Document KGvkNdKVLmpNzLXXBRENE1kKo

Korean J Lab Med 2010;30:469-73 DOI 10.3343/kjlm.2010.30.5.469 Brief CommunicationDiagnostic Hematology Level of HOXA5 Hypermethylation in Acute Myeloid Leukemia is Associated with Short-term Outcome Shine Young Kim, M.D.1, Sang-Hyun Hwang, M.D.1,3, Eun Joo Song, M.D.1, Ho Jin Shin, M.D.2, Joo Seop Jung, M.D.2, and Eun Yup Lee, M.D.1,3 Departments of Laboratory Medicine1 and Internal Medicine2, Medical Research Institute3, Pusan National University School of Medicine, Busan, Korea Hypermethylation of the homeobox (HOX) gene promoter leads to decreased expression of the gene during tumor development and is thought to be correlated with the clinical outcome in leukemia. In this study, we performed pyrosequencing to quantify the methylation level of HOXA5 genes in the bone marrow samples obtained from 50 patients with AML and 19 normal controls. The methylation percentage of HOXA5 in AML patients (median=65.4%, interquartile range=35.9-72.3%) was higher than that of HOXA5 in control patients (median=43.1%, interquartile range=36.7-49.6%, MannWhitney U test, P=0.012). The patients of the AML group who had a high methylation percentage (>70%) had a good prognosis with a 3-yr overall survival (OS) of 82.5%, whereas the patients with a low methylation percentage (70%) showed a 3-yr OS of 40.5% (P=0.048). Cox proportional hazards regression showed that the methylation percentages of HOXA5 were independently associated with the 3-yr OS of AML patients, regardless of their karyotypes. We propose that the quantification of HOXA5 methylation by pyrosequencing may be useful for predicting short-term prognosis in AML. However, the limitations of our study are the small sample size and its preliminary nature. Thus, a larger study should be performed to clearly determine the relationships among HOXA5 methylation levels, cytogenetics, and prognosis in AML patients. (Korean J Lab Med 2010;30:469-73) Key Words : AML, Methylation, HOXA5 Homeobox (HOX) genes are members of a transcrip- tion factor family and play a crucial role in embryonic development and in the control of differentiation of adult hematopoietic cells [1, 2]. DNA methylation in the 5region of the genes, particularly of the CpG islands, has been clearly linked to the repression of gene transcription and is known to play an important role in tumor development [3, 4]. Concerning hematologic malignan- cies, recent studies have shown that hypermethylation of the HOXA4 gene is associated with transcriptional Received : August 31, 2009 Manuscript No : KJLM09-108 Revision received : August 4, 2010 Accepted : August 13, 2010 Corresponding author : Sang-Hyun Hwang, M.D. Department of Laboratory Medicine, Pusan National University Hospital, 1-10 Ami-dong, Seo-gu, Busan 602-739, Korea Tel : +82-51-240-7418, Fax : +82-51-247-6560 E-mail : mindcatch@hanmail.net ISSN 1598-6535 The Korean Society for Laboratory Medicine repression in chronic lymphocytic leukemia and that extremely high methylation levels of HOXA5 are observed in AML, far greater than that seen in normal hematopoietic cells [5, 6]. The HOXA genes are associated not only with the development of hematologic malignancies but also with the prognosis of these conditions. The inactivation of HOXA4 and HOXA5 genes by hypermethylation is strongly correlated with progression to blast crisis in CML [6], and some recent studies have shown that increased expression of HOXA genes is correlated to the cytogenetic findings associated with poor prognosis in AML and mixed lineage leukemia [7, 8]. Nevertheless, the clear-cut correlation of HOXA5 with prognosis in AML has not been reported. In this study, we quantified the level of HOXA5 methylation by pyrosequencing and examined its correlation with the clinical outcome. We 469 470 Korean J Lab Med 2010;30:469-73 also considered other factors influencing the prognosis of AML patients. DNA was isolated from bone marrow samples obtained at diagnosis from 50 consecutive AML patients who were admitted at the Pusan National