Administration (HRSA) or any other agency of the U.S. Government.
Footnotes
Corporate Members
DISCLOSURE OF CONFLICT OF INTEREST
The authors do not have any conflicts specific to this study to disclose.
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.
REFERENCES
- 1.Current Uses and Outcomes of Hematopoietic Cell Transplantation (HCT): CIBMTR Summary Slides. 2016. http://www.cibmtr.org.
- 2.Anasetti C, Logan BR, Lee SJ, et al. Peripheral-blood stem cells versus bone marrow from unrelated donors. The New England journal of medicine. 2012;367(16):1487–1496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lee SJ, Logan B, Westervelt P, et al. Comparison of Patient-Reported Outcomes in 5-Year Survivors Who Received Bone Marrow vs Peripheral Blood Unrelated Donor Transplantation: Long-term Follow-up of a Randomized Clinical Trial. JAMA oncology. 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bolanos-Meade J, Koreth J, Reshef R. A multi-center phase II trial randomizing novel approaches for graft-versus-host disease prevention compared to contemporary controls (1203). 2015. [Google Scholar]
- 5.Sankoh AJ, Li H, D’Agostino RB Sr., Use of composite endpoints in clinical trials. Statistics in medicine. 2014;33(27):4709–4714. [DOI] [PubMed] [Google Scholar]
- 6.Mehta RS, Peffault de Latour R, DeFor TE, et al. Improved graft-versus-host disease-free, relapse-free survival associated with bone marrow as the stem cell source in adults. Haematologica. 2016;101(6):764–772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Spellman S, Setterholm M, Maiers M, et al. Advances in the selection of HLA-compatible donors: refinements in HLA typing and matching over the first 20 years of the National Marrow Donor Program Registry. Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation. 2008;14(9 Suppl):37–44. [DOI] [PubMed] [Google Scholar]
- 8.Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on Acute GVHD Grading. Bone marrow transplantation. 1995;15(6):825–828. [PubMed] [Google Scholar]
- 9.Lee SJ, Klein JP, Barrett AJ, et al. Severity of chronic graft-versus-host disease: association with treatment-related mortality and relapse. Blood. 2002;100(2):406–414. [DOI] [PubMed] [Google Scholar]
- 10.Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. The American journal of medicine. 1980;69(2):204–217. [DOI] [PubMed] [Google Scholar]
- 11.Cox DR. Regression Models and Life-Tables. Journal of the Royal Statistical Society Series B (Methodological). 1972;34(2):187–220. [Google Scholar]
- 12.Ringden O, Remberger M, Runde V, et al. Peripheral blood stem cell transplantation from unrelated donors: a comparison with marrow transplantation. Blood. 1999;94(2):455–464. [PubMed] [Google Scholar]
- 13.Eapen M, Logan BR, Horowitz MM, et al. Bone marrow or peripheral blood for reduced-intensity conditioning unrelated donor transplantation. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2015;33(4):364–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nagler A, Labopin M, Shimoni A, et al. Mobilized peripheral blood stem cells compared with bone marrow as the stem cell source for unrelated donor allogeneic transplantation with reduced-intensity conditioning in patients with acute myeloid leukemia in complete remission: an analysis from the Acute Leukemia Working Party of the European Group for Blood and Marrow Transplantation. Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation. 2012;18(9):1422–1429. [DOI] [PubMed] [Google Scholar]
- 15.Eapen M, Logan BR, Appelbaum FR, et al. Long-term survival after transplantation of unrelated donor peripheral blood or bone marrow hematopoietic cells for hematologic malignancy.Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation. 2015;21(1):55–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Holtick U, Albrecht M, Chemnitz JM, et al. Bone marrow versus peripheral blood allogeneic haematopoietic stem cell transplantation for haematological malignancies in adults. The Cochrane database of systematic reviews. 2014(4):Cd010189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Shin SH, Yoon JH, Yahng SA, et al. PBSC vs BM grafts with myeloablative conditioning for unrelated donor transplantation in adults with high-risk ALL. Bone marrow transplantation. 2014;49(6):773–779. [DOI] [PubMed] [Google Scholar]
- 18.Shin SH, Kim JH, Jeon YW, et al. Similar outcomes of peripheral blood stem cells vs. bone marrow for human leukocyte antigen-matched unrelated donor transplantation in adult patients with acute myeloid leukemia using risk-adapted graft-versus-host disease prophylaxis. European journal of haematology. 2014;93(1):19–28. [DOI] [PubMed] [Google Scholar]
- 19.Auberger J, Clausen J, Kircher B, Kropshofer G, Lindner B, Nachbaur D. Allogeneic bone marrow vs. peripheral blood stem cell transplantation: a long-term retrospective single-center analysis in 329 patients. European journal of haematology. 2011;87(6):531–538. [DOI] [PubMed] [Google Scholar]
- 20.Craddock C, Szydlo RM, Klein JP, et al. Estimating leukemia-free survival after allografting for chronic myeloid leukemia: a new method that takes into account patients who relapse and are restored to complete remission. Blood. 2000;96(1):86–90. [PubMed] [Google Scholar]
- 21.Ruggeri A, Labopin M, Bacigalupo A, et al. Post-transplant cyclophosphamide for graft-versus- host disease prophylaxis in HLA matched sibling or matched unrelated donor transplant for patients with acute leukemia, on behalf of ALWP-EBMT. Journal of hematology & oncology. 2018;11(1):40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mielcarek M, Furlong T, O’Donnell PV, et al. Posttransplantation cyclophosphamide for prevention of graft-versus-host disease after HLA-matched mobilized blood cell transplantation. Blood. 2016;127(11):1502–1508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bolaños-Meade J LBA1: Novel Approaches for Graft-Versus-Host Disease (GvHD) Prophylaxis: Primary Results of Progress I Multicenter Trial of Matched Allogeneic Hematopoietic Cell Transplantation (alloHCT) Using Reduced Intensity Conditioning (RIC) BMT CTN 1203 BMT Tandem Meetings 2018; https://bmt.confex.com/tandem/2018/meetingapp.cgi/Paper/11767. [Google Scholar]
- 24.Mielcarek M, Storer B, Martin PJ, et al. Long-term outcomes after transplantation of HLA- identical related G-CSF-mobilized peripheral blood mononuclear cells versus bone marrow. Blood. 2012;119(11):2675–2678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.O’Donnell PV, Eapen M, Horowitz MM, et al. Comparable outcomes with marrow or peripheral blood as stem cell sources for hematopoietic cell transplantation from haploidentical donors after non-ablative conditioning: a matched-pair analysis. Bone marrow transplantation. 2016;51(12):1599–1601. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.