Work

HLA alleles, blocks, and haplotypes associated with the hematological diseases of AML, ALL, MDS, and AA in the Han population of Southeastern China

Year: 2025

Type: article

Abstract: To the Editor: Previous studies have reported the increased frequency of human leukocyte antigen (HLA) alleles in patients with hematological diseases, including acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), myelodysplastic syndrome (MDS), and aplastic anemia (AA).[1] However, there is currently a lack of large-sample research analyzing HLA blocks and haplotypes in relation to hematological diseases. HLA is inherited in the form of haplotypes, and there is a strong linkage disequilibrium between adjacent HLA loci on chromosomes. HLA-A is located within the alpha block; HLA-B and C are located within the beta block; HLA-DRB1, DQB1, DRB3/4/5, and DQA1 are located within the delta block; and HLA-DPA1 and DPB1 are located within the epsilon block. When the frequencies of HLA alleles increase, the alleles at other loci linked within the same block or haplotype may also exhibit increased frequencies. This study aimed to analyze whether alleles with increased frequency are located within the same block or haplotype, and to investigate whether HLA block analysis or haplotype analysis is more meaningful for association studies of hematological diseases. In this study, 3546 hematological disease patients from southeastern China were enrolled who had undergone high-resolution HLA typing at the Department of HLA Laboratory between April 2019 and July 2022. This study was approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (Approval Number: 2020 Ethics Approval [Declaration] No.322). Informed consent was obtained from all participants. A total of 1623 AML, 929 ALL, 493 MDS, and 501 AA cases were included for allele frequency (AF) analysis. Additionally, 1217 AML, 749 ALL, 355 MDS, and 425 AA patients with five-locus haplotypes (HLA-A~B~C~DRB1~DQB1) were included, as well as 610 AML, 370 ALL, 195 MDS, and 192 AA patients with nine-locus haplotypes (HLA-A~B~C~DRB1~DQB1~DRB3/4/5~DQA1~DPA1~DPB1) for block and haplotype analysis. The beta block (HLA-B~C) and delta block (HLA-DRB1~DQB1) analyses were based on the five-locus haplotypes, while the delta block (HLA-DRB3/4/5~DRB1~DQA1~DQB1) and epsilon block (HLA-DPA1~DPB1) analyses were based on the nine-locus haplotype. The control group comprised the data from the common and well-documented allele list version 2.2 (CWD v2.2) for the HLA-A, B, C, DRB1, and DQB1 loci released by the China Marrow Donor Program,[2] the CWD data for the HLA-DRB3/4/5, DQA1, DPA1, and DPB1 loci established in our laboratory, and published five-locus and nine-locus haplotype data from our laboratory.[3,4] Next-generation sequencing was employed to perform HLA high-resolution genotyping at the HLA-A, B, C, DRB1, DQB1, DPB1, DQA1, DPA1, and DRB3/4/5 loci. HLA gene data analysis software (software copyright number: 2022SR1412487) was utilized to calculate the AFs, block frequencies (BFs), and haplotype frequencies (HFs) of patients. Statistical analysis was performed using R (v4.2.2, Vienna, Austria), and a chi-squared test was conducted to compare the AFs, BFs, and HFs between case and control groups. An odds ratio (OR) greater than 1 indicated that the allele was a susceptibility factor for the disease. To address multiple testing, Bonferroni-corrected P-value (Pc) <0.05 was considered statistically significant. HLA-A*02:07, A*11:01, A*30:01, B*38:02, B*46:01, C*01:02, DRB1*09:01, DRB1*12:02, DQB1*03:03, DQA1*01:04, DQA1*03:01, DRB3*01:01, DRB3*02:02, and DPB1*05:01 showed increased frequencies in AML patients compared with the control group. HLA-A*02:07, B*13:02, B*46:01, C*01:02, DRB1*04:06, DRB1*09:01, DQB1*02:02, DQB1*03:03, and DQA1*03:01 exhibited elevated frequencies in ALL patients compared with the control group. HLA-A*02:01, A*02:06, B*15:11, C*03:03, C*14:03, DQB1*04:01, DQB1*06:04, DQA1*03:03, DRB3*01:01, and DPB1*04:02 showed higher frequencies in MDS patients compared with the control group. The frequencies of