Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Mar 20.
Published in final edited form as: Bone Marrow Transplant. 2014 Jul 7;49(9):1184–1186. doi: 10.1038/bmt.2014.135

Donor-Recipient Allele Level HLA-Matching of Unrelated Cord Blood Units Reveals High Degrees of Mismatch and Alters Graft Selection

Parastoo B Dahi 1,4, Doris M Ponce 1,4, Sean Devlin 2, Katherine L Evans 1, Marissa Lubin 1, Anne Marie Gonzales 1, Courtney Byam 3, Melissa Sideroff 3, Deborah Wells 3, Sergio Giralt 1,4, Nancy A Kernan 3, Andromachi Scaradavou 3, Juliet N Barker 1,4
PMCID: PMC4367540  NIHMSID: NIHMS669157  PMID: 25000459

Abstract

The feasibility of selecting cord blood (CB) units at high-resolution human leukocyte antigen (HLA)-match has not been investigated. We analyzed the high-resolution donor-recipient HLA-match of 100 double-unit 4–6/6 HLA-A,-B antigen, -DRB1 allele-matched CB grafts (units 1a and 1b) and their back-up units (n = 377 units in total). The median cryopreserved graft dose was 2.9 × 107/kg/unit, and at high-resolution these units had a median donor-recipient HLA-allele match of 5/8 (range 2–8/8) and 6/10 (range 2–9/10), respectively. We then evaluated how often use of high-resolution HLA-match criteria would change the original graft selection to substitute one or both of the back-up units for units 1a and/or 1b. Using a model in which both a higher 8-allele HLA-match and cell dose ≥ 2.0 × 107/kg/unit were required, graft selection changed in 33% of transplants with minimal effect on cell dose (8.3% reduction). In summary, while units chosen based on HLA-A,-B antigen, -DRB1 allele-match have substantial mismatch at higher resolution, CB selection based on high-resolution HLA-match is possible in a significant proportion of patients without compromise in cell dose.

Keywords: cord blood transplantation, human leukocyte antigen allele match, unit selection

Introduction

Unrelated donor cord blood (CB) is increasingly used as an alternative hematopoietic stem cell (HSC) source in patients without suitable adult donors. Given the delayed engraftment but reduced graft-versus-host disease (GVHD) incidence for the degree of HLA-mismatch associated with CB transplantation (CBT), total nucleated cell (TNC) dose has traditionally taken priority in unit selection. HLA-match requirement has been less stringent (intermediate level resolution for HLA-A,-B antigens and high resolution for HLA-DRB1 alleles) with 0–2 mismatches allowed. While initial reports found no association between HLA-A,-B,-C,-DR,-DQ allele match and either grade II-IV acute GVHD or 2-year survival after single-unit CBT1, in a larger series Kurtzberg et al reported increased grade II-IV and III-IV acute GVHD in single-unit CBT recipients < 5/6 HLA-A,-B,-DRB1 allele matched2. A significant association between intermediate resolution HLA-C mismatch and transplant-related mortality (TRM) risk has been reported in single-unit CBT recipients3. More recently, a 1,568 patient registry series of single unit CBT recipients has reported progressively increasing TRM with increasing HLA-allele mismatch ranging from 9% in recipients of 8-allele matched units to 41% in recipients of 5 allele mismatched units with this effect being independent of cell dose and patient age4. Our group has also demonstrated a decreased day 180 grade III-IV acute GVHD incidence if the donor-recipient HLA-match of the engrafting unit of a double-unit pair was ≥ 4/6 HLA-A,-B,-DRB1 alleles5.

While comparable disease-free survival has been observed after single-unit CBT (children) or double-unit CBT (adults) and volunteer donor HSC transplantation using current unit selection criteria6–11, taken together the emerging data suggests allele level HLA-matching should now be considered in both single-unit and double-unit CBT. However, the extent of HLA-allele mismatch of the CB grafts currently used has not been widely appreciated, and how to select CB units using high-resolution HLA-matching has not been studied. The aims of this analysis were to examine the extent of allele level HLA-mismatch of 100 double-unit CB grafts transplanted at Memorial Sloan-Kettering Cancer Center (MSKCC), and to assess the feasibility of CB graft selection if high-resolution HLA-allele matching was required.

Methods

This analysis was performed in 100 consecutive double-unit CBT recipients [median age 41 years (range 1–69), median weight 66 kilograms (range 10–125)] transplanted between 1/1/2009 and 6/30/2012 for the treatment of hematologic malignancies. The 6/30/12 cut-off was used as after this time high resolution HLA-match was routinely considered in unit selection. All patients from the study period could be included as all patients and units considered for transplantation were typed at high resolution. Forty-six patients had European ancestry including 6 north-western, 9 eastern, 11 southern, and 20 mixed Europeans. Fifty-four had non-European ancestry including 17 Asian, 17 African (Hispanic and non-Hispanic), 8 white Hispanic, 7 Middle Eastern, and 5 mixed non-European backgrounds. The study was approved by the MSKCC Institutional Review and Privacy Board.

