Abstract
Background:
Donors other than matched siblings and low intensity conditioning regimens are increasingly used to restore normal hematopoiesis for sickle cell disease. We aimed to compare the relative risk of donor type and conditioning regimen intensity on transplant-outcomes.
Methods:
Between 15/01/2008and 28/12/2017, 910 patients < 50 years underwent hematopoietic cell transplant from HLA matched sibling (558 of 910; 61%), haploidentical related (137 of 910; 15%), matched unrelated (111 of 910; 12%) and mismatched unrelated (104 of 910; 11%) donors. . Data were reported to the Center for International Blood and Marrow Transplant Research (CIBMTR) from 90 US centers. The main outcome was event-free survival. The effect of donor type and conditioning regimen intensity was studied using Cox regression models.
Findings:
The median follow-up after HLA-matched sibling, haploidentical related, HLA-matched and mismatched unrelated donor transplants were 36 months (IQR 18-60), 25 months (IQR 12-48), 37 months (IQR 23-60) and 47 months (24-72), respectively. Event-free survival was lower in recipients aged ≥13 years (hazard ratio [HR] 1.74, 95% CI 1.24-2.45, p=0.0014), those transplanted with grafts from haploidentical relative (HR 5.30, 95% CI 3.17-8.86, p<0.0001) matched unrelated (HR 3.71, 95% CI 2.39-5.75, p<0.0001) and mismatched unrelated (HR 4.34, 95% CI 2.58-7.32, p<0.0001) donors compared to matched siblings and those conditioned with reduced intensity (HR 1.97, 95% CI 1.15-3.36, p=0.013) compared to nonmyeloablative regimens. Event-free survival did not differ between myeloablative and nonmyeloablative regimens (HR 1.57, 95% CI 0.95-2.61, p=0.079).
Interpretation:
Event-free survival was highest in children (≤13 years old) and with an HLA matched sibling donor. For those without a matched sibling, the data does not favor one alternative donor over another.
INTRODUCTION
Sickle cell disease (SCD) is the most common inherited hemoglobinopathy and occurs in 1 in 500 African-American and 1 in 1000 – 1400 Hispanic-American births. For children with access to modern health care, overall survival by 18 years of age is 85.6%.1 The life expectancy of adults with sickle cell disease is shortened by at least two decades compared to the general population.2,3 Risk factors associated with mortality in adults include age, gender, elevated tricuspid valve regurgitation velocity, intensity of hemolytic anemia and elevated ferritin, creatinine and aspartate transaminase. 3,4 Hematopoietic cell transplantation is potentially curative for sickle cell disease but the treatment itself is associated with substantial risks which can significantly limit its use. While the probability of survival is approximately 90% at 5 years in recipients of an HLA-matched sibling transplant,5–8 access is limited by donor availability. As a result, alternative donor sources have been explored. Because of the rarity of HLA haplotypes observed in patients of African-descent, matched unrelated donors are infrequent9, often necessitating the use of grafts from haploidentical related or mismatched unrelated donors.10–13 Treatment failure after mismatched related donor transplants is primarily attributed to graft failure10, and after matched unrelated donor transplantation, to chronic graft-versus-host disease (GVHD).11
Data reported to the Center for International Blood and Marrow Transplant Research (CIBMTR) indicate increasing numbers of alternative donor transplantation, increasingly utilized for young adults and conditioning regimens that are not full intensity (i.e., myeloablation) in the United States. Therefore, the current study sought to compare the relative risks of donor type and conditioning regimen intensity on outcomes after allogeneic transplantation for sickle cell disease in a more contemporary period (2008-2017) during which alternative donors and reduced intensity and nonmyeloablative conditioning regimens were increasingly used for transplantation.
PATIENTS AND METHODS
Study design and patient characteristics
The CIBMTR is a working group of >300 transplant centers that contribute data on consecutive allogeneic and autologous transplants. Patients are followed longitudinally until death or lost to follow-up. Accuracy of data reported to the CIBMTR and compliance are monitored by on-site audits. Consent is sought from patients and/or their legal guardians for research. The Institutional Review Board of the National Marrow Donor Program approved the study.
Nine hundred and ninety-six patients with SCD (Hb SS or Hb Sβ-thalassemia) aged less than 50 years received their first allogeneic HCT in the U.S. between 15/01/2008 and 28/12/2017 at 90 centers. With only 7 patients aged ≥50 years we were unable to study a group more likely to experience transplant-related complications including death. Nine hundred and ten patients (910 of 996, 91%) were eligible for this study (Figure 1). Excluded were patients from two transplant centers that failed data accuracy audit (43 of 996, 4%), uncommon conditioning regimens (10 of 996, 1%; treosulfan-containing [n=8] and TBI dose ≥1000 cGy [n=2]), did not receive GVHD prophylaxis (13 of 996, 1%) and inadequate follow-up after transplantation (20 of 996, 2%). Treosulfan is an investigational agent in the US and may be used only under an IND (investigational new drug) and high dose TBI (≥1000 cGy) regimens are rarely used for SCD. Twenty percent of patients (185 of 910) were included in earlier reports (HLA-matched sibling [111 of 910, 12%], haploidentical relative [39 of 910, 4%], HLA-matched unrelated [28 of 910, 3%], and HLA-mismatched unrelated [7 of 910, 0.7%]).7,8,10–13
Figure 1.
Patient flow diagram
Outcomes
The main outcome was event-free survival (death or graft failure were considered events). Graft failure was defined as failure to achieve absolute neutrophil recovery (ANC) ≥0⋅5 × 109/L for 3 consecutive days, or ANC decline to <0⋅5 × 109/L without recovery after having achieved ANC ≥0⋅5 × 109/L, or myeloid donor chimerism (<5%), or second transplant.14 Other outcomes studied were overall survival (death from any causes was considered an event), and acute and chronic GVHD graded using standard crietria.15,16 GVHD is an immunologically mediated complication of transplantation affecting multiple organs. Acute GVHD occurs primarily within the first 3 months after transplantation and chronic GVHD, thereafter. Occurrence of grade II-IV acute GVHD and any chronic GVHD were considered events.
Statistical Methods
The characteristics of patients by donor type were compared using the chi-square test for categorical variables. The probabilities of graft failure, acute and chronic GVHD were calculated using the cumulative incidence estimator to accommodate competing risks (death).17 Risk factors associated with event-free survival, overall survival, graft failure, acute and chronic GVHD were examined using the Cox proportional hazards model.18 The probabilities of event-free and overall survival were generated from final Cox regression models.19 Surviving patients were censored at last follow-up. Variables considered included age, sex, performance score, comorbidity index, recipient cytomegalovirus serostatus, donor type, conditioning regimen intensity, graft type and transplant period (Table 1). Age was treated as a binary variable (≤12 versus 13-49 years). The age cutpoint at below 13 years was determined statistically using the minimum p-value approach. To determine the optimal cutpoint, we used a series of two-sample tests for the multiple possible candidate dichotomizations of age. For each candidate cutpoint, an appropriate Cox model with a single binary covariate for age was constructed and the p-value for the Wald test was obtained. The optimal cutpoint is defined as that candidate cutpoint with the smallest p-value. All variables met the assumption of proportional hazards and there were no first order interactions between the variables held in Cox models. The level of significance set for the study was <0.05. All p-values are two-sided and all analyses were done using SAS version 9⋅4 (Cary, NC).
Table 1.
Patient and Transplant Characteristics
| Donor type | ||||
|---|---|---|---|---|
| Characteristic | HLA-matched sibling | Haploidentical relative | HLA-matched unrelated donor | Mismatched unrelated donor |
| Number | 558 | 137 | 111 | 104 |
| Genotype | ||||
| Hb SS | 534 (96%) | 132 (96%) | 106 (95%) | 95 (91%) |
| Hb Sβ thal | 24 (4%) | 5 (4%) | 5 (5%) | 9 (9%) |
| Age, transplant | ||||
| Median, years | 11 (<1 – 48) | 19 (3 – 47) | 14 (3 – 46) | 10 (2 – 44) |
| ≤ 10 years | 295 (53%) | 22 (16%) | 35 (32%) | 56 (53%) |
| 11 – 17 years | 139 (25%) | 38 (28%) | 54 (49%) | 34 (33%) |
| 18 – 29 years | 82 (15%) | 50 (36%) | 17 (15%) | 12 (12%) |
| 30 – 49 years | 42 (8%) | 27 (20%) | 5 (5%) | 2 (2%) |
| Sex | ||||
| Male | 311 (56%) | 74 (54%) | 53 (48%) | 51 (49%) |
| Female | 247 (44%) | 63 (46%) | 58 (52%) | 53 (51%) |
| CMV serostatus | ||||
| Positive | 288 (52%) | 64 (47%) | 54 (48%) | 49 (48%) |
| Negative | 269 (48%) | 72 (53%) | 57 (52%) | 55 (52%) |
| Not reported | 1 (<1) | 1 (<1) | — | — |
| Performance score | ||||
| ≤ 80 | 80 (14%) | 42 (31%) | 18 (16%) | 20 (19%) |
| 90 - 100 | 472 (85%) | 90 (66%) | 90 (81%) | 80 (77%) |
| Not reported | 6 (1%) | 5 (4%) | 3 (3%) | 4 (4%) |
| HCT-comorbidity index | ||||
| 0 – 2 | 404 (72%) | 76 (55%) | 79 (71%) | 79 (76%) |
| ≥ 3 | 154 (28%) | 61 (45%) | 32 (29%) | 25 (24%) |
| Graft type | ||||
| Bone marrow | 437 (78%) | 63 (46%) | 99 (89%) | 32 (31%) |
| Peripheral blood | 78 (14%) | 74 (54%) | 12 (11%) | 13 (12%) |
| Umbilical cord blood | 43 (8%) | — | — | 59 (56%) |
| HLA-match | ||||
| 8/8 match | 558 (100%) | — | 111 (100%) | — |
| 7/8 match | — | — | — | 55 (53%) |
| ≤6/8 match | — | 137 (100%) | — | 49 (46%) |
| Conditioning regimen | ||||
| Myeloablative | ||||
| Busulfan + Cy | 250 (45%) | 21 (15%) | 14 (13%) | 27 (26%) |
| Busulfan + Flu | 85 (15%) | 6 (4%) | 26 (24%) | 17 (16%) |
| Busulfan + Flu + TT | 2 (<1%) | 7 (5%) | 2 (2%) | 2 (2%) |
| Melphalan + Flu | 11 (2%) | — | 7 (6%) | 1 (<1%) |
| Reduced intensity | ||||
| Busulfan + Flu | 5 (<1%) | — | — | — |
| Melphalan + Flu | 110 (20%) | 1(<1%) | 41 (37%) | 24 (23%) |
| Melphalan + Flu + TT | 14 (3%) | 13 (9%) | 16 (14%) | 27 (27%) |
| Nonmyeloablative* | ||||
| TBI + Cy | 14 (3%) | 6 (4%) | — | — |
| TBI + Cy + Flu | 9 (2%) | 55 (40%) | 3 (3%) | 3 (3%) |
| TBI + Cy + Flu + TT | 1 (<1%) | 9 (7%) | — | — |
| TBI + melphalan | 2 (<1%) | — | 1 (<1%) | 2 (2%) |
| TBI + flu | — | 4 (3%) | 1 (<1%) | — |
| TBI alone | 55 (10%) | 15 (11%) | — | 1 (<1%) |
| In vivo T cell depletion | ||||
| Anti-thymocyte globulin | 249 (45%) | 95 (69%) | 29 (26%) | 25 (24%) |
| Alemtuzumab | 289 (52%) | 30 (22%) | 75 (68%) | 66 (64%) |
| None | 20 (3%) | 12 (9%) | 7 (6%) | 13 (12%) |
| GVHD prophylaxis | ||||
| Ex vivo T cell depletion | — | 10 (7%) | 1 (1%) | — |
| CD34 selection | 2 (<1%) | 22 (16%) | 4 (4%) | 12 (11%) |
| Post-HCT Cy + sirolimus + MMF | 3 (<1%) | 47 (34%) | 1 (<1%) | 1 (<1%) |
| Post-HCT Cy + sirolimus | — | 21 (15%) | — | — |
| Post-HCT Cy + CNI + MMF | 4 (<1%) | 14 (10%) | 1 (<1%) | — |
| CNI + MMF | 129 (23%) | 10 (7%) | 27 (25%) | 57 (54%) |
| CNI + methotrexate | 324 (58%) | 4 (3%) | 65 (58%) | 26 (26%) |
| CNI + sirolimus | 2 (<1%) | — | 2 (2%) | 1 (<1%) |
| CNI alone | 26 (5%) | 9 (7%) | 10 (9%) | 7 (7%) |
| Sirolimus alone | 68 (12%) | — | — | — |
| Transplant period | ||||
| 2008 – 2012 | 211 (38%) | 33 (24%) | 38 (35%) | 45 (43%) |
| 2013 – 2017 | 347 (62%) | 104 (76%) | 72 (65%) | 60 (57%) |
TBI dose for the non-myeloablative regimens:
HLA-matched sibling HCT: TBI dose was 300 cGy for TBI alone and TBI/Cy regimens. TBI dose for all other TBI-containing regimens were 200 cGy except for 2 patients who received TBI 300 cGy with the TBI/Cy/Flu regimen.
