DNA Methylation Changes in Multiple Myeloma

O Galm1, S Wilop1, J Reichelt1, E Jost1, G Gehbauer1, JG Herman2 and R Osieka1

1Medizinische Klinik IV, Universitaetsklinikum Aachen, 52074 Aachen, Germany; and 2The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD 21231, USA

Using a candidate gene approach, we analyzed the methylation It could be demonstrated that there are specific methylation status of the promoter-associated CpG islands of 11 well- patterns in various types of human malignancies.13 Unlike characterized tumor suppressor genes by methylation-specific genetic alterations in cancer, epigenetic changes are potentially polymerase chain reaction in five multiple myeloma (MM) cell 0 lines and 56 patients with malignant plasma cell disorders. The reversible. Demethylating agents such as 5-aza-2 -deoxycyti- frequency of aberrant methylation among the patient samples dine are capable of reactivating tumor suppressor genes that are was 46.4% for SOCS-1, 35.7% for p16, 21.4% for E-cadherin, silenced by promoter hypermethylation.4,14 Characterization of 12.5% for DAP kinase and p73, 1.8% for p15, MGMT as well as the methylation profile of candidate tumor suppressor genes that RARb, and 0% for TIMP-3, RASSF1A and hMLH1. We found at have been established to play a role in tumorigenesis may least one hypermethylated gene in 80.4% of the primary patient samples, while 33.9% harbored two or more hypermethylated provide significant insight into the molecular pathogenesis of genes. For the first time, we show that p73 may be hypermethy- MM. Furthermore, methylation patterns may be predictive for 15,16 lated in MM and thus be involved in the pathogenesis of plasma clinical behavior and response to therapy. cell disorders. Hypermethylation of p16 at diagnosis was To gain further insight into specific pathways that are relevant associated with a poorer prognosis. In patients with plasma in the multistep transformation process of normal plasma cells to cell leukemia, we found frequent simultaneous hypermethyla- MM, we have therefore applied methylation analysis of 11 tion of p16, E-cadherin and DAP kinase. We conclude that aberrant methylation of tumor suppressor genes is a common candidate genes to five MM cell lines and 56 samples from event in malignant plasma cell disorders and that there is a patients with plasma cell disorders as well as 20 nonmalignant correlation between methylation patterns and clinical charac- bone marrow samples. The list of tumor suppressor genes teristics in MM patients. includes the cell cycle regulators p15, p16 and p73, the tissue Leukemia (2004) 18, 1687–1692. doi:10.1038/sj.leu.2403434 inhibitor of matrix metalloproteinases-3 (TIMP-3), E-cadherin, Published online 19 August 2004 the DNA repair genes O6-methylguanine DNA methyltransfer- Keywords: DNA methylation; tumor suppressor genes; multiple myeloma; epigenetics; tumor marker; methylation profile ase (MGMT) and hMLH1, DAP kinase, RASSF1A, the retinoic acid receptor b (RARb) and the cytokine regulator SOCS-1. It could be shown that each of these genes may be affected by aberrant CpG island methylation in association with transcrip- Introduction tional silencing in various types of human malignancies.17–27 Multiple myeloma (MM) is a B-cell neoplasm that is characterized by the accumulation of malignant plasma cells in the bone Material and methods marrow. Previous molecular studies have largely focused on acquired genetic aberrations in MM. Important changes in DNA Human tissue samples sequence include chromosomal translocations involving the immunoglobulin heavy chain locus on chromosome 14q32 and Bone marrow (BM) and peripheral blood (PB) specimens were various partner genes such as cyclin D1, cyclin D3, fibroblast obtained during routine clinical assessment of 53 MM patients growth factor receptor 3 and c-maf, as well as mutations of N- and three patients with monoclonal gammopathy of undeter- ras and K-ras.1–3 The accumulation of genetic events is thought mined significance (MGUS), who presented at the University to be crucial for the malignant transformation of plasma cells.1 Hospital Aachen, Germany, between 1995 and 2002. MM There is increasing evidence that in addition to genetic diagnosis and staging classification were made in accordance aberrations epigenetic processes play a major role in carcino- with standard criteria.28 The main clinical characteristics of the genesis. Aberrant methylation of gene promoter regions is the patient cohort are summarized in Table 1. Control BM samples most widely studied epigenetic abnormality in human malig- were aspirated from 20 patients with nonmetastatic solid tumors nancies and is associated with loss of gene function. This or malignant lymphoma without BM infiltration or hematopoie- epigenetic event acts as an alternative to mutations and tic dysfunction as part of the routine staging procedures. deletions to disrupt tumor suppressor gene function.4,5 It was Informed consent was obtained from all patients following reported previously that hypermethylation of the cell cycle institutional guidelines. Normal PB samples were collected from inhibitors p15 and p16,6–9 the apoptosis regulator DAP kinase10 healthy volunteers. Mononuclear cells from BM and PB were and the tumor suppressor RASSF1A11 may occur in MM separated by density gradient centrifugation prior to further patients. Additionally, we have recently identified hypermethy- analysis. lation-associated silencing of the suppressor of cytokine signaling-1 (SOCS-1) gene to be a frequent epigenetic aberration in 12 MM. Cell lines