During the period of the study, unit selection (units 1a, 1b and 1–2 back-up units) was based on 4–6/6 HLA-A,-B antigen, -DRB1 allele donor-recipient HLA-match, cryopreserved TNC dose ≥ 1.5 (increased to ≥ 2.0 × 107/kg/unit in 2011), and the bank of origin as previously described12,13. HLA-match was usually given priority over TNC dose above the TNC dose threshold. The bank of origin was considered according to the MSKCC experience with that bank from the standpoint of post-thaw CD34+ cell viability as well as other factors such as reliability of information provided and processing method. RBC deplete units were given priority over RBC replete units, and other factors such as availability of attached segments for identity testing was considered mandatory. The vector of mismatch was considered in unit selection with preference given to units with a unidirectional mismatch in the GVHD direction over units that had either bidirectional mismatch or mismatch in the direction of rejection only14 but the unit-unit HLA-match was not. Following MSKCC policy, after initial review of all potential 4–6/6 matched CB units for a patient, up to 6–8 units were selected, from which 2 units were chosen for the graft (assigned as units 1a and 1b). Additionally, 1–2 back-up units were chosen in case of problems with shipping, thaw, low post-thaw CD34+ cell viability, or graft failure, as previously reported13. Back-up units were the next best from the stand-point of dose and HLA-match, but were always from domestic banks to permit shipment at short notice. While selected units were typed at HLA-A,-B,-C,-DRB1,-DQ alleles, high-resolution HLA-match at loci other than HLA-DRB1 was not used in graft selection.

Results

Extent of HLA-Mismatch at High Resolution

We evaluated 377 CB units: 200 were used as double-unit grafts (units 1a and 1b), and 177 were back-up units (77 patients had 2 back-up units and 23 had one). The median cryopreserved TNC dose of the originally selected graft (units 1a and 1b) was 2.9 × 107/kg/unit (range 1.5–17.8). At high-resolution their median donor-recipient HLA-match at HLA-A,-B,-DRB1 alleles was 4/6 (range 1–6/6). For 8 alleles (including HLA-C) it was 5/8 (range 2–8/8), and for 10 alleles (including HLA-C and -DQ) it was 6/10 (range 2–9/10). The median 8 allele HLA-match of units 1a and 1b was not different in 46 European (n = 92 units) and 54 non-European (n = 108 units) CBT recipients: 5/8 (range 2–8/8) in both groups. However, Non-Europeans received units with a greater mismatch at 10 HLA-alleles (median 6/10, range 2–9) than Europeans (7/10, range 3–9), p = 0.001. The overall distribution of the original 4–6/6 HLA-A,-B antigen, -DRB1 allele matched CB grafts at 8 HLA-alleles, the new recommended standard in CBT4, is shown in Figure 1. While “6/6” matched units were at least 5/8 HLA-allele matched, “5/6” units were as low as 3/8 matched, and “4/6” units were as low as 2/8 matched to the recipient.

Figure 1. The distribution of 8 allele donor-recipient HLA-match of 100 double-unit CB grafts (units 1a and 1b, 200 units) originally selected based on 4–6/6 HLA-A,-B antigen, -DRB1 allele match.

Figure 1

Open in a new tab

The majority of CB units (CBU) were either 5/6 or 4/6 at traditional HLA-match grade. At high resolution, 6/6 matched units were at least 5/8 HLA-allele matched, 5/6 units were as low as 3/8 matched, and 4/6 units were as low as 2/8 matched to the recipient.

Modeling the Feasibility of Unit Selection Using High Resolution HLA-Match

We then evaluated how often the use of higher resolution HLA-match criteria would have changed the original graft selection to substitute units 1a and/or 1b with one or both of the backup units (“new graft selection”), and the effect on the TNC dose (Table 1). The new graft required both a better allele level HLA-match and a cryopreserved TNC ≥ 2.0 × 107/kg/unit. Using this model, the graft would have changed in 33/100 (33%) transplants for 8 allele HLA-match. While the median HLA-match improvement was 1 allele, it ranged up to 4 alleles. Comparing the median TNC dose of the original graft and the new choice, an 8.3% and 8.5% graft TNC dose reduction was seen at 8 and 10 allele HLA-match grades, respectively. Thus, in grafts in which the unit selection changed, the effect on graft TNC dose of selecting units based on optimal high-resolution HLA-match was modest. Of the 33 patients whose graft would have changed, 15 had non-European ancestry indicating that improved matching was possible even in minority patients.

Table 1.