HLA-A*02:01, A*02:06, A*30:01, B*13:02, B*40:02, B*46:01, B*54:01, C*01:02, DRB1*09:01, DRB1*15:01, DQB1*03:03, DQB1*06:02, DQA1*03:02, and DRB4*01:03 in AA patients were higher than those in the control group [Supplementary Table 1, https://links.lww.com/CM9/C359]. Following Bonferroni correction, HLA-A*11:01, C*01:02, DRB1*09:01, and DQA1*03:01 in AML patients; HLA-B*46:01, B*13:02, C*01:02, DRB1*09:01, and DQB1*03:03 in ALL patients; HLA-DQA1*03:03 and DRB3*01:01 within the MDS patients; and HLA-B*40:02, B*46:01, C*01:02, DRB1*09:01, DQB1*03:03, and DQB1*06:02 in the AA patients all continued to show elevated frequencies with statistical significance [Figure 1A]. This study found that HLA-A*02:01, A*02:06, DRB1*09:01, DQB1*03:03, DRB1*15:01, and DQB1*06:02 exhibited elevated frequencies in AA patients, while HLA-A*02:01 and DQB1*06:02 exhibited elevated frequencies in MDS patients, consistent with previous reports.[1]Figure 1: Summary of HLA alleles (A), blocks (B), and haplotypes (C) with significantly increased frequency in AML (a), ALL (b), MDS (c), and AA (d). The enlarged and bolded a/b/c/d indicates HLA alleles/blocks/haplotypes with Pc <0.05 after Bonferroni correction. In haplotypes (C), bold alleles show increased frequency, thick wireframes denote frequency-increased blocks. AA: Aplastic anemia; ALL: Acute lymphoblastic leukemia; AML: Acute myeloid leukemia; HLA: Human leukocyte antigen; MDS: Myelodysplastic syndrome.The blocks HLA-B*15:02~C*08:01, B*46:01~C*01:02, B*52:01~C*12:02, and DRB1*12:02~DQB1*03:01 showed higher frequencies in AML patients compared with the control group. The blocks HLA-B*38:02~C*07:02, DRB1*15:01~DQB1*06:01, DRB1*16:02~DQB1*05:02, DRB1*14:54~DQB1*05:02, and DPA1*02:02~DPB1*05:01 showed increased frequencies in ALL patients compared with the control group. The blocks HLA-DRB1*15:01~DQB1*06:02 and DRB1*13:01~DQB1*06:03 were observed to have elevated frequencies in MDS patients compared with the controls. The blocks HLA-B*13:01~C*03:04, DRB1*09:01~DQB1*03:03, DRB1*15:01~DQB1*06:02, DRB5*01:01~DRB1*15:01~DQA1*01:02~DQB1*06:02, and DPA1*02:01~DPB1*17:01 exhibited higher frequencies in AA patients than in the controls [Supplementary Table 2, https://links.lww.com/CM9/C359]. After Bonferroni correction, only the AML and AA groups retained statistically significant blocks. In AML patients, the HLA-B*15:02~C*08:01 and DRB1*12:02~DQB1*03:01 blocks remained statistically significant, while the statistically significant blocks in the AA patients included HLA-DRB1*09:01~DQB1*03:03, DRB1*15:01~DQB1*06:02, and DRB5*01:01~DRB1*15:01~DQA1*01:02~DQB1*06:02 [Figure 1B]. This study found that the frequency of HLA-DRB1*15:01~DQB1*06:02 increased in AA patients, which is consistent with previous reports.[5] The five-locus haplotype HLA-A*11:01~B*15:02~C*08:01~DRB1*12:02~DQB1*03:01 exhibited higher frequency in the AML patients compared with the control group. The HLA-A*02:01~B*46:01~C*01:02~DRB1*09:01~DQB1*03:03, A*02:03~B*38:02~C*07:02~DRB1*16:02~DQB1*05:02, and A*11:01~B*13:01~C*03:04~DRB1*15:01~DQB1*06:01 haplotypes showed significantly higher frequencies in ALL patients than in the control group. The HLA-A*02:01~B*40:01~C*07:02~DRB1*09:01~DQB1*03:03 haplotype exhibited increased frequency in MDS patients compared with the control group. In AA patients, the haplotypes HLA-A*02:01~B*46:01~C*01:02~DRB1*09:01~DQB1*03:03 and A*11:01~B*46:01~C*01:02~DRB1*09:01~DQB1*03:03 exhibited elevated frequencies compared with the control group [Supplementary Table 3, https://links.lww.com/CM9/C359]. The nine-locus haplotype HLA-A*02:01~B*40:01~C*07:02~DRB1*09:01~DQB1*03:03~DRB4*01:03~DQA1*03:02~DPA1*02:02~DPB1*05:01 showed elevated frequency in MDS patients in comparison to the control group. HLA-A*02:07~B*46:01~C*01:02~DRB1*08:03~DQB1*06:01~DRB3/4/5*NP~DQA1*01:03~DPA1*02:02~DPB1*02:02 showed higher frequency in AA patients than in control group [Supplementary Table 4, https://links.lww.com/CM9/C359]. After Bonferroni correction, only HLA-A*11:01~B*46:01~C*01:02~DRB1*09:01~DQB1*03:03 in AA patients exhibited a significantly higher frequency. However, the alleles and blocks with