Jeongsook Yoon1. 1. Hyundai Medical Clinic Research Center, Seoul, Republic of Korea.
Abstract
BACKGROUND: We aimed to observe human leukocyte antigen (HLA) associations with human acute myeloid leukemia (AML) in a large population, in order to investigate the roles of HLA in leukemogenesis. Furthermore, we examined the HLA association according to morphological, cytogenetic, immunological, and clinical classifications. MATERIALS AND METHODS: We performed HLA genotyping, bone marrow studies, cytogenetic analyses, and fluorescence-activated cell sorting analyses. A clinical outcome database was constructed, and the HLA frequency, gene frequency, relative risk (RR), linkage disequilibrium, and the 2-locus and 3-locus haplotype frequency using the Mattiuz formula were calculated. For the healthy controls, Korean HLA data published by Park and co-workers were used. RESULTS: AML was found to be associated with HLA-C3 (RR = 1.46; p < 0.001). In the French-American-British (FAB) classification, acute myelomonocytic leukemia (AML-M4) was associated with HLA-C3 (47.2 vs. 74.1%; RR = 3.13; p = 0.005), in cytogenetic classification, del(9), which is frequently observed in AML-M4, was also associated with HLA-C3 (47.2 vs. 100%; RR = 13.43; p = 0.024), and in clinical classification, incomplete remission was associated with HLA-C3 as well (47.2 vs. 63.2%; RR = 1.92; p = 0.002). No correlations between AML and immunological classifications were observed. Moreover, and in terms of 2-locus haplotypes, AML was found to be associated with HLA-C3/B62 (HLA-C3 gene frequency 0.3415; HLA-B62 gene frequency 0.1361; linkage disequilibrium 0.0136; haplotype frequency 4.15 vs. 6.0%; p < 0.05). In clinical classification, incomplete remission (linkage disequilibrium 0.0136; haplotype frequency 4.15 vs. 13.6%; p = 0.013) and relapse (linkage disequilibrium 0.0136; haplotype frequency 4.15 vs. 71.0%; p = 0.044) were associated with HLA-C3/B62, whereas no association was observed for FAB, cytogenetic and immunological classifications. No association was observed for the 3-locus haplotype. CONCLUSION: The HLA-C3 antigen and the 2-locus haplotype are associated with AML.
BACKGROUND: We aimed to observe human leukocyte antigen (HLA) associations with humanacute myeloid leukemia (AML) in a large population, in order to investigate the roles of HLA in leukemogenesis. Furthermore, we examined the HLA association according to morphological, cytogenetic, immunological, and clinical classifications. MATERIALS AND METHODS: We performed HLA genotyping, bone marrow studies, cytogenetic analyses, and fluorescence-activated cell sorting analyses. A clinical outcome database was constructed, and the HLA frequency, gene frequency, relative risk (RR), linkage disequilibrium, and the 2-locus and 3-locus haplotype frequency using the Mattiuz formula were calculated. For the healthy controls, Korean HLA data published by Park and co-workers were used. RESULTS:AML was found to be associated with HLA-C3 (RR = 1.46; p < 0.001). In the French-American-British (FAB) classification, acute myelomonocytic leukemia (AML-M4) was associated with HLA-C3 (47.2 vs. 74.1%; RR = 3.13; p = 0.005), in cytogenetic classification, del(9), which is frequently observed in AML-M4, was also associated with HLA-C3 (47.2 vs. 100%; RR = 13.43; p = 0.024), and in clinical classification, incomplete remission was associated with HLA-C3 as well (47.2 vs. 63.2%; RR = 1.92; p = 0.002). No correlations between AML and immunological classifications were observed. Moreover, and in terms of 2-locus haplotypes, AML was found to be associated with HLA-C3/B62 (HLA-C3 gene frequency 0.3415; HLA-B62 gene frequency 0.1361; linkage disequilibrium 0.0136; haplotype frequency 4.15 vs. 6.0%; p < 0.05). In clinical classification, incomplete remission (linkage disequilibrium 0.0136; haplotype frequency 4.15 vs. 13.6%; p = 0.013) and relapse (linkage disequilibrium 0.0136; haplotype frequency 4.15 vs. 71.0%; p = 0.044) were associated with HLA-C3/B62, whereas no association was observed for FAB, cytogenetic and immunological classifications. No association was observed for the 3-locus haplotype. CONCLUSION: The HLA-C3 antigen and the 2-locus haplotype are associated with AML.