Change in double-unit CB graft selection using the model of higher allele (8 or 10) level HLA-match and TNC dose ≥ 2 × 107/kg/unit

Match Grade Number (%)Grafts That Would Change of 100 CB Grafts Grafts That Would Change Median (Range) Cryopreserved TNC × 107/kg
Original Graft Selection New Graft Selection
Unit 1a Unit 1b Total Unit 1a Unit 1b Total
8 Allele: HLA-A,-B,-C, -DRB1 33 (33%) 3.6 (1.9–10.5) 2.7 (1.9–6.4) 6.4 (4.2–16.9) 3.2 (2.0–7.6) 2.5 (*1.9–6.8) 5.7 (4.1–14.4)
10 Allele: HLA-A,-B,-C, -DRB1,-DQ 38 (38%) 3.6 (2.4–10.5) 2.5 (1.6–6.4) 6.3 (4.2–16.9) 3.1 (2.0–8.7) 2.5 (*1.6–7.6) 5.5 (3.8–16.3)
Open in a new tab
*

The lower limit of the range < 2.0 × 107/kg reflects one of original units that was not replaced.

Discussion

How to select CB units based on HLA-allele match is not established. We demonstrate that CB units routinely chosen for transplantation based on donor-recipient HLA-A,-B antigen, -DRB1 allele match have substantial disparity with the recipient when typed at HLA-A,-B,-C,-DRB1,-DQB1 alleles with units as low as 3/10 HLA-allele match being transplanted in patients of European ancestry and 2/10 HLA-match in those of non-European backgrounds. Moreover, the adoption of higher HLA-match criteria is feasible and will frequently change the selection of the “optimal” graft. Modeling the feasibility of such an approach in 100 recent transplants, we found this can be achieved without significant compromise of TNC dose in approximately one-third of patients if 8 allele HLA-match is considered. This is highly relevant given the recent publication by Eapen et al that suggests HLA-matching at 8 HLA-alleles should now be considered a new standard in CBT as improved HLA-match at 8 HLA-alleles mitigates TRM4. We limited our evaluation to the CB units originally selected as the best by standard criteria. However, if more units were considered, with increasing public inventory size it is likely that more patients could have had access to an even better matched unit with adequate TNC, especially in smaller children. While performing confirmatory HLA-typing at high-resolution could add cost to the search, the current CBT standard is to type HLA-DRB1 at high-resolution, and the increase associated with upgrading HLA-A,-B,-C from intermediate to high-resolution is modest (approximately $500 US). Furthermore, with current improvements in HLA methodologies, some banks are now typing CB units at the allele level, and this information will assist in future CB graft selection.

The recent Eapen analysis recommends that units with less than 4/8 HLA-allele match not be used4. However, how to trade off cell dose against high resolution HLA-match in patients whose best units are < 3 × 107/kg (the recommended lower limit of cryopreserved TNC dose in this analysis4) is not established, and will likely become even more complex when cell dose is evaluated by CD34+ cell and not TNC dose. The lower limit of acceptable HLA-match at high-resolution also requires further investigation and the comparison of transplantation results with CB units with a high degree of mismatch versus alternative treatment approaches such as transplantation with mismatched unrelated or haplo-identical donors is warranted. Whether there is any disadvantage to fully matched CB units in the setting of hematologic malignancies also remains highly controversial. The impact of allele level HLA-mismatch at different loci is also unknown and, therefore, at this time we consider mismatches at any of the HLA-A, -B, -C, -DRB1 loci equally.

In approximately two-thirds of patients in this analysis the graft choice did not change. Nonetheless, our findings have practical implications for unit selection for both single-unit and double-unit CBT. In this study, only the 2 backup units were considered for potential substitution of either units of the graft as they reflected the 3rd and 4th best units by the standards employed during the period of the study. If additional units were considered beyond the best 4 units it is possible that the percentage of grafts that would have changed would have increased. Additionally, we also agree that as the global inventory of quality units increases the number of units that are optimal from the standpoint of high resolution HLA-match and cell dose will likely increase. In the interim, this study suggests that high-resolution HLA-matching could be implemented in a significant proportion of patients of both European and non-European ancestry. This finding has special significance for smaller children in whom the TNC dose is less limiting and thus unit selection based on high resolution HLA-match is more feasible.

Acknowledgments

This work was supported in part by the Gabrielle’s Angel Foundation for Cancer Research, the Memorial Sloan-Kettering Cancer Center Society, the Translational and Integrative Medicine Research Grant, and P01 CA23766 from the National Cancer Institute, National Institutes of Health.

Footnotes

Author Contributions and Disclosure of Conflicts of Interest

P.B.D. and A. S. interpreted the data and wrote the manuscript. K.L E, M.L., A.G., C.B, R.F, M.S and D.W collected the data and wrote the manuscript, D.P, S.G, N.A.K wrote the manuscript. J.N.B. designed the study, interpreted the data, and wrote the manuscript. The authors have no relevant conflicts of interest to declare.