increased frequencies observed in this study, which form the haplotype HLA-A*11:01~B*15:02~C*08:01~DRB1*12:02~DQB1*03:01 in AML and the haplotype HLA-A*02:01~B*46:01~C*01:02~DRB1*09:01~DQB1*03:03 in AA, showed statistically significant differences. Haplotypes with increased frequency for both five-locus and nine-locus haplotypes, or haplotypes with frequency-increased alleles and blocks in HLA class I and class II, are considered potentially associated with hematological diseases. These HLA haplotypes are summarized in Figure 1C. In this study, HLA alleles, blocks, and haplotypes with increased frequencies were observed in a large sample of AML, ALL, MDS, and AA patients from the Chinese Han population. AML and AA patients exhibited frequency-increased alleles and blocks, and the haplotypes composed of these alleles and blocks also showed elevated frequencies. Some of the alleles with increased frequencies in MDS and AA patients in this study were consistent with previous reports.[1] However, this study found additional alleles that may be associated with hematological diseases, including HLA-DRB1*12:02 in AML patients, HLA-DRB1*09:01 and DQB1*03:03 in ALL patients, and HLA-B*46:01 and C*01:02, which exhibited significantly increased frequency in ALL and AA patients. In addition, some of the alleles discovered in this study were inconsistent with previous reports. For example, HLA-A*11:01 was observed to have an increased frequency in AML patients in this study, whereas a previous study by Wang et al[1] found that its frequency was lower than that in the CWD data. Based on previous reports, this study found increased frequencies of some blocks and haplotypes composed of alleles with elevated frequencies, an unreported finding. In AML patients, a haplotype HLA-A*11:01~B*15:02~C*08:01~DRB1*12:02~ DQB1*03:01 showed increased frequency, with elevated allele or block frequencies at all its loci. Although previous work identified many increased allele frequencies in AA patients, this study first found that the block HLA-DRB1*09:01~DQB1*03:03 and the haplotype HLA-A*02:01~B*46:01~C*01:02~DQB1*09:01~DQB1*03:03, composed of some of these alleles, are over-represented in AA patients. Therefore, HLA distribution differences between AML/AA patients and healthy individuals are more likely due to haplotype distribution differences. Combined block and haplotype analysis is more valuable than allele analysis alone. In addition, unlike MDS and AA, this study’s AML and ALL findings conflicted with past reports, maybe because of sample size diffs or diverse gene changes in patients. Compared with AA patients, leukemia patients at initial diagnosis have highly inconsistent genetic mutations and fusion genes, and HLA distribution may vary among those with different genetic changes. Therefore, HLA analysis results for leukemia patients without gene change stratification vary greatly between studies. To further investigate the correlations between HLA and hematological diseases, we analyzed how HLA alleles/blocks relate to immunology and gene mutations of AML and ALL patients at our hospital. Among AML patients with the RUNX1 mutation, the HLA-DRB1*12:02 allele was more common. In ALL patients, B-ALL (not T-ALL) had higher frequencies of alleles HLA-DRB1*09:01, DQB1*03:03 and the block HLA-DRB1*09:01~DQB1*03:03. These HLA alleles and block, which differ in genetic changes, may aid leukemia prognosis. They’re also important components of haplotypes associated with hematological diseases here. These findings may help understand how HLA affects hematological diseases and prognosis. In summary, this study outlined the distribution characteristics of HLA in Han Chinese patients with AML, ALL, MDS, and AA in southeast China, providing a basis for further research on the correlation between HLA and risk stratification, clinical treatment, and transplantation prognosis in these four hematologic diseases. Conflicts of Interest None. Funding This work was supported by the National Natural Science Foundation of China (No. 82070180).

Source: Chinese Medical Journal

DOI: 10.1097/cm9.0000000000003521