University Hospital, Busan, Korea, between October 2003 and January 2007. Peripheral blood samples were also obtained from 17 healthy male and 2 healthy female volunteers (controls). The Human Studies Committee at our institution approved this study. The patients were divided into the following 3 prognostic groups according to their cytogenetic findings as described by Grimwade et al. [9]: favorable karyotype (N=13), intermediate karyotype (N=29), and unfavorable karyotype (N=7). Cytogenetic studies failed for one of the cases. DNA was extracted from EDTA-decalcified bone marrow samples by using a DNeasy kit (Qiagen Ltd, Hamburg, Germany). The DNA concentration was measured using NanoDrop (Labtech International Ltd, Ringmer, UK). Bisulphite treatment of 2 mg of each sample was performed using the EZ DNA methylation kit (Zymo Research, Orange, CA, USA), and the modified DNA was eluted in 50 mL of 0.1Tris-EDTA (TE) buffer. PCR and pyrosequencing analysis were performed according to the established protocols [10]. The forward PCR primer was 5-GGAAATGTAATAATTTTGTTATAATGGGTTG-3, and the reverse PCR primer was 5-CTAAAACATATACTTAATTCCCTCCTAC-3, with a 5-biotin label on the reverse primer. Hot-start PCR was performed with 3 mL of bisulphite-treated DNA by using a HotStar Taq Master Mix Kit (Qiagen Ltd.). PCR was performed with 1 cycle of 95 for 6 min; 45 cycles of 95 for 30 sec, 58 for 30 sec, and 72 for 30 sec; and finally, 1 cycle of 72 for 5 min. Pyrosequencing with single-strand binding proteins (PyroGold reagents, Biotage AB, Uppsala, Sweden) was performed using the PSQ96MA System (Biotage AB), according to the manufacturer's protocol. The sequencing primer was 5GATGAGTTTTGTTTTTAG-3. Pyrosequencing primers were subsequently designed to focus on a series of 8 "target"CpG dinucleotides and thereby avoid CpGs with- Sensitivity (%) in the primer sequence. In sequencing, the assay was validated using an internal control (non-CpG cytosine in target methylation sequence region). The"mean HOXA5 methylation percentage"for each sample was calculated as the mean of the methylation percentages of 8 CpGs (mC/total C) of the promoter CpG islands. The patients were classified into 2 groups (high and low HOXA5 methylation percentage) according to a 70% cut-off, which was derived from the ROC curve analysis of the HOXA5 methylation levels of AML patients and healthy controls. Statistical correlations between methylation percentages and clinical variables were calculated using SPSS version 12.0 (SPSS Inc., Chicago, IL, USA). The 50 AML patients showed significantly higher percentages of methylation (Mann-Whitney U test, P=0.012) (median=65.4%, interquartile range=35.9-72.3%) than the 19 healthy control individuals (median=43.1%, interquartile range=36.7-49.6%). The ROC curve analysis showed that a cut-off of 70% HOXA5 methylation yielded sensitivity and specificity of 60% and 100%, respectively (Fig 1). We divided the AML patients into 2 subgroups on the basis of the 70% cut-off for HOXA5 methylation derived from the ROC curve analysis. The patients whose HOXA5 methylation percentage was greater than 70% showed more favorable karyotypes at the time of diagnosis than those whose HOXA5 methylation percentage was less than 70% (Table 1). The other clinical variables were not 100 80 60 40 20 0 0 20 40 60 80 100 100-Specificity (%) Fig 1. ROC curve analysis showed that a cut-off of 70% HOXA5 methylation yielded sensitivity and specificity of 60% and 100%, respectively. Kim SY, et al., Effect of HOXA5 Hypermethylation in AML 471 statistically significant. Among AML patients, the patients with high HOXA5 methylation percentage (>70%) showed a better 3-yr overall survival than those with low methylation percentage (HOXA5 methylation percentage 70%) (82.5% vs. 40.5%, respectively, P=0.048, Fig. 2). Multivariate Cox proportional hazards regression showed that HOXA5 methylation was an independent factor associated with the 3-yr overall survival of AML patients and with the karyotypic findings of the leukemic cells (Table 2). In this study, the level of HOXA5 methylation showed significant association with short-term mortality in AML Table 1. Characteristics of AML patients according to HOXA5 methylation Mean of MtP Less than 70% More than 70% P value (N=32) (N=18) Age (meanSD) 51.914.2 47.116.5 Sex (%) Male 15 (46.9) 13 (72.2) Female 17 (53.1) 5 (27.8) FAB classification (%) AML, unclassified 3 (9.3) 0 (0.0) M0 2 (6.3) 0 (0.0) M1 7 (21.9) 7 (38.9) M2 9 (28.1) 3 (16.7) M3 4 (12.5) 6 (33.3) M4 5 (15.6) 1 (5.6) M5 1 (3.1) 1 (5.6) M6 1 (3.1) 0 (0.0) Karyotyping (%) Unfavorable 4 (12.5) 3 (17.6) Intermediate 24 (75.0) 5 (29.4) Favorable 4 (12.5) 9 (52.9) Normal karyotype 17 (70.8) 4 (80.0) (%) among intermediate group Chemotherapy regimens (%) AI 25 (78.1) 9 (50.0) AIDA 4 (12.5) 6 (33.3) Other 3 (9.4) 3 (16.7) LDH (meanSD)* 1,597.51,464.6 1,458.91,505.3 Hb (meanSD)* 8.51.5 8.51.2 WBC (103/mL, 29.560.6 38.067.7 meanSD)* 0.201 0.452 0.004 1.000 0.114 0.303 0.973 0.412 *each value was determined at the time of diagnosis. Abbreviations: MtP, methylation percentage; AI, Ara-C+Idarubicin; AIDA, ATRA+Idarubicin; LDH, lactate dehydrogenase; FAB, French-American-British; WBC, white blood cells. patients. Interestingly, HOXA5 methylation showed correlation with the 3-yr overall survival of AML patients, regardless of other clinical variables considered at diagnosis, such as age, lactate dehydrogenase level, cytogenetic findings, and white blood cell count. However, the rate of disease-free survival (DFS) was not statistically different between the patients showing higher methylation percentages and those showing lower methylation percentages (data not shown). Because most patients in this study belonged to the intermediate and favorable groups, the mean HOXA5 methylation percentage had a limited significance. HOX genes are known to be involved in the maintenance and expansion of hematopoietic stem cells [11, 12], and HOXA5-11 genes are expressed throughout the CD34+ compartment. High methylation levels of HOXA5 are frequently observed in AML patients; in contrast, low methylation levels are observed in patients with solid tumors. HOXA5 methylation may play an important role in the 1.0 Cumulative survival percentage (%) 0.8 Methylation >70% 0.6 P=0.048 0.4 Methylation 70% 0.2 0.0 0 6 12 18 24 30 36 3-yr overall survival (months) Fig 2. Percentage of 3-yr overall survival of patients with "mean HOXA5 methylation percentage greater than 70%" and of those with "mean HOXA5 methylation less than 70%". Table 2. Cox proportional hazard regression analysis for 3-yr overall survival of AML patients Variables Hazard ratio 95% confidence interval P value HOXA5 methylation 70% vs. >70% Karyotype (favorable) Karyotype (intermediate) Karyotype (unfavorable) 5.685 - 0.635 10.455 1.308 24.714 0.020 0.111 1.809 3.651 60.427 0.611 0.009 472 Korean J Lab Med 2010;30:469-73 arrest of normal differentiation during the development of AML [10]. Even though in this study, the proportion of favorable AML karyotypes was greater in patients showing high methylation percentage (>70%) than in those with low methylation percentage (70%), hypermethylation proved to be an independent factor for the overall survival. There have been conflicting data on the prognostic impact of HOXA5 methylation. One study showed that hypermethylation of both HOXA5 and HOXA4 was strongly correlated with progression to blast crisis in CML and that hypermethylation was more strongly correlated to normal karyotype than favorable karyotype in AML patients [6]. However, Strathdee et al. [10] also reported that hypermethylation was observed in all cytogenetic-risk groups. Recently, several researches consistently reported that low HOX expression levels were correlated with favorable AML. Andreeff et al. [13] showed that downregulated HOX expression was a consistent feature in favorable AML patients. Debernardi et al. [14] showed similar results in their study. Since gene methylation downregulates its expression, low HOXA5 expression in favorable AML would be directly related to hypermethylation of HOXA5. In our study, HOXA5 was hypermethylated in the favorable group rather than in the intermediate or unfavorable groups. Our results indicate that hypermethylation shows a good correlation with prognosis in AML patients, thereby corroborating the findings of Andreeff et al. [13]. Other HOX genes (HOXA7-11 genes) are"upregulated together"in AML, and their increased expression is also correlated with poor prognosis in AML [15]. Because there is no surrogate marker other than cytogenetic classification for prognosis of AML patients, HOX gene methylation can be considered as a surrogate predictor of survival or response to therapy. In conclusion, the quantification of HOXA5 methylation level by pyrosequencing was found to be useful for predicting short-term prognosis in AML patients. This finding indicates that downregulation of HOXA5 by methylation could be related to favorable AML. However, the small sample size and the preliminary nature of this study are the limitations of our study. Thus, a larger study should be performed to clearly determine the relationships among HOXA5 methylation levels, cytogenetics, and prognosis of AML patients. ACKNOWLEDGMENT This study was supported by Medical Research Institute Grant (2009-09), Pusan National University. REFERENCES 1. Cillo C, Cantile M, Faiella A, Boncinelli E. Homeobox genes in normal and malignant cells. J Cell Physiol 2001;188:161-9. 2. Abramovich C and Humphries RK. Hox regulation of normal and leukemic hematopoietic stem cells. Curr Opin Hematol 2005;12: 210-6. 3. Bird AP and Wolffe AP. Methylation-induced repression--belts, braces, and chromatin. Cell 1999;99:451-4. 4. Herman JG and Baylin SB. Gene silencing in cancer in association with promoter hypermethylation. N Engl J Med 2003;349:2042-54. 5. Strathdee G, Sim A, Parker A, Oscier D, Brown R. Promoter hypermethylation silences expression of the HoxA4 gene and correlates with IgVh mutational status in CLL. Leukemia 2006;20:1326-9. 6. Strathdee G, Holyoake TL, Sim A, Parker A, Oscier DG, Melo JV, et al. Inactivation of HOXA genes by hypermethylation in myeloid and lymphoid malignancy is frequent and associated with poor prognosis. Clin Cancer Res 2007;13:5048-55. 7. Drabkin HA, Parsy C, Ferguson K, Guilhot F, Lacotte L, Roy L, et al. Quantitative HOX expression in chromosomally defined subsets of acute myelogenous leukemia. Leukemia 2002;16:186-95. 8. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531-7. 9. Grimwade D, Walker H, Oliver F, Wheatley K, Harrison C, Harrison G, et al. The importance of diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC AML 10 trial. The Medical Research Council Adult and Children's Leukaemia Working Parties. Blood 1998;92:2322-33. 10. Strathdee G, Sim A, Soutar R, Holyoake TL, Brown R. HOXA5 is targeted by cell-type-specific CpG island methylation in normal Kim SY, et al., Effect of HOXA5 Hypermethylation in AML 473 cells and during the development of acute myeloid leukaemia. Carcinogenesis 2007;28:299-309. 11. Vieille-Grosjean I, Roullot V, Courtois G. Identification of homeoboxcontaining genes expressed in hematopoietic blast cells. Biochem Biophys Res Commun 1992;185:785-92. 12. Moretti P, Simmons P, Thomas P, Haylock D, Rathjen P, Vadas M, et al. Identification of homeobox genes expressed in human haemopoietic progenitor cells. Gene 1994;144:213-9. 13. Andreeff M, Ruvolo V, Gadgil S, Zeng C, Coombes K, Chen W, et al. HOX expression patterns identify a common signature for favor- able AML. Leukemia 2008;22:2041-7. 14. Debernardi S, Lillington DM, Chaplin T, Tomlinson S, Amess J, Rohatiner A, et al. Genome-wide analysis of acute myeloid leukemia with normal karyotype reveals a unique pattern of homeobox gene expression distinct from those with translocation-mediated fusion events. Genes Chromosomes Cancer 2003;37:149-58. 15. Celetti A, Barba P, Cillo C, Rotoli B, Boncinelli E, Magli MC. Characteristic patterns of HOX gene expression in different types of human leukemia. Int J Cancer 1993;53:237-44.