Volume 16, Issue 6 , Pages 865-866, June 2010
Re: An Approach to Predicting HSCT Outcome Using HLA-Mismatch Information Mapped on Protein Structure Data
Article Outline
We wish to comment on 3 aspects of the recently published article by Dudkiewicz et al [1]: (1) Table 1, which erroneously summarizes current practice in donor selection; (2) our concerns about the analyses attempting to validate the recommendations in Table 1; and (3) the use of contact energy calculations to predict immunogenicity of specific HLA mismatches.
Table 1. Corrections to the donor selection guidelines listed in the article by Dudkiewicz et al [1].
| Statements from Dudkiewicz et al [1] | Results in cited NMDP publications |
|---|---|
| Search for 10/10 match at 5 loci (HLA-A, B, C, DR, and DQ) | Both the Flomenberg et al. [2]and Lee et al. [4] studies show that 8/8 matching (HLA-A, B, C, and DRB1) was the minimum level of matching associated with highest survival; DQ mismatches did not have a significant impact on outcome. |
| Allele mismatch is less harmful than antigen mismatch | The Flomenberg et al. [2] study suggested that allele mismatches had less of an impact on outcome; however, the later and larger NMDP study by Lee and colleagues[4] showed that both types of mismatches had an indistinguishable impact on outcome. This was discussed in the updated guidelines published by Bray et al. [5]. |
| Chose a donor with a mismatch in HLA-A over a mismatch in HLA-B | In the Lee [4] study, single mismatches at HLA-A appeared to be more poorly tolerated than at HLA-B and HLA-C, but the limited number of transplants with isolated mismatches at HLA-B or HLA-C suggested that more research was needed. |
| Chose a mismatch with a lower number of amino acid substitutions between alleles | None of the NMDP publications cited addressed the impact of the number of amino acid substitutions on outcome; however, the results of the Lee et al. study [4] related to antigen versus allele differences suggest that the number of amino acids that differ do not have an impact on outcome. In fact, other published studies, some supported by the NMDP, did not find any evidence for selecting an allele with a lower number of amino acid substitutions 6, 7 |
| If there are 2 1-mismatch donors with single amino acid substitution in the same locus, choose the donor with the substitution outside the binding groove | None of the NMDP publications cited addressed the impact of amino acid differences outside of the antigen binding site. A study by Xiao et al. [8] (published after Dudkiewicz et al. [1] submitted their study) indicated that it would not be possible to analyze the impact of these differences with the current number of transplants. |
| Because graft-versus-host disease (GVHD) may increase the risk of hematopoietic stem cell transplant (HSCT)-related mortality, avoid mismatches that could lead to GVHD | The NMDP guidelines for matching published by Bray et al. [5] state that the impact of HLA matching on survival should be of primary importance. The effect of mismatching on GVHD, treatment-related mortality (TRM), and rejection should be considered in the development of a risk-adapted treatment strategy. |
First, the guidelines listed in Table 1 are not supported by the referenced 3 articles 2, 3, 4 published by the National Marrow Donor Program (NMDP) and Center for International Blood and Marrow Transplant Research (CIBMTR). Furthermore, the authors did not cite an update to the NMDP matching guidelines published by Bray et al. [5] (free full text available online), which incorporates the more recent findings by Lee et al. in 2007 [4].
Second, the authors' misrepresentation of the donor matching recommendations confounds their validation assessment because they are not testing current best practices. Dudkiewicz et al. also were not able to adjust for the numerous other clinical variables known to influence transplant outcome beyond HLA matching (disease, disease stage, patient age, cytomegalovirus [CMV] status, etc.). Our corrections to the guidelines listed in Table 1.
Finally, we look forward to learning more about the contact energy calculations used to predict immunogenicity of specific HLA mismatches. However, based on the methods and data presented in this article, we question the authors' conclusions. The authors chose to highlight 2 particular mismatches, HLAB*3501-HLAB*3503 and HLAB*2702-HLAB*2705, because they were the 2 most numerous groups in the database. However, Figure 9, showing the survival curves for these 2 groups, is based on a total of 18 patients (4% of all mismatches in the database), and not adjusted for any clinical characteristics. Conclusions based on such a small number of cases are questionable because small population imbalances can have dramatic effects.
We appreciate the opportunity to comment on the manuscript, and caution readers to refer to the most recent donor selection recommendations [5].
References
- . An approach to predicting hematopoietic stem cell transplantation outcome using HLA-mismatch information mapped on protein structure data. Biol Blood Marrow Transplant. 2009;15:1014–1025
- Impact of HLA class I and class II high-resolution matching on outcomes of unrelated donor bone marrow transplantation: HLA-C mismatching is associated with a strong adverse effect on transplantation outcome. Blood. 2004;104:1923–1930
- National Marrow Donor Program HLA-matching guidelines for unrelated marrow transplants. Biol Blood Marrow Transplant. 2003;9:610–615
- High-resolution donor-recipient HLA matching contributes to the success of unrelated donor marrow transplantation. Blood. 2007;110:4576–4583
- National marrow donor program HLA matching guidelines for unrelated adult donor hematopoietic cell transplants. Biol Blood Marrow Transplant. 2008;14(Suppl):45–53
- Allogeneic MHC class I molecules with numerous sequence differences do not elicit a CTL response. Hum Immunol. 2005;66:969–976
- Highly diverged MHC class I mismatches are acceptable for haematopoietic stem cell transplantation. Bone Marrow Transplant. 2007;40:193–200
- Evaluating the potential impact of mismatches outside the antigen recognition site in unrelated hematopoietic stem cell transplantation: HLA-DRB1*1454 and DRB1*140101. Tissue Antigens. 2009;73:595–598
PII: S1083-8791(10)00083-2
doi:10.1016/j.bbmt.2010.02.012
© 2010 American Society for Blood and Marrow Transplantation. Published by Elsevier Inc. All rights reserved.
Volume 16, Issue 6 , Pages 865-866, June 2010