References

  • 1.Kogler G, Enczmann J, Rocha V, Gluckman E, Wernet P. High-resolution HLA typing by sequencing for HLA-A, -B, -C, -DR, -DQ in 122 unrelated cord blood/patient pair transplants hardly improves long-term clinical outcome. Bone Marrow Transplant. 2005;36 (12):1033–41. doi: 10.1038/sj.bmt.1705189. [DOI] [PubMed] [Google Scholar]
  • 2.Kurtzberg J, Prasad VK, Carter SL, Wagner JE, Baxter-Lowe LA, Wall D, et al. Results of the Cord Blood Transplantation Study (COBLT): clinical outcomes of unrelated donor umbilical cord blood transplantation in pediatric patients with hematologic malignancies. Blood. 2008;112(10):4318–4327. doi: 10.1182/blood-2007-06-098020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Eapen M, Klein JP, Sanz GF, Spellman S, Ruggeri A, Anasetti C, et al. Effect of donor-recipient HLA matching at HLA A, B, C, and DRB1 on outcomes after umbilical-cord blood transplantation for leukaemia and myelodysplastic syndrome: a retrospective analysis. Lancet Oncol. 2011;12(13):1214–21. doi: 10.1016/S1470-2045(11)70260-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Eapen M, Klein JP, Ruggeri A, Spellman S, Lee SJ, Anasetti C, et al. Impact of allele-level HLA matching on outcomes after myeloablative single unit umbilical cord blood transplantation for hematologic malignancy. Blood. 2014;123(1):133–40. doi: 10.1182/blood-2013-05-506253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ponce DM, Gonzales A, Lubin M, Castro-Malaspina H, Giralt S, Goldberg JD, et al. Graft-versus-Host Disease after Double-Unit Cord Blood Transplantation Has Unique Features and an Association with Engrafting Unit-to-Recipient HLA Match. Biol Blood Marrow Transplant. 2013;19(6):904–911. doi: 10.1016/j.bbmt.2013.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Eapen M, Rubinstein P, Zhang MJ, et al. Comparison of outcomes after transplantation of unrelated donor umbilical cord blood and bone marrow in children with acute leukemia. Lancet. 2007;369(9577):1947–1954. doi: 10.1016/S0140-6736(07)60915-5. [DOI] [PubMed] [Google Scholar]
  • 7.Eapen M, Rocha V, Sanz G, et al. Effect of graft source on unrelated donor haemopoietic stem-cell transplantation in adults with acute leukaemia: a retrospective analysis. Lancet Oncol. 2010;11(7):653–660. doi: 10.1016/S1470-2045(10)70127-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Brunstein CG, Gutman JA, Weisdorf DJ, Woolfrey AE, Defor TE, Gooley TA, et al. Allogeneic hematopoietic cell transplantation for hematological malignancy: relative risks and benefits of double umbilical cord blood. Blood. 2010;116(22):4693–4699. doi: 10.1182/blood-2010-05-285304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Brunstein CG, Fuchs EJ, Carter SL, Karanes C, Costa LJ, Wu J, et al. Alternative donor transplantation after reduced intensity conditioning: results of parallel phase 2 trials using partially HLA-mismatched related bone marrow or unrelated double umbilical cord blood grafts. Blood. 2011;118(2):282–8. doi: 10.1182/blood-2011-03-344853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ponce DM, Zheng J, Gonzales AM, Lubin M, Heller G, Castro-Malaspina H, et al. Reduced late mortality risk contributes to similar survival after double-unit cord blood transplantation compared with related and unrelated donor hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2011;17(9):1316–1326. doi: 10.1016/j.bbmt.2011.01.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Brunstein CG, Eapen M, Ahn KW, Appelbaum FR, Ballen KK, Champlin RE, et al. Reduced-intensity conditioning transplantation in acute leukemia: the effect of source of unrelated donor stem cells on outcomes. Blood. 2012;119(23):5591–8. doi: 10.1182/blood-2011-12-400630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Barker JN, Byam C, Scaradavou A. How I treat: the selection and acquisition of unrelated cord blood grafts. Blood. 2011;117(8):2332–2339. doi: 10.1182/blood-2010-04-280966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ponce DM, Lubin M, Gonzales AM, Byam C, Wells D, Ferrante R, et al. The use of back-up units to enhance the safety of unrelated donor cord blood transplantation. Biol Blood Marrow Transplant. 2012;18(4):648–51. doi: 10.1016/j.bbmt.2011.12.588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Stevens CE, Carrier C, Carpenter C, Sung D, Scaradavou A. HLA mismatch direction in cord blood transplantation: impact on outcome and implications for cord blood unit selection. Blood. 2011;118(14):3969–78. doi: 10.1182/blood-2010-11-317271. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES