|Home | About | Journals | Submit | Contact Us | Français|
Designed research and drafted the manuscript (AW, SJL, CH, MH, SS), performed the statistical analysis (JPK, MH), interpreted data and critically revised the manuscript (all authors).
The association between human leukocyte antigen (HLA) matching and outcome in unrelated donor, peripheral blood stem cell (PBSC) transplantation has not been established.
1933 unrelated donor-recipient pairs transplanted between 1999-2006 for AML, ALL, MDS or CML and who had high resolution HLA typing for HLA-A, B, C, DRB1, DQA1 and DQB1 were included in the analysis. Outcomes were compared between HLA-matched and HLA-mismatched pairs, adjusting for patient and transplant characteristics.
Matching for HLA-A, -B, -C and DRB1 alleles [8/8 match] was associated with better survival at one year compared with 7/8 HLA-matched pairs (56% vs. 47%). Using 8/8 HLA-matched patients as the baseline (n=1243), HLA-C antigen mismatches (n=189) were statistically significantly associated with lower LFS (RR 1.36 [95% CI 1.13-1.64] p=0.0010), and increased risk for mortality (RR=1.41 [1.16-1.70], p=0.0005), treatment-related mortality (RR=1.61 [1.25-2.08], p=0.0002), and grades III-IV graft-versus-host disease (RR=1.98 [1.50-2.62], p<0.0001). HLA-B antigen or allele mismatching was associated with a higher risk for acute GVHD grades III-IV. No statistically significant differences in outcome were observed for HLA-C allele mismatch (n=61), nor for mismatches at HLA-A antigen/allele (n=136), HLA-DRB1 allele (n=39) or HLA-DQ antigen/allele (n=114) compared to 8/8 HLA-matched pairs. HLA mismatching was not associated with relapse or chronic GVHD.
HLA-C antigen mismatched unrelated PBSC donors are associated with worse outcomes compared with 8/8 HLA-matched donors. Limited power due to small sample sized prevents comment about other mismatches.
Unrelated donors have provided a vital resource for patients who do not have an HLA-matched relative. Approximately 50% of allogeneic transplants reported to the Center for International Blood and Marrow Transplant Research (CIBMTR) utilize unrelated donors. Over the past ten years, the number of peripheral blood stem cell (PBSC) grafts facilitated by the National Marrow Donor Program (NMDP) has grown substantially, such that currently around 75% of unrelated grafts are PBSC [NMDP statistics]. Additionally, around 30% of all PBSC products are mismatched for one or more of the recipient’s human leukocyte antigen (HLA) loci. Previous NMDP/CIBMTR studies evaluating the effects of HLA mismatching included predominantly bone marrow recipients. Since the number of unrelated donor PBSC transplants in the NMDP registry has now reached sufficient quantity for preliminary analysis, this study was designed to determine the association of HLA-mismatching with survival, relapse, graft-vs-host disease, and transplant-related mortality.
Previous studies from the NMDP/CIBMTR in the setting of bone marrow transplantation have shown that HLA-mismatching is associated with worse outcomes.1,2 In particular, single mismatches at HLA-A, -B, -C or DRB1 were associated with a higher risk for treatment-related mortality (TRM) and acute graft-versus-host disease (GVHD) compared to 8/8 HLA-matched pairs. Isolated HLA-DQ mismatches did not appear detrimental. Reports from the Fred Hutchinson Cancer Research Center and the Japanese Marrow Donor Program also support the concept that disparities involving HLA-class I alleles are independent risk factors for acute GVHD, TRM, and overall survival.3,4
In the 1990s, collection of granulocyte colony-stimulating factor (G-CSF) mobilized PBSC was introduced as an alternative to bone marrow donation for volunteer unrelated donors.5 The advantage of PBSC compared to marrow is faster engraftment of neutrophils and platelets for patients, and the ability to avoid the operating room for donors and physicians. Retrospective studies show similar rates of acute GVHD, TRM, relapse and survival with unrelated donor PBSC and bone marrow (BM), but the incidence of extensive chronic GVHD was increased with PBSC.6 Although PBSC has supplanted BM as the most common source of unrelated hematopoietic stem cells, the impact of HLA mismatching on outcomes after unrelated PBSC transplantation has not been well studied. The present study was undertaken to compare the outcomes of HLA-mismatched compared to HLA-matched unrelated donor transplantation when PBSC is used as the graft source. Identification of mismatched HLA loci associated with particularly poor outcomes may help guide donor selection when an 8/8 HLA-matched donor is not available and allogeneic transplantation is recommended.
The study population included all patients reported to the NMDP/CIBMTR registries who received an unrelated PBSC transplant between 1999 and 2006 for AMl, ALL, CML or MDS, and for which there existed retrospective high-resolution HLA typing results for both patient and unrelated donor. Diseases were categorized as early phase (acute leukemia in first complete remission (CR1), CML in first chronic phase, and MDS-refractory anemia (RA), intermediate phase (acute leukemia in second remission (CR2) and CML in accelerated or second chronic phase), or advanced phase (acute leukemia advanced beyond CR2 or not in remission, CML in blast crisis, MDS – refractory anemia with excess blasts (RAEB) or in transformation (RAEB-T). Conditioning regimens were defined as “myeloablative” if the patient received total body irradiation (TBI) at a dose greater than 500 cGy if given as a single dose or greater than 800 cGy if given in fractions, or if the patient received busulfan dosed at or greater than 9.5 mg/kg, or if the patient received melphalan at a dose greater than 150 mg/m2. All other regimens were considered either “reduced intensity conditioning (RIC)” or “non-myeloablative.”7 All patients received T-replete grafts.
All patients included in this study signed informed consent for reporting of clinical information to the NMDP/CIBMTR registries in accordance with the Declaration of Helsinki. Twenty seven (1.3%) of otherwise eligible cases were removed to account for lack of consent to use the data of surviving patients or to adjust for potential bias by excluding appropriately the same percentage of deceased patients using a biased coin randomization with exclusion probabilities based on characteristics associated with not providing consent for use of the data in survivors.
High resolution typing was performed for HLA-A, B, C, DRB1, DQA1, DQB1, DPA1 and DPB1, as described.1 Low resolution (serological or antigen level) disparities involved conversion of the DNA-based typing to its lower-level serologic equivalent, usually by collapsing the 4-digit typing result back to its first 2 digits, with the exception of a few HLA-B alleles that were mapped to their corresponding serologic specificities. Antigen and allele mismatches at HLA-DRB1 were combined. Mismatches at HLA-DQ were scored if there was disparity for either the DQA1 or the DQB1 sequence, since both DQA1 and DQB1 genes contribute to the expression of a single heterodimeric HLA-DQ protein. HLA-DQA1 was not considered for determination of antigen matching. Directional mismatches (“graft versus host” or “host versus graft”) were considered as appropriate in the analysis of GVHD and engraftment, as described.8 Mismatches at homozygous alleles were considered single mismatches.
Probabilities for mortality and leukemia-free survival (LFS) were calculated using the Kaplan-Meier estimator, and survival curves were compared using the log-rank test. All other outcomes were estimated using the cumulative incidence function.9 Death was considered a competing risk for all of the endpoints except mortality and LFS. Relapse was also considered a competing event for treatment-related mortality (TRM). Patients were censored when they underwent a second hematopoietic cell transplant (HCT) procedure or if alive at last follow-up.
The association between number and type of HLA-mismatches was evaluated using separate multivariate proportional hazards models, adjusting for significant clinical covariates. Similar to the 2007 NMDP/CIBMTR paper,2 this approach compares subgroups of HLA-mismatched pairs with 8/8 matched pairs. A p-value less than 0.01 was considered statistically significant because of multiple testing.
All models were tested for significant clinical covariates including disease, disease stage, Karnofsky performance status, patient race, patient age, GVHD prophylaxis, conditioning regimen, donor age, donor-patient cytomegalovirus (CMV) serology, T cell depletion, use of total body irradiation (TBI), patient-donor sex match and year of transplant. Models adjusted for any clinical factors that were related to a given outcome at p less than 0.05. All variables were tested for affirmation of the proportional hazards assumption and to look for interactions with HLA matching. No significant interactions were identified.
Characteristics of the study population are shown in Table 1. The median follow-up of the study population was 2 years (range 0.3-7.4 years).