Correspondence: Dr O Galm, Medizinische Klinik IV, Universitaets- We obtained the MM cell lines U266, OPM-2, RPMI8226 and klinikum Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany; Fax: þ 49 241 8082449; E-mail: [email protected] LP-1 from the German Collection of Microorganisms and Cell Received 25 March 2004; accepted 11 June 2004; Published online Cultures (Braunschweig, Germany). XG1 cells were kindly 19 August 2004 provided by PC Heinrich (RWTH Aachen, Germany). DNA methylation in multiple myeloma O Galm et al 1688 Table 1 Patient characteristics specific for either the methylated or the unmethylated sequence. Primers for MSP analysis are designed to discriminate between Number of patients 56 methylated and unmethylated alleles after bisulfite treatment Median age, years (range) 65.5 (40–81) and to anneal only to DNA that has been bisulfite-converted. A purification of plasma cells prior to further analysis was not Gender (n) performed because of the high sensitivity of the MSP technique Male 35 Female 21 with a detection limit of approximately 0.1% of methylated DNA present in an otherwise unmethylated sample.29 Paraprotein isotype (n) Genomic DNA was isolated from cell lines and primary IgA 12 tissues using standard methods. Approximately 1 mg of DNA was IgG 33 sodium bisulfite-modified and subjected to MSP as described IgM 1 previously.29 A detailed list of MSP primers that specifically Light chain only 6 Nonsecretory 4 recognize either unmethylated (U) or methylated (M) gene sequences is given in Table 2. Reactions were hot-started at Stage (n) 951C for 5 min and held at 801C before addition of 0.625 U of MGUS 3 Taq polymerase (Sigma, St Louis, MO, USA). Temperature I2conditions for PCR for all genes except TIMP-3 were as follows: II 12 35 cycles of 951C for 30 s, 601C for 30 s and 721C for 30 s, III 39 followed by one cycle of 721C for 5 min. For TIMP-3 analysis, Plasma cell leukemia (n) 5 conditions were 35 cycles of 951C for 30 s, 591C for 30 s and 721C for 30 s, followed by one cycle of 721C for 5 min. PCR Lytic bone lesions (n) products were separated on 2.5% agarose gels and visualized by Yes 44 ethidium bromide staining. No 12

At diagnosis (n)36 During follow-up (n) 20 Statistical methods WBC, G/l (range) 6.2 (0.9–63.4) Hb, g/l (range) 113.5 (57–192) Overall survival curves were plotted according to the method of PLT, G/l (range) 225.5 (22–479) Kaplan and Meier, and compared using the log-rank test. LDH, U/l (range) 185.5 (63–908) Survival was calculated from the date of diagnosis until the Creatinine, mg/dl (range) 1.1 (0.3–6.9) patients’ death or last visit. Correlation between variables was Serum calcium, mmol/l (range) 2.3 (2.04–4.41) estimated using the Fisher’s exact two-sided test. All calculations were performed using the SAS statistical software (version 8.02, SAS Institute, Cary, NC, USA). Methylation-speciﬁc polymerase chain reaction (MSP)

MSP29 was used for qualitative analysis of the methylation status Results of the 11 tumor suppressor genes. After bisulfite treatment of genomic DNA, which converts all unmethylated cytosines to We determined the methylation status of 11 tumor suppressor uracils, the subsequent PCR amplification is performed in two genes in five MM cell lines, 56 patients with neoplastic plasma separate reactions with two different primer pairs that are cell disorders and 20 nonmalignant BM specimens. Figure 1

Table 2 Primer sets for MSP analysis

Gene Forward primer (50- 30) Reverse primer (50-30)

p15 U: GGTTGGTTTTTTATTTTGTTAGAGTGAGGT AACCACTCTAACCACAAAATACAAACACA M GGTTTTTTATTTTGTTAGAGCGAGGC TAACCGCAAAATACGAACGCG p16 U: TTATTAGAGGGTGGGGTGGATTGT CAACCCCAAACCACAACCATAA M: TTATTAGAGGGTGGGGCGGATCGC GACCCCGAACCGCGACCGTAA TIMP-3 U: TTTTGTTTTGTTATTTTTTGTTTTTGGTTTT CCCCCAAAAACCCCACCTCA M: CGTTTCGTTATTTTTTGTTTTCGGTTTC CCGAAAACCCCGCCTCG E-cadherin U: TGGTTGTAGTTATGTATTTATTTTTAGTGGTGTT ACACCAAATACAATCAAATCAAACCAAA M: TGTAGTTACGTATTTATTTTTAGTGGCGTC CGAATACGATCGAATCGAACCG MGMT U: TTTGTGTTTTGATGTTTGTAGGTTTTTGT AACTCCACACTCTTCCAAAAACAAAACA M: TTTCGACGTTCGTAGGTTTTCGC GCACTCTTCCGAAAACGAAACG p73 U: AGGGGATGTAGTGAAATTGGGGTTT ATCACAACCCCAAACATCAACATCCA M: GGACGTAGCGAAATCGGGGTTC ACCCCGAACATCGACGTCCG DAP kinase U: GGAGGATAGTTGGATTGAGTTAATGTT CAAATCCCTCCCAAACACCAA M: GGATAGTCGGATCGAGTTAACGTC CCCTCCCAAACGCCGA RASSF1A U: GGGGTTTGTTTTGTGGTTTTGTTT AACATAACCCAATTAAACCCATACTTCA M: GGGTTCGTTTTGTGGTTTCGTTC TAACCCGATTAAACCCGTACTTCG RARb U: TTGGGATGTTGAGAATGTGAGTGATTT CTTACTCAACCAATCCAACCAAAACAA M: TGTCGAGAACGCGAGCGATTC CGACCAATCCAACCGAAACGA hMLH1 U: TTTTGATGTAGATGTTTTATTAGGGTTGT ACCACCTCATCATAACTACCCACA M: ACGTAGACGTTTTATTAGGGTCGC CCTCATCGTAACTACCCGCG SOCS-1 U: TTATGAGTATTTGTGTGTATTTTTAGGTTGGTT CACTAACAACACAACTCCTACAACAACCA M: TTCGCGTGTATTTTTAGGTCGGTC CGACACAACTCCTACAACGACCG

Leukemia DNA methylation in multiple myeloma O Galm et al 1689 two or more hypermethylated gene loci (Po0.05, two-sided Fisher’s exact). Aberrant methylation of p16 was associated with lower Hb levels (P ¼ 0.04, two-sided Fisher’s exact), higher LDH levels (P ¼ 0.02, two-sided Fisher’s exact) and higher creatinine (P ¼ 0.02, two-sided Fisher’s exact). Furthermore, hypermethylation of E-cadherin and DAP kinase was associated with a lower platelet count (P ¼ 0.01 and 0.03, respectively; two-sided Fisher’s exact). There were no correlations between the methylation status of the other eight genes and any of the clinical parameters analyzed. For survival analysis only those 36 patients, from whom material was obtained at diagnosis, were included. We found that in these patients, hypermethylation of p16 was correlated with a poorer survival (P ¼ 0.03, log-rank test; Figure 3), while the methylation status of any of the other 10 genes had no prognostic impact.

Discussion

Figure 1 Representative MSP analysis of patient samples. Lanes U: The aim of this study was to determine the methylation patterns amplified products with primers recognizing the unmethylated gene of patients with malignant plasma cell disorders and explore sequence. Lanes M: amplified products with primers recognizing the possible correlations between aberrant methylation and clinical methylated gene sequence. Peripheral blood (PB) from healthy parameters. To our knowledge, this study is the first to analyze volunteers, in vitro methylated DNA (IVD) and water served as the methylation status of multiple promoter-associated CpG controls. islands from a cohort of patients with MM and MGUS. Our data provide evidence that a variety of genes regulating different pathways may be affected by aberrant methylation in MM. shows representative MSP results of the 11 genes analyzed. An We found a higher methylation frequency in MM cell lines overview of the methylation profile of the five MM cell lines and compared to the patient samples. There was aberrant methyla- the 56 patient samples is given in Figure 2. tion of up to eight of the 11 tumor suppressor genes analyzed, In MM cell lines, multiple epigenetic changes could be and all genes except p15 and hMLH1 were hypermethylated in demonstrated. Aberrant methylation of up to eight of the 11 at least one MM cell line. These findings are in accordance with genes per cell line was observed. All genes except p15 and data obtained from solid tumors,30 indicating that cancer cell hMLH1 were hypermethylated in at least one MM cell line. lines tend to exhibit higher levels of CpG island hypermethyla- We then analyzed 56 samples from patients with plasma cell tion than the primary malignancies they represent. disorders. The frequency of aberrant methylation among the The majority of the 56 patients (80.4%) exhibited hyper- patient samples was 46.4% (26/56) for SOCS-1, 35.7% (20/56) methylation of at least one of the 11 candidate genes. The genes for p16, 21.4% (12/56) for E-cadherin, 12.5% (7/56) for DAP most frequently affected by aberrant methylation were SOCS-1, kinase and p73, 1.8% (1/56) for p15, MGMT as well as RARb, p16, E-cadherin, DAP kinase and p73. Hypermethylation of and 0% (0/56) for TIMP-3, RASSF1A and hMLH1. At least one MGMT, p15 and RARb were rare events, while there was no hypermethylated gene was found in 80.4% (45/56) of the methylation of TIMP-3, hMHL1 and RASSF1A. These methyla- primary patient samples, while 33.9% (19/56) harbored two or tion patterns in malignant plasma cell disorders differ from more aberrantly methylated genes. Among the 11 candidate previously reported data in other hematopoietic malignancies. genes analyzed, SOCS-1 was most often affected by hyper- In acute leukemias, aberrant methylation of p15, E-cadherin, methylation in our patient cohort. In patients with plasma cell SOCS-1, DAP kinase and p7313,31–33 are the most common leukemia (PCL), there was frequent simultaneous hypermethyla- events, while in non-Hodgkin’s lymphomas DAP kinase, p16, tion of p16, E-cadherin and DAP kinase. For the first time, we MGMT, RARb and p7313,34 are predominantly affected by CpG demonstrate that p73 may be hypermethylated in MM. MSP for island hypermethylation. p15 was additionally performed using alternative primer pairs We found a significantly higher methylation frequency of p16, for MSP analysis,29 yielding identical results in all patient DAP kinase and E-cadherin as well as a higher total number of samples and cell lines. In contrast to the results from patients hypermethylated genes in PCL. These data from the PCL samples with MM and MGUS, no methylation of any of the 11 tumor indicate that the accumulation of epigenetic events affecting suppressor genes was found in 20 nonmalignant BM samples. genes regulating cell cycle control, cell adhesion and apoptosis We have then explored possible correlations between may in addition to genetic aberrations contribute to a more methylation data and clinical parameters of the 56 patients. aggressive phenotype in MM or reflect a molecular progression Factors analyzed included age, gender, white blood cell count from MM with few silenced genes. PCL is a distinct clinical MM (WBC), hemoglobin levels (Hb), platelet counts (PLT), creati- subgroup that appears to have a characteristic methylation nine, serum calcium levels, serum lactate dehydrogenase (LDH), profile. paraprotein isotype, disease stage and presence of lytic bone The role of p16 in MM pathogenesis has been discussed lesions. Hypermethylation of p16 (Po0.01, two-sided Fisher’s controversially. Previous reports have already shown that exact), DAP kinase (P ¼ 0.01, two-sided Fisher’s exact) and E- hypermethylation of p16 is a common phenomenon in MM cadherin (Po0.01, two-sided Fisher’s exact) was associated and MGUS.6–9,35 While two studies found a correlation of with PCL. Among PCL samples, there was a higher incidence of aberrant p16 methylation with a poorer outcome,8,9 Guillerm

Leukemia DNA methylation in multiple myeloma O Galm et al 1690

Figure 2 Methylation proﬁle of 11 tumor suppressor gene promoter regions in ﬁve MM cell lines and 56 patients with plasma cell disorders. Gray grid squares indicate methylated sites, open grid squares indicate unmethylated sites.

Leukemia DNA methylation in multiple myeloma O Galm et al 1691 Acknowledgements

We thank Sandra Mellen and Ingeborg Wiegand for expert technical assistance and Albert Esser for help with the statistical analysis. This work was supported by a grant from the Rheinisch- Westfaelische Technische Hochschule Aachen (START program) and the Samuel Waxman Cancer Research Foundation. JGH is a paid consultant to and receives research support from Onco- Methylome Sciences. The terms of this arrangement are being managed by the Johns Hopkins University in accordance with its conﬂict of interest policies.

Figure 3 Overall survival (Kaplan–Meier curve) in 36 MM patients References with or without aberrant methylation of p16 at diagnosis. 1 Hallek M, Bergsagel PL, Anderson KC. Multiple myeloma: increasing evidence for a multistep transformation process. Blood et al36 could not demonstrate a negative prognostic impact of 1998; 91: 3–21. p16 methylation. In our patient cohort, hypermethylation of p16 2 Kastrinakis NG, Gorgoulis VG, Foukas PG, Dimopoulos MA, Kittas was associated with a shortened survival, thus providing further C. Molecular aspects of multiple myeloma. Ann Oncol 2000; 11: evidence for the role of methylation markers as a prognostic 1217–1228. 3 Kuehl WM, Bergsagel PL. Multiple myeloma: evolving genetic factor. events and host interactions. Nat Rev Cancer 2002; 2: 175–187. 37 In agreement with Depil et al, we found that SOCS-1 4 Baylin SB, Herman JG. DNA hypermethylation in tumorigenesis: hypermethylation appears to have no impact on MM prognosis, epigenetics joins genetics. Trends Genet 2000; 16: 168–174. and there was no correlation of the SOCS-1 methylation status 5 Jones PA, Baylin SB. The fundamental role of epigenetic events in with clinical parameters. We discovered aberrant methylation of cancer. Nat Rev Genet 2002; 3: 415–428. p73 to be a novel epigenetic event in MM. Hypermethylation of 6 Ng MH, Chung YF, Lo KW, Wickham NW, Lee JC, Huang DP. Frequent hypermethylation of p16 and p15 genes in multiple p73 has been previously implicated in the pathogenesis of acute myeloma. Blood 1997; 89: 2500–2506. 38,39 leukemias and malignant lymphomas. 7 Gonzalez M, Mateos MV, Garcia-Sanz R, Balanzategui A, Lopez- Our results regarding the methylation status of p15 are in Perez R, Chillon MC et al. De novo methylation of tumor contrast to previously published data.6,35 Using two different suppressor gene p16/INK4a is a frequent finding in multiple primer pairs for MSP analysis, we found p15 to be hypermethy- myeloma patients at diagnosis. Leukemia 2000; 14: 183–187. lated in only one MM patient and in none of the five MM cell 8 Kramer A, Schultheis B, Bergmann J, Willer A, Hegenbart U, Ho AD et al. Alterations of the cyclin D1/pRb/p16(INK4A) pathway in lines analyzed. Given the fact that cell lines tend to carry a multiple myeloma. Leukemia 2002; 16: 1844–1851. higher number of hypermethylated genes in comparison to their 9 Mateos MV, Garcia-Sanz R, Lopez-Perez R, Moro MJ, Ocio E, corresponding primary tumors,30 we draw the conclusion that Hernandez J et al. Methylation is an inactivating mechanism the role of p15 hypermethylation in MM may have been of the p16 gene in multiple myeloma associated with high plasma overestimated. Hypermethylation of p15 appears to be restricted cell proliferation and short survival. Br J Haematol 2002; 118: mostly to acute leukemias.13 1034–1040. 10 Ng MH, To KW, Lo KW, Chan S, Tsang KS, Cheng SH et al. In summary, this study demonstrates that aberrant gene Frequent death-associated protein kinase promoter hypermethyla- promoter methylation is a common phenomenon in malignant tion in multiple myeloma. Clin Cancer Res 2001; 7: 1724–1729. plasma cell disorders. In addition to the accumulation of genetic 11 Ng MH, Lau KM, Wong WS, To KW, Cheng SH, Tsang KS et al. aberrations, this epigenetic phenomenon may further contribute Alterations of RAS signalling in Chinese multiple myeloma to the phenotype of MM cells. In order to explore possible links patients: absent BRAF and rare RAS mutations, but frequent between genetic and epigenetic events in the multistep inactivation of RASSF1A by transcriptional silencing or expression of a non-functional variant transcript. Br J Haematol 2003; 123: transformation process of plasma cells, future studies are needed 637–645. to examine a broad spectrum of methylation changes as well as 12 Galm O, Yoshikawa H, Esteller M, Osieka R, Herman JG. SOCS-1, genetic alterations in different stages of MM at diagnosis and a negative regulator of cytokine signaling, is frequently silenced by during follow-up. methylation in multiple myeloma. Blood 2003; 101: 2784–2788. Our data provide further evidence that methylation patterns 13 Esteller M, Corn PG, Baylin SB, Herman JG. A gene hypermethyla- may be useful as prognostic indicators at MM diagnosis. tion profile of human cancer. Cancer Res 2001; 61: 3225–3229. 14 Issa JP. Decitabine. Curr Opin Oncol 2003; 15: 446–451. However, the number of cases studied limits the clinical 15 Esteller M. CpG island hypermethylation and tumor suppressor implications of our results. To better define and assess the genes: a booming present, a brighter future. Oncogene 2002; 21: clinical impact of methylation markers in MM, it is therefore 5427–5440. necessary to establish a detailed methylation profile in a larger 16 Laird PW. Early detection: the power and the promise of DNA number of patient samples, where multivariable analysis will be methylation markers. Nat Rev Cancer 2003; 3: 253–266. possible. This knowledge may help to develop new markers for 17 Bachman KE, Herman JG, Corn PG, Merlo A, Costello JF, Cavenee WK et al. Methylation-associated silencing of the tissue inhibitor of detection of minimal residual disease and determine candidate metalloproteinase-3 gene suggest a suppressor role in kidney, patients for epigenetically targeted salvage therapy regimens. brain, and other human cancers. Cancer Res 1999; 59: 798–802. The demethylating drugs 5-aza-20-deoxycytidine and 5- 18