Haploidentical related donor HCT: TBI dose was 400 cGy for TBI alone regimen except for 1 patient who received TBI dose 300 cGy. TBI dose was 400 cGy for TBI/Cy regimen. TBI dose was 200 cGy was the predominant dose for TBI/Cy/Flu regimen except for 6 patients who received TBI 300 cGy and 14 patients who received TBI 400 cGy. TBI dose was 200 cGy for TBI/Cy/Flu/TT and TBI/Flu regimens.
Matched unrelated donor HCT: TBI dose was 300 cGy for TBI/Cy/Flu and TBI/Flu regimens and 200 cGy fpr TBI/Mel regimen
Mismatched unrelated donor HCT: TBI dose was 400 cGy for TBI alone regimen, 300 cGy for TBI/Cy/Flu regimen. One patient received 200 cGy and 1 patient, 400 cGy with TBI/Mel regimen.
Abbreviations:
CMV = cytomegalovirus
HCT = hematopoietic cell transplant
HLA = human leukocyte antigen
Cy = cyclophosphamide
Flu = fludarabine
TT = thiotepa
TBI = total body irradiation
GVHD = graft-versus-host disease
MMF = mycophenolate
CNI = calcineurin inhibitor
Role of the funding source
The CIBMTR is supported by grant U24-CA76518 from the National Institutes of Health and HHSH 250201200016C from Health Services Research Administration, Department of Health and Human Services. The sponsors of the study had no role in the study design, data collection, data analysis, interpretation or writing of the report. The corresponding author had full access to all data and final responsibility for the decision to submit for publication.
RESULTS
Table 1 shows the characteristics of 910 patients with SCD by donor type and HLA-match (HLA-matched sibling, haploidentical relative, HLA-matched and mismatched unrelated donors). Donor-recipient pairs were HLA-matched at the allele-level (HLA-A, -B, -C and -DRB1). There were no differences in sex distribution or recipient cytomegalovirus serostatus between the donor groups. However, compared to recipients of haploidentical related donor transplants, recipients of HLA-matched sibling, HLA-matched and HLA-mismatched unrelated donor transplants were younger, more likely to have performance scores of 90 or 100 and HCT-comorbidity index 0-2. HLA-matched siblings were the predominant donor group and bone marrow the predominant graft for HLA-matched sibling and HLA-matched unrelated donor transplants. The predominant graft source for haploidentical related donor transplants was peripheral blood and for HLA-mismatched unrelated donor transplants, umbilical cord blood. Conditioning regimen intensity was either, myeloablative, reduced intensity or nonmyeloablative based on previously defined criteria.20 In regards to regimen intensity, myeloablative and nonmyeloablative conditioning are at opposite ends of the regimen intensity spectrum. The reduced intensity regimens are an intermediate “intensity” category that do not fit the definition for myeloablative or nonmyeloablative. The predominant regimen intensity for HLA-matched sibling donor transplants was myeloablative (348 of 558, 62%) and for haploidentical related donor transplants was nonmyeloablative (89 of 137, 65%). Myeloablative and reduced intensity regimens were equally likely to be used for HLA-matched ([49 + 57] of 111, 95%) and mismatched unrelated donor transplants ([47 + 51] of 104, 98%). Tacrolimus or cyclosporine with methotrexate, mycophenolate, sirolimus or alone was used for most transplants (699 of 910, 77%). Posttransplant cyclophosphamide was used exclusively for haploidentical related donor transplants and was the predominant regimen in that group (82 of 137, 60%). Transplantations were more common in the period 2013-2017 compared to 2008-2012 especially haploidentical related donor transplants. The median follow-up of surviving patients after HLA-matched sibling, haploidentical related, HLA-matched and mismatched unrelated donor transplants were 36 months (IQR 18-60), 25 months (IQR 12-48), 37 months (IQR 23-60) and 47 months (24-72), respectively.
Event-free survival was lower in patients aged ≥13 years, after reduced intensity conditioning regimens compared to nonmyeloablative regimens and after transplantation of grafts from donors who were not HLA-matched siblings (Table 2, Figure 2A). Event-free survival did not differ between myeloablative and reduced intensity regimens (HR 0⋅80, 95% CI 0⋅56-1⋅13, p=0⋅21). Event-free survival did not differ between HLA-matched unrelated donors and haploidentical related (HR 1⋅43, 95% CI 0⋅81-2⋅50, p=0⋅21) or mismatched unrelated (HR 1⋅17, 95% CI 0⋅67-2⋅05, p=0⋅58) donors. Similarly, event-free survival did not differ between mismatched unrelated and haploidentical related donors (HR 1⋅22, 95% CI 0⋅65-2⋅27, p=0⋅98).
Table 2.
Risk factors for event-free survival, graft failure and overall survival
| Events / Evaluable | Hazard Ratio (95% confidence interval) | P-value | |
|---|---|---|---|
| Event-free survival | |||
| Age | |||
| <1 – 12 years | 72/491 | 1⋅00 | |
| 13 – 49 years | 102/418 | 1⋅74 (1⋅24 – 2⋅45) | 0⋅0014 |
| Regimen intensity | 0⋅046 | ||
| Nonmyeloablative | 36/181 | 1⋅00 | |
| Myeloablative | 75/478 | 1⋅57 (0⋅95 – 2⋅61) | 0⋅079 |
| Reduced intensity | 63/250 | 1⋅97 (1⋅15 – 3⋅36) | 0⋅013 |
| Donor type | <0⋅0001 | ||
| HLA-matched sibling | 52/557 | 1⋅00 | |
| Haploidentical related | 45/137 | 5⋅30 (3⋅17 – 8⋅86) | <0⋅0001 |
| HLA-matched unrelated | 38/111 | 3⋅71 (2⋅39 – 5⋅75) | <0⋅0001 |
| HLA-mismatched unrelated | 39/104 | 4⋅34 (2⋅58 – 7⋅32) | <0⋅0001 |
| Graft type | |||
| Bone marrow | 105/630 | 1⋅00 | 0⋅33 |
| Peripheral blood | 40/177 | 1⋅01 (0⋅66 – 1⋅54) | 0⋅98 |
| Umbilical cord blood | 29/102 | 1⋅52 (0⋅87 – 2⋅65) | 0⋅14 |
| HCT comorbidity index | |||
| 0-2 | 125/637 | 1⋅00 | |
| ≥ 3 | 49/272 | 0⋅86 (0⋅61 – 1⋅24) | 0⋅42 |
| Performance score | |||
| ≤ 80 | 29/160 | 1⋅00 | |
| 90 – 100 | 143/731 | 1⋅33 (0⋅87 – 2⋅04) | 0⋅19 |
| Recipient CMV serology | |||
| Negative | 81/454 | 1⋅00 | |
| Positive | 93/455 | 1⋅33 (0⋅98 – 1⋅80) | 0⋅065 |
| Sex | |||
| Male | 95/488 | 1⋅00 | |
| Female | 79/421 | 0⋅86 (0⋅63 – 1⋅16) | 0⋅31 |
| Transplant period | |||
| 2008 – 2012 | 73/327 | 1⋅00 | |
| 2013 – 2017 | 101/582 | 0⋅98 (0⋅71 – 1⋅36) | 0⋅89 |
| Graft failure | |||
| Age | |||
| <1 – 12 years | 54/491 | 1⋅00 | |
| 13 – 49 years | 59/418 | 1⋅17 (0⋅76 – 1⋅82) | 0⋅47 |
| Regimen intensity | 0⋅28 | ||
| Nonmyeloablative | 31/181 | 1.00 | |
| Myeloablative | 44/478 | 1⋅04 (0⋅56 – 1⋅91) | 0⋅91 |
| Reduced intensity | 38/250 | 1⋅46 (0⋅77 – 2⋅79) | 0⋅25 |
| Donor type | <0⋅0001 | ||
| HLA-matched sibling | 32/557 | 1⋅00 | |
| Haploidentical related | 36/137 | 6⋅58 (3⋅55 – 12⋅21) | <0⋅0001 |
| HLA-matched unrelated | 16/111 | 2⋅88 (1⋅54 – 5⋅37) | 0⋅00090 |
| HLA-mismatched unrelated | 29/104 | 4⋅38 (2⋅31 – 8⋅31) | <0⋅0001 |
| Graft type | |||
| Bone marrow | 60/630 | 1⋅00 | 0⋅09 |
| Peripheral blood | 28/177 | 1⋅02 (0⋅61 – 1⋅70) | 0⋅95 |
| Umbilical cord blood | 25/102 | 2⋅07 (1⋅08 – 3⋅96) | 0⋅028 |
| HCT comorbidity index | |||
| 0-2 | 81/637 | 1⋅00 | |
| ≥ 3 | 32/272 | 0⋅77 (0⋅49 – 1⋅21) | 0⋅26 |
| Performance score | |||
| ≤ 80 | 23/160 | 1⋅00 | |
| 90 – 100 | 88/731 | 1⋅03 (0⋅62 – 1⋅70) | 0⋅92 |
| Recipient CMV serology | |||
| Negative | 53/454 | 1⋅00 | |
| Positive | 60/455 | 1⋅32 (0⋅90 – 1⋅92) | 0⋅16 |
| Sex | |||
| Male | 71/488 | 1⋅00 | |
| Female | 42/421 | 0⋅61 (0⋅42 – 0⋅90) | 0⋅013 |
| Transplant period | |||
| 2008 – 2012 | 47/327 | 1⋅00 | |
| 2013 – 2017 | 66/582 | 1⋅01 (0⋅67 – 1⋅51) | 0⋅96 |
| Overall Survival | |||
| Age | |||
| <1 – 12 years | 22/491 | 1⋅00 | |
| 13 – 49 years | 54/418 | 3⋅15 (1⋅86 – 5⋅34) | <0⋅0001 |
| Regimen intensity | 0⋅004 | ||
| Nonmyeloablative | 7/181 | 1⋅00 | |
| Myeloablative | 41/478 | 4⋅62 (1⋅87 – 11⋅44) | 0⋅00093 |
| Reduced intensity | 28/250 | 3⋅79 (1⋅46 – 9⋅84) | 0⋅0062 |
| Donor type | <0⋅0001 | ||
| HLA-matched sibling | 21/557 | 1⋅00 | |
| Haploidentical related | 13/137 | 2⋅94 (1⋅26 – 6⋅87) | 0⋅013 |
| HLA-matched unrelated | 26/111 | 5⋅12 (2⋅79 – 9⋅40) | <0⋅0001 |
| HLA-mismatched unrelated | 16/104 | 4⋅88 (2⋅22 – 10⋅75) | <0⋅0001 |
| Graft type | |||
| Bone marrow | 50/630 | 1⋅00 | 0⋅09 |
| Peripheral blood | 19/177 | 1⋅02 (0⋅61 – 1⋅70) | 0⋅95 |
| Umbilical cord blood | 7/102 | 0⋅58 (0⋅22 – 1⋅55) | 0⋅27 |
| HCT comorbidity index | |||
| 0-2 | 53/637 | 1⋅00 | |
| ≥ 3 | 23/272 | 1⋅15 (0⋅68 – 1⋅93) | 0⋅60 |
| Performance score | |||
| ≤ 80 | 10/160 | 1⋅00 | |
| 90 – 100 | 65/731 | 1⋅71 (0⋅85 – 3⋅45) | 0⋅13 |
| Recipient CMV serology | |||
| Negative | 34/454 | 1⋅00 | |
| Positive | 42/455 | 1⋅35 (0⋅85 – 2⋅14) | 0⋅20 |
| Sex | |||
| Male | 31/488 | 1⋅00 | |
| Female | 45/421 | 1⋅49 (0⋅94 – 2⋅38) | 0⋅09 |
| Transplant period | |||
| 2008 – 2012 | 31/327 | 1⋅00 | |
| 2013 – 2017 | 45/582 | 0⋅94 (0⋅57 – 1⋅56) | 0⋅81 |
Abbreviation:
GVHD = graft versus-host disease
HCT = hematopoietic cell transplant
CMV = cytomegalovirus
Figure 2.
A. Event-free survival: The 3-year probability of event-free survival after A: HLA-matched sibling (89%, 95% CI 85-91), B: Haploidentical relative (49%, 95% CI 34-62), C: HLA-matched unrelated (69%, 95% CI 59-77) and D: HLA-mismatched unrelated (63%, 95% CI 52-72) donor transplantation, adjusted for age and transplant conditioning regimen intensity.
B. Overall survival: The 3-year probability of overall survival after A: HLA-matched sibling (96%, 95% CI 93-97), B: Haploidentical relative (87%, 95% CI 76-93), C: HLA-matched unrelated (81%, 95% CI 73-88) and D: HLA-mismatched unrelated (82%, 95% CI 70–90) donor transplantation, adjusted for age and transplant conditioning regimen intensity.
C. Graft failure: The 3-year probability of graft failure after A: HLA-matched sibling (7% (95% CI 5–10), B: Haploidentical relative (44% (95% CI 29–58), C: HLA-matched unrelated (17% (95% CI 10–26) and D: HLA-mismatched unrelated (21% (95% CI 13–31) donor transplantation.
Donor type was associated with graft failure. Compared to HLA-matched sibling transplants graft failure was higher after haploidentical relative, HLA-matched and mismatched unrelated donor transplants (Table 2, Figure 2B). Compared to HLA-matched unrelated donor transplants, graft failure was higher after haploidentical related (HR 2⋅27, 95% CI 1⋅09-4⋅76, p=0⋅028) but not HLA-mismatched unrelated (HR 1⋅52, 95% CI 0⋅71-3⋅24, p=0⋅28). . Graft failure did not differ between mismatched unrelated and haploidentical related donor transplants (HR 0⋅67, 95% CI 0⋅31-1⋅44, p=0⋅30). Graft failure was lower in females compared to males.
Overall survival was lower in patients aged ≥13 years, myeloablative and reduced intensity conditioning compared to nonmyeloablative regimens and after transplantation of grafts from donors who were not HLA-matched siblings (Table 2, Figure 2C). Survival did not differ between myeloablative and reduced intensity regimens (HR 1⋅22, 95% CI 0⋅74-2⋅02, p=0⋅44). Compared to HLA-matched unrelated donor transplants, survival did not differ between haploidentical related (HR 0⋅57, 95% CI 0⋅24-1⋅35, p=0⋅20) and mismatched unrelated (HR 0⋅95, 95% CI 0⋅44-2⋅05, p=0⋅90) donor transplants. Similarly survival did not differ between mismatched unrelated and haploidentical related donor transplants (HR 0⋅60, 95% CI 0⋅24-1⋅49, p=0⋅27). The 3-year probabilities of overall and event-free survival and graft failure are shown in Supplemental Table 1, page 1.
Acute GVHD was higher with myeloablative and reduced intensity conditioning compared to nonmyeloablative regimens, after transplantation of grafts from HLA-matched and mismatched unrelated donors compared to HLA-matched siblings and transplant period 2013-2017 compared to 2008-2012 (Table 3). Acute GVHD risks did not differ between myeloablative and reduced intensity conditioning regimens (HR 0⋅93, 95% CI 0⋅60-1⋅46, p=0⋅77). Compared to HLA-matched unrelated donor transplants, acute GVHD risks were not different between haploidentical related (HR 0⋅63, 95% CI 0⋅27-1⋅47, p=0⋅29) and mismatched unrelated (HR 1⋅27, 95% CI 0⋅62-2⋅59, p=0⋅51) unrelated donor transplants. Similarly, risks did not differ between mismatched unrelated and haploidentical related donor transplants (HR 0⋅37, 95% CI 0⋅20-1⋅20, p=0.12). The severity of acute GVHD did not differ by donor type (p=0⋅35).
Table 3.
Risk factors for acute and chronic GVHD
| Events / Evaluable | Hazard Ratio (95% confidence interval) | P-value | |
|---|---|---|---|
| Grade II-IV acute GVHD | |||
| Age | |||
| <1 – 12 years | 46/446 | 1⋅00 | |
| 13 – 49 years | 49/377 | 1⋅39 (0⋅89 – 2⋅18) | 0⋅14 |
| Regimen intensity | 0⋅0062 | ||
| Nonmyeloablative | 7/171 | 1⋅00 | |
| Myeloablative | 51/440 | 4⋅14 (1⋅68 – 10⋅22) | 0⋅0020 |
| Reduced intensity | 37/212 | 4⋅43 (1⋅73 – 11⋅35) | 0⋅0019 |
| Donor type | <0⋅0001 | ||
| HLA-matched sibling | 32/513 | 1⋅00 | |
| Haploidentical related | 11/126 | 2⋅27 (1⋅08 – 4⋅77) | 0⋅03 |
| HLA-matched unrelated | 23/97 | 3⋅84 (2⋅22 – 6⋅63) | <0≥0001 |
| HLA-mismatched unrelated | 29/87 | 6⋅14 (3⋅66 – 10⋅28) | <0⋅0001 |
| Graft type | |||
| Bone marrow | 58/565 | 1⋅00 | 0⋅37 |
| Peripheral blood | 15/163 | 0⋅98 (0⋅50 – 1⋅91) | 0⋅95 |
| Umbilical cord blood | 22/95 | 1⋅61 (0⋅81 – 3⋅20) | 0⋅18 |
| HCT comorbidity index | |||
| 0-2 | 65/564 | 1⋅00 | |
| ≥ 3 | 30/259 | 0⋅94 (0⋅59 – 1⋅49) | 0⋅79 |
| Performance score | |||
| ≤ 80 | 22/149 | 1⋅00 | |
| 90 – 100 | 72/658 | 0⋅71 (0⋅42 –1⋅19) | 0⋅19 |
| Recipient CMV serology | |||
| Negative | 45/411 | 1⋅00 | |
| Positive | 50/412 | 1⋅19 (0⋅79 – 1⋅80) | 0⋅39 |
| Sex | |||
| Male | 45/439 | 1⋅00 | |
| Female | 50/384 | 1⋅16 (0⋅77 – 1⋅76) | 0⋅47 |
| Transplant period | |||
| 2008 – 2012 | 24/285 | 1⋅00 | |
| 2013 – 2017 | 71/538 | 1⋅74 (1⋅06 – 2⋅84) | 0⋅028 |
| Chronic GVHD | |||
| Age | |||
| <1 – 12 years | 98/491 | 1⋅00 | |
| 13 – 49 years | 92/419 | 1⋅46 (1⋅06 – 2⋅00) | 0⋅019 |
| Regimen intensity | <0⋅0001 | ||
| Nonmyeloablative | 17/181 | 1⋅00 | |
| Myeloablative | 97/478 | 2⋅82 (1⋅51 – 5⋅27) | 0⋅0012 |
| Reduced intensity | 76/250 | 4⋅00 (2⋅11 – 7⋅55) | <0⋅0001 |
| Donor type | 0⋅017 | ||
| HLA-matched sibling | 101/557 | 1⋅00 | |
| Haploidentical related | 22/137 | 1⋅75 (0⋅55 – 3⋅11) | 0⋅055 |
| HLA-matched unrelated | 37/111 | 1⋅70 (1⋅14 – 2⋅54) | 0⋅0087 |
| HLA-mismatched unrelated | 30/104 | 1⋅59 (0⋅95 – 2⋅67) | 0⋅08 |
| Graft type | |||
| Bone marrow | 146/630 | 1⋅00 | 0⋅09 |
| Peripheral blood | 20/177 | 0⋅55 (0⋅32 – 0⋅94) | 0⋅028 |
| Umbilical cord blood | 24/102 | 0⋅84 (0⋅49 – 1⋅46) | 0⋅54 |
| HCT comorbidity index | |||
| 0-2 | 135/637 | 1⋅00 | |
| ≥ 3 | 55/272 | 1⋅13 (0⋅81 – 1⋅57) | 0⋅47 |
| Performance score | |||
| ≤ 80 | 34/160 | 1⋅00 | |
| 90 – 100 | 152/731 | 0⋅86 (0⋅57 –1⋅28) | 0⋅44 |
| Recipient CMV serology | |||
| Negative | 93/454 | 1⋅00 | |
| Positive | 97/455 | 1⋅05 (0⋅79 – 1⋅40) | 0⋅74 |
| Sex | |||
| Male | 89/488 | 1⋅00 | |
| Female | 101/421 | 1⋅32 (0⋅99 – 1⋅77) | 0⋅06 |
| Transplant period | |||
| 2008 – 2012 | 76/327 | 1.00 | |
| 2013 – 2017 | 114/583 | 1.07 (0.79 – 1.46) | 0.65 |
Abbreviation:
GVHD = graft versus-host disease
HCT = hematopoietic cell transplant
CMV = cytomegalovirus
Chronic GVHD was higher in patients aged ≥13 years, those treated with myeloablative and reduced intensity conditioning and transplanted with grafts from an HLA-matched unrelated compared to HLA-matched sibling donors (Table 3). Chronic GVHD was also higher after reduced intensity compared to myeloablative conditioning regimens (HR 1⋅41, 95% CI 1⋅03-1⋅96, p=0⋅031). Compared to HLA-matched unrelated donor transplants, chronic GVHD risks did not differ between haploidentical related (HR 1⋅03, 95% CI 0⋅54-1⋅96, p=0⋅93) and mismatched unrelated (HR 0⋅93, 95% CI 0⋅52-1⋅67, p=0⋅82). Similarly, chronic GVHD did not differ between mismatched unrelated and haploidentical related donor transplants (HR 1⋅09, 95% CI 0⋅54-2⋅22, p=0⋅79). The severity of chronic GVHD did not differ by donor type (p=0⋅12). The day-100 incidences of acute GVHD and 3-year incidences of chronic GVHD are shown in Supplemental Table 1, page 1.
A subset analysis of 558 HLA-matched sibling transplant recipients was undertaken to primarily study the effect of conditioning regimen intensity, not previously studied. Conditioning regimen intensity was not associated with survival. However, graft failure was higher after reduced intensity conditioning compared to myeloablative (HR 0.28, 95% CI 0.13-0.57, p<0.0001) and nonmyeloablative (HR 0.29, 95% CI 0.08-1.00, p=0.049) regimens. Consequently, event-free survival was lower after reduced intensity conditioning compared to myeloablative (HR 0.38, 95% CI 0.21-0.67, p=0.00080) and nonmyeloablative (HR 0.36, 95% CI 0.13-0.94, p=0.036) regimens. Consistent with the main analysis, the risk of death was three time higher in patients aged ≥13 years (HR 3.25, 95% CI 1.27-8.29, p=0.014).
Six patients developed malignant neoplasm (acute myeloid leukemia [n=2], myelodysplastic syndrome [n=2], hepatic myelofibroblastic tumor [n=1] and TCR-β gene rearrangement positive T-cell large granular lymphocytic leukemia [n=1]), Supplemental Table 2, page 2. Three of six patients with malignant neoplasm are dead. The risk of developing a malignancy post-transplant was seven times higher with non-myeloablative compared to reduced intensity regimens although this did not reach the level of significance set for this analysis (HR 7⋅08, 95% CI 0⋅82-60⋅63, p=0⋅07). None of the six patients had chronic GVHD. There were no cases of posttransplant malignant neoplasm with myeloablative regimens. Nine patients developed Epstein-Barr virus (EBV) positive lymphoproliferative disease at a median 5 months (range 2 – 15) after transplant with all having had in vivo T-cell depletion with anti-thymocyte globulin or alemtuzumab. Notably, acute chest syndrome (n=1) and stroke (n=6) were reported in a subset of patients for whom pre- and post-transplant sickle cell disease-specific data were available (421 of 910; 46%); all 7 had full donor myeloid chimerism.
DISCUSSION
There were two key findings in this retrospective analysis that warrant caution when considering alternative donor transplantation as a potentially curative option for SCD. Mortality risks and graft failure were higher after any alternative donor transplantation compared to HLA-matched sibling transplantation and that resulted in substantially lower event-free survival. To our knowledge this is the first study on allogeneic transplantation for SCD that has compared outcomes after HLA-matched sibling donors to that after alternative donors (haploidentical related, HLA-matched unrelated and HLA-mismatched unrelated donors). The study also addressed the effect of conditioning regimen intensity on transplant outcomes, not previously reported to our knowledge. Myeloablative and reduced intensity regimens were associated with higher mortality, and higher acute and chronic GVHD. Transplant regimen intensity was not associated with graft failure.
Higher mortality and graft failure after HLA mismatched unrelated donor transplantation has been reported for non-malignant diseases but those studies had very patients who were transplanted for hemoglobinopathy.21,22 Our findings confirm the adverse effect of HLA disparity on survival and graft failure in SCD. The timing of graft failure differed by donor type in SCD. Graft failure after HLA-matched sibling and unrelated donor transplantation primarily occurred within 1-2 years after transplantation. In contrast, graft failure was more common 2-3 years after haploidentical related donor transplantation. This is particularly relevant in light of the excellent outcomes of two recent phase II trials of haploidentical related donor transplant with 15 and 12 patients with SCD.23,24 Both trials used the post-transplant cyclophosphamide approach to overcome the HLA barrier One trial reported >95% donor engraftment in 14 of 15 patients at 6 months23 and the other trial,24 reported full or mixed donor engraftment 11 patients in the 1st year after transplantation. Longer follow up of those patients are needed to confirm sustained donor engraftment. Conditioning regimen intensity was not associated with graft failure.
Higher mortality after myeloablative conditioning has been recorded for several decades and the concept of less intense conditioning regimens (reduced or nonmyeloablative) was introduced to overcome mortality risks associated with myeloablation in less fit patients or for diseases for which myeloablation was not desirable. So, it was surprising that we observed mortality risks were four-fold higher with both myeloablative and reduced intensity regimens compared to nonmyeloablative regimens. The recorded higher mortality may be explained in part by higher acute and chronic GVHD risks with myeloablative and reduced intensity regimens. These regimens relied on standard GVHD prophylaxis that included a calcineurin inhibitor with methotrexate, mycophenolate, sirolimus or calcineurin inhibitor as single agent. The nonmyeloablative regimens used a different strategy. In the setting of HLA-matched sibling transplantation, the low dose TBI approach incorporates an attempt at tolerance induction through the use of lymphocyte reduction with alemtuzumab along with mTOR inhibition with sirolimus during recovery.25 This tolerance is blocked in vitro and in animal models with calcineurin inhibitor.25 Similarly, for haploidentical related donor transplantation, posttransplant cyclophosphamide induces early immune tolerance that is mediated by destruction of alloreactive donor and recipient T-cells and any remaining alloreactivity is counterbalanced by increasing T regulatory cells.26 A delayed but long-lasting intra-thymic clonal deletion of anti-host T cells maintains long term immune tolerance.26 Tolerance induction and low rates of GVHD coupled with low intensity regimens likely contributed to the high event-free and overall survival in these patients. Higher acute GVHD after the period 2012 is explained by increasing numbers of unrelated donor transplantation.27 Taken together, we hypothesize immune tolerance induction rather than regimen intensity per se is a key driver of survival after transplantation for SCD.
Age at transplantation was an important predictor of survival. The risk of death is three times higher in a patient aged 13 years or older compared to a younger patient assuming the donor and conditioning regimen intensity are the same for both patients. Although almost half the study population was older than 13 years we did not find another age cut-off associated with survival differences. Only 7 transplants were reported in patients aged 50 years and older and were not included in the current analyses. The effect of age in adults can only be studied properly in a larger adult population than represented in our population and is a limitation of this study. The timing of transplantation is dependent on physician and patient choice, donor availability and access to health care. With the exception of stroke, severity of sickle cell symptoms that prompts referral for transplantation is variable. In the absence of a comparative study of transplant recipients and those receiving non-transplant therapies with comparable disease severity, we cannot comment on whether HLA-matched sibling transplantation should be offered in the first decade of life. With 3-year incidence of chronic GVHD after HLA-matched sibling transplantation at 18% (95% CI 15-22) we cannot recommend transplantation for asymptomatic children or for children without severe disease. Consistent with published literature we did not record an association between comorbidity index and survival.28
There are several limitations to our study. These include, the decision to offer transplantation and its timing, choice of conditioning regimen intensity, and choice of alternative donor in the absence of a matched sibling. We acknowledge transplant strategies are best studied in the setting of multi-center trials. Yet, this is challenging with accrual extending over 5 years for funded multi-center transplant trials.5,11 We did not collect hemoglobin S concentration consistently after transplantation. Our definition of graft failure considered donor chimerism collected up to 2 years after transplantation. Thereafter, our standardized data collection forms ask whether the patient experienced graft failure (<5% donor) and the date of failure. We do not have data on red blood cell chimerism. The occurrence of events such as acute chest syndrome and stroke in the setting of full donor chimerism deserves further study and best achieved through careful longitudinal follow-up focusing on sickle cell related complications. We do not collect data in sufficient dependent to address this within the context of data as it is collected now. Similarly, the recorded higher risk of myeloid malignancy post-transplant after nonmyeloablative regimens merits further investigation. Whether this is attributed to age, transplantation per se, regimen intensity or some other unknown or unmeasured factor cannot be addressed in this study. The standardized incidence ratio for hematologic cancer in an unselected SCD population is 1.72 (95% CI 1.17-2.44) and patients were diagnosed with their first primary cancer at a median age of 46 years.29 With increasing numbers of transplants in young adults it might be possible several years later to design comparative studies on cancer incidence amongst transplanted and non-transplanted SCD patients. Although our data does not favor one alternative donor over another, a 20% absolute decrement in event-free survival between haploidentical and HLA-matched unrelated donor transplantation at 3-years cannot be ignored. To detect a significant difference with 80% power, 271 patients are needed but the current analyses only has 248 patients.
Allogeneic transplantation is potentially curative, but we do not know whether the up-front mortality from the procedure may with follow-up overcome mortality associated with complications of SCD. A phase II trial (NCT02766465) in the United States for young adults with severe SCD addresses this question by assigning participants eligible for the trial (based on disease severity and organ function) to a “donor arm” if they have a suitably matched sibling or unrelated donor (expected to undergo transplantation) or a “no donor” arm (expected to receive best available standard of care) for participants without a suitable donor. We acknowledge monetary coverage for access to health care is critical to improve survival. Access to transplantation for young adults with SCD is also challenging in the United States. The recent introduction of the Coverage with Evidence Determination (CED) program by Center for Medicare and Medicaid Services is likely to broaden access to allogeneic transplantation. Another potentially curative treatment for SCD that is being pursued, is with lentiviral gene addition of an anti-sickling β-globin variant into autologous hematopoietic cells.30 This approach is being studied in limited numbers of patients and definitive conclusions will require confirmation of success in larger numbers of patients as well as longer follow-up.
Supplementary Material
RESEARCH IN CONTEXT.
Evidence before the study
We searched MEDLINE (29/08/2018) for articles on sickle cell disease published after 2010, with the search terms “sickle cell disease”, “HLA-matched sibling”, “haploidentical”, “unrelated” and “transplant”. In addition to two reports of HLA-matched sibling donor transplants7,8 there were 5 reports on haploidentical and unrelated donor transplants but limited to small numbers of patients.10–13,25 We did not identify any reports that compared outcomes after HLA-matched sibling, haploidentical and unrelated donor transplantation. Further, the heterogeneity of transplant conditioning regimens in the reports made it impossible to assess the effect of conditioning regimen intensity on transplant outcomes. Therefore, our aim was to compare the relative risks of donor type and conditioning regimen intensity on outcomes after allogeneic HCT in patients with sickle cell disease with the hypothesis that HLA-matched sibling transplantation with myeloablative conditioning would lead to fewer graft failure and better survival.
Added value of the study
To our knowledge this is the first study to compare transplant outcomes by donor type for sickle cell disease. In a population of 910 patients we recorded highest event-free and overall survival after HLA-matched sibling transplants. Graft failure was higher after alternative donor compared to HLA-matched sibling transplantation.
Implications of all the available evidence
There are publications with fewer than 30 patients that have reported outcomes using alternative donors but not a direct comparison as we have undertaken.
Acknowledgement:
This work was undertaken through conversations with the members of the ASH Guideline Panel on Sickle-cell Disease-Related Transplantation.
Funding:
The CIBMTR is supported by grant U24-CA76518 from the National Institutes of Health and HHSH 250201200016C from Health Services Research Administration, Department of Health and Human Services, United States.
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: Authors declare none
REFERENCES
- 1.Quinn CT, Rogers ZR, Buchanan GR. Survival of children with sickle cell disease. Blood 2004; 103: 4023–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lanzkron S, Carroll CP, Haywood C Jr, et al. Mortality rates and age at death from sickle cell disease: U.S., 1979 – 2005. Public Health Rep 2013; 128: 110–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Elmariah H, Garret ME, De Castro LM, et al. Factors associated with survival in a contemporary adult sickle cell disease cohort. Am J Hematol 2014; 89: 530–5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gladwin MT, Barst RJ, Gibbs JS, et al. ; walk-PHAaSST Investigators and Patients. Risk factors for death in 632 patients with sickle cell disease in the United States and United Kingdom. PLos One 2014; 9: e99489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Walters MC, Patience M, Leisering M, et al. Bone marrow transplantation for sickle cell disease. N Engl J Med 1996; 335: 369–76 [DOI] [PubMed] [Google Scholar]
- 6.Bernaudin F, Socie G, Kuentz M, et al. Long-term results of related myeloablative stem-cell transplantation to cure sickle cell disease. Blood 2007; 110: 2749–56 [DOI] [PubMed] [Google Scholar]
- 7.Hsieh MM, Fitzhugh C, Weltzel P, et al. Nonmyeloablative HLA-matched sibling allogeneic hematopoietic stem cell transplantation for severe sickle cell phenotype. JAMA 2014; 312: 48–56 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gluckman E, Cappelli B, Bernaudin F, et al. Sickle cell disease: an international survey of results of HLA-identical sibling hematopoietic stem cell transplantation. Blood 2017; 129: 1548–56 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Garget L, Eapen M, Williams E, et al. HLA-match likelihoods for hematopoietic stem-cell grafts in the U.S. registry. N Engl J Med 2014; 371: 339–48 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bolanos-Meade J, Fuchs EJ, Luznik L, et al. HLA-haploidentical bone marrow transplantation with posttransplant cyclophosphamide expands the donor pool for patients with sickle cell disease. Blood 2012; 120; 4285–91 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shenoy S, Eapen M, Panepinto JA, et al. A trial of unrelated donor marrow transplantation for children with severe sickle cell disease. Blood 2016; 128: 2561–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Foell J, Pfirstinger B, Rehe K, Wolff D, Holler E, Corbacioglu S. Haploidentical stem cell transplantation with Cd3+/CD19+ depleted peripheral stem cells for patients with advanced stage sickle cell disease and no alternative donor: results of a pilot study. Bone Marrow Transplant 2017; 52: 938–40 [DOI] [PubMed] [Google Scholar]
- 13.Fitzhugh CO, Hsieh MM, Taylor T, et al. Cyclophosphamide improves engraftment in patients with SCD and severe organ damage who undergo haploidentical PBSCT. Blood Adv. 2017; 1: 652–61 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Olsson R, Remberger M, Schaffer M, et al. Graft failure in the modern era of allogeneic hematopoietic SCT. Bone Marrow Transplant 2013; 48: 537–43 [DOI] [PubMed] [Google Scholar]
- 15.Przepiorka D, Weisdorf D, Martin P, et al. 1994 consensus conference on acute GVHD grading. Bone Marrow Transplant. 1995; 15: 825–8 [PubMed] [Google Scholar]
- 16.Atkinson K, Horowitz MM, Gale RP, Lee MB, Rimm AA, Bortin MM. Committee of the International Bone Marrow Transplant Registry. Consensus among bone marrow transplanters for diagnosis, grading and treatment of chronic graft versus host disease. Bone Marrow Transplant 1989; 4: 247–54 [PubMed] [Google Scholar]
- 17.Lin DY. Non-parametric inference for cumulative incidence functions in competing risks. Stat Med 1997; 16: 901–910 [DOI] [PubMed] [Google Scholar]
- 18.Cox DR. Regression model and life-tables. J R Stat Soc Series B 1972; 34: 187–200 [Google Scholar]
- 19.Zhang X, Loberiza FR, Klein J, Zhang MJ. A SAS macro for estimation of direct adjusted survival curves based on a stratified Cox regression model. Comput Methods Programs Biomed 2007; 88: 95–101 [DOI] [PubMed] [Google Scholar]
- 20.Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning regimens: working definitions. Biol Blood Marrow Transplant 2009; 15: 1628–33 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Horan J, Wang T, Haagenson M, et al. Evaluation of HLA matching in unrelated hematopoietic stem cell transplantation for nonmalignant disorders. Blood 2012; 120: 2918–24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Eapen M, Wang T, Veys PA, et al. Allele-level HLA matching for umbilical cord blood transplantation for non-malignant diseases in children: a retrospective analysis. Lancet Haematol 2017; 4: e325–33 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.de la Fuente J, Dhedin N, Koyama T et al. Haploidentical bone marrow transplant with post-transplant cyclophosphamide plus thiotepa improves donor engraftment in patients with sickle cell anemia: results of an international learning collaborative. Biol Blood Marrow Transplant 2018; S1083-8791 (18)30772–9. Doi. 10.1016/j.bbmt.2018.11.027 (Epub ahead of print) [DOI] [PubMed] [Google Scholar]
- 24.Bolaños-Meade J, Cooke KR, Gamper CJ et al. Effect of increased dose of total body irradiation on graft failure associated with HLA-haploidentical transplantation in patients with severe hemoglobinopathies: a prospective clinical trial. Lancet Haematol. 2019. 4: e183–93 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fitzhugh C, Weitzel RP, Hsieh MM, et al. Sirolimus and post transplant CY synergistically maintain mixed chimerism in a mismatched murine model. Bone Marrow Transplant 2013; 48: 1335–41 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Al-Homsi AS, Roy TS, Cole K, Feng Y, Duffner U. Post-transplant cyclophosphamide for the prevention of graft-versus-host disease. Biol Blood Marrow Transplant. 2015; 21: 604–11 [DOI] [PubMed] [Google Scholar]
- 27.Jagasia M, Arora M, Flowers MED, et al. Risk factors for acute GVHD and survival after hematopoietic cell transplantation. Blood 2012; 119: 296–307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Thakar M, Broglie L, Logan B, et al. The hematopoietic cell transplant comorbidity index predicts survival after allogeneic transplant for non-malignant diseases. Blood. 2019; 133: 754–62 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Brunson A, Keegan THM, Bang H, Mahajan A, Paulukonis S, Wun T. Increased risk of leukemia among sickle cell disease patients in California. Blood 2017; 130: 1597–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Urbinati F, Campo Fernandez B, Masiuk KE, et al. Gene therapy for sickle cell disease: a lentiviral vector comparison study. Hum Gene Ther 2018; 29:1153–66 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.