HLA-DQ mismatching was not statistically associated with survival if patients were otherwise matched for HLA-A, B, C, and DRB1. The relative risk for mortality for single DQ allele (n=68) or antigen (n=46) mismatch was 0.97 [0.71-1.34], p=0.87 and 1.35 [0.95-1.96]. p=0.10, respectively, compared to a full match (n=1125). There were no statistically significant differences in disease-free survival, relapse, TRM, and acute and chronic GVHD, so HLA-DQ mismatching was not considered further in determination of HLA-matching status. Information about HLA-DP was available in only 20% of donor-recipient pairs, therefore too few to be sufficient for analysis; accordingly, HLA-DP mismatching was not considered in the subsequent analyses.
Table 2 shows the association between 1 or 2 allele and/or antigen mismatches and the transplant outcomes evaluated. HLA-mismatched pairs that contained at least one antigen mismatch had statistically worse survival and disease-free survival than pairs who were 8/8 matched: however, survival of 7/8 allele mismatches was not statistically different than 8/8 matched pairs. Among 6/8 matched pairs, 29 pairs with double allele mismatches did not have worse survival than 8/8 matched pairs (RR 1.21 [0.77-1.90], p=0.42), but the small number of patients limited the power of this analysis. 6/8 matched pairs that contained at least one antigen mismatch had statistically worse survival than 8/8 matched pairs.
For TRM, any degree of HLA-mismatch was associated with worse outcomes. In contrast, HLA-mismatching was not associated with a lower risk of relapse.
Grade III-IV acute GVHD was increased with any degree of HLA-mismatching (Table 2). There was no association between number and type of HLA mismatches and grade II-IV acute GVHD or chronic GVHD.
Table 3 shows results of locus-specific analysis of single mismatched pairs for the outcomes of interest, although readers are cautioned that power is limited for some subgroups because of small sample sizes. Thus, while evidence of a statistically significant worse outcome can be accepted, absence of such a finding does not mean that a mismatch is “safe.” Mismatch of a single HLA-C antigen was associated with a statistically significant higher risk for mortality, TRM and grade III-IV acute GVHD and lower disease-free survival. ‘At two years, unadjusted survival was 32% for HLA-C mismatches compared to 44% of 8/8 matches (p=0.003), DFS was 26% compared to 40% (p=0.0002), and cumulative incidence of TRM was 40% compared to 28% (p=0.002), respectively. Mismatching at a single HLA-B allele or antigen also was associated with increased grade III-IV acute GVHD. The risks of relapse and chronic GVHD, were not statistically different for recipients of 8/8 matches compared to any locus-specific mismatch, including HLA-C antigen mismatched pairs.
Table 4 shows the comparison of HLA-C antigen mismatched pairs, other 7/8 antigen (non-C) mismatched pairs and 7/8 allele mismatched pairs compared with 8/8 matched pairs, separated by conditioning regimen intensity. Mismatching of HLA-C antigen was associated with an increase in the risk of mortality compared to 8/8 matches for patients given either myeloablative conditioning (n=122, RR 1.40 [1.10-1.78], p=0.006) or nonmyeloablative or RIC (n=65, RR 1.40 [1.01-1.95], p=0.04). In contrast, other 7/8 (non-C) antigen mismatched pairs and 7/8 allele mismatched pairs did not have statistically higher mortality than 8/8 matched pairs in either myeloablative or nonmyeloablative groups.
In order to gain insight into whether switching to marrow instead of PBSC would be advantageous if use of an HLA-mismatched donor is planned, we compared the PBSC recipients in our research dataset to recipients of marrow grafts in the analysis reported by Lee, et al.2 No statistically significant differences in mortality were seen when HLA-mismatched transplants were performed with PBSC or bone marrow. Comparing recipients of 7/8 antigen mismatched marrow (n=547) and 7/8 antigen mismatched PBSC (n=293) grafts, the relative risk for mortality at one year was not different (OR 1.13, 95% CI 0.93-1.40, p=0.26); nor was it different for the subgroups of marrow (n=321) and PBSC (n=187) recipients when the mismatch involved a single HLA-C antigen (RR 1.08, 95% CI 0.84-1.70, p=0.55). These results, which are adjusted for disease, disease status and Karnofsky performance score pre-transplant, suggest there is no advantage to changing the graft source from PBSC to marrow when using a HLA-mismatched donor even if the antigen mismatch is at HLA-C. Year of transplant and intensity of conditioning were not found to be statistically significant in these models.
This study of HLA-matching and outcomes of unrelated donor PBSC transplantation shows that HLA-mismatching in general, and HLA-C antigen and HLA-B allele and antigen mismatching in particular, are associated with statistically worse outcomes compared to 8/8 HLA-matched unrelated donors. No statistically significant association between HLA-mismatching and relapse or chronic GVHD were observed. Mismatching at HLA-DQ was not associated with statistically significantly worse outcomes and so was subsequently disregarded in determining HLA-matching. Overall, these results are similar to previous reports of the effects of HLA-mismatching in bone marrow transplantation within largely Caucasian populations.1,2
Most patients included in the previously reported sequential retrospective studies of high-resolution HLA matching received bone marrow grafts. Compared to marrow, PBSC contains on average 10-fold more CD3+ cells and 4-fold higher CD34+ cells.10 The relative contribution of cell subsets also differ, for example in PBSC the CD3:CD34 ratio is approximately 3-fold and the CD14:CD34 ratio approximately 25-fold higher, respectively, and the proportion of CD4 cells that have an anti-inflammatory (Th2) phenotype increases compared to marrow.11-14 Other studies have indicated relatively more DC2 dendritic cells and skewing of the DC1:DC2 ratio to DC2 cells within PBSC.14 All PBSC donors receive G-CSF while most unrelated bone marrow donors do not. The differences in cellular characteristics of the two products suggest, at a minimum, that the effect of HLA mismatching shown for bone marrow transplants can not be assumed to be applicable to PBSC products.
Although our results could not define an “optimal” mismatch for a PBSC transplant, they clearly show that mismatching for HLA-C antigen is associated with lower survival and higher TRM in either the single or the double mismatch setting. This conclusion is consistent with the observations from both the Flomenberg and Lee studies.2 In the Lee study, HLA-C antigen mismatch was associated with worse survival (RR 1.22, 95% CI 1.06-1.39, p=.004), as was mismatch at HLA-A antigen (RR 1.24, 95% CI 1.02-1.52, p<.001) and HLA-DRB1 allele (RR 1.42, 95% CI 1.13-1.80, p=.003), compared to the 8/8 matches.2 Our study of PBSC also showed a statistically significant relationship between HLA-B mismatching and a higher risk for grades III-IV acute GVHD although there was no association with survival. It is notable that the higher rates of severe acute GVHD observed in HLA-C antigen and HLA-B antigen and allele mismatched pairs did not translate into higher chronic GVHD or lower relapse rates. We hypothesize that this could be due to the higher TRM generally associated with grade III-IV acute GVHD, the fact that chronic GVHD is more closely linked with prevention of relapse, or that small sample sized limited power.
The main limitation of this study is the small number of observations in some of the subgroups, which may lead to erroneous estimation of the effect of a specific HLA-mismatch and limited power in comparisons. In addition, the median follow-up of 2 years is relatively short compared to the studies of HLA-matching in bone marrow transplantation. Nonmyeloablative/RIC transplants comprised 35% of our study population but are approximately 50% of procedures currently being performed. As the number of PBSC transplants increases and follow-up lengthens, a subsequent analysis will be important to update our observations, and to consider other factors such as KIR status15-19 or HLA-DP matching that could affect outcome.19,20 For example, in the bone marrow setting, the earlier Flomenberg study (2004, n=1,874)1 did not observe an increased risk associated with single allele mismatches whereas the larger Lee study (2007, n=3,857) found an adverse outcome was associated with either a single allele or antigen mismatch.2
Neither the Flomenberg nor Lee bone marrow studies included substantial numbers of nonmyeloablative (NM) or reduced intensity conditioning (RIC) transplants, which typically use PBSC grafts. A reasonable concern with these conditioning regimens is that rejection of a mismatched graft or risk for relapse might be amplified because host T or NK cells might survive less intensive conditioning. Our analysis showed that HLA-C antigen mismatching is associated with higher risk for mortality and transplant-related mortality, but not relapse after NM/RIC PBSC HCT, similar to myeloablative HCT procedures (relapse data not shown). Unfortunately, we lack data on KIR genotyping to allow refined analysis of possible natural killer effects.
In the instance when HLA-C antigen mismatching cannot be avoided, one might wonder whether a marrow graft might be better tolerated than PBSC. In an exploratory analysis, we did not detect any advantage to marrow as the cell source from a donor with an isolated HLA-C antigen mismatch. We caution that this retrospective analysis could not take into consideration all factors between the groups that might introduce bias. Our presentation of these results is not intended to address the question of whether one graft source is preferable to another but to provide the best available data pending larger studies. In an analysis by Eapen et al, outcomes of 7/8 matched bone marrow and 7/8 matched peripheral blood appeared similar although direct comparisons were not performed and locus specific data were not provided.21 In October 2009, the Blood and Marrow Transplant Clinical Trials Network finished enrollment of a 550 patient, prospective, multi-center, randomized trial to assess the risks and benefits of bone marrow versus PBSC from unrelated donors. A planned subgroup analysis of HLA-mismatched grafts has been included in the study design, the results of which will be important for addressing the issues raised in our analysis.
It is important to remember that these results are not meant to imply that an HLA-mismatched graft should not be used, only that higher risks than 8/8 HLA-matched donors should be recognized if present. For many patients, the best hope for long-term disease-free survival will still be allogeneic transplantation, even if from a less than optimal donor.
The CIBMTR is supported by Public Health Service Grant/Cooperative Agreement U24-CA76518 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 5U01HL069294 from NHLBI and NCI; a contract HHSH234200637015C with Health Resources and Services Administration (HRSA/DHHS); two Grants N00014-06-1-0704 and N00014-08-1-0058 from the Office of Naval Research; and grants from AABB; Aetna; American Society for Blood and Marrow Transplantation; Amgen, Inc.; Anonymous donation to the Medical College of Wisconsin; Astellas Pharma US, Inc.; Baxter International, Inc.; Bayer HealthCare Pharmaceuticals; Be the Match Foundation; Biogen IDEC; BioMarin Pharmaceutical, Inc.; Biovitrum AB; BloodCenter of Wisconsin; Blue Cross and Blue Shield Association; Bone Marrow Foundation; Canadian Blood and Marrow Transplant Group; CaridianBCT; Celgene Corporation; CellGenix, GmbH; Centers for Disease Control and Prevention; Children’s Leukemia Research Association; ClinImmune Labs; CTI Clinical Trial and Consulting Services; Cubist Pharmaceuticals; Cylex Inc.; CytoTherm; DOR BioPharma, Inc.; Dynal Biotech, an Invitrogen Company; Eisai, Inc.; Enzon Pharmaceuticals, Inc.; European Group for Blood and Marrow Transplantation; Gamida Cell, Ltd.; GE Healthcare; Genentech, Inc.; Genzyme Corporation; Histogenetics, Inc.; HKS Medical Information Systems; Hospira, Inc.; Infectious Diseases Society of America; Kiadis Pharma; Kirin Brewery Co., Ltd.; The Leukemia & Lymphoma Society; Merck & Company; The Medical College of Wisconsin; MGI Pharma, Inc.; Michigan Community Blood Centers; Millennium Pharmaceuticals, Inc.; Miller Pharmacal Group; Milliman USA, Inc.; Miltenyi Biotec, Inc.; National Marrow Donor Program; Nature Publishing Group; New York Blood Center; Novartis Oncology; Oncology Nursing Society; Osiris Therapeutics, Inc.; Otsuka America Pharmaceutical, Inc.; Pall Life Sciences; Pfizer Inc; Saladax Biomedical, Inc.; Schering Corporation; Society for Healthcare Epidemiology of America; Soligenix, Inc.; StemCyte, Inc.; StemSoft Software, Inc.; Sysmex America, Inc.; THERAKOS, Inc.; Thermogenesis Corporation; Vidacare Corporation; Vion Pharmaceuticals, Inc.; ViraCor Laboratories; ViroPharma, Inc.; and Wellpoint, Inc. 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, or any other agency of the U.S. Government.
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.
The authors have no conflicts of interest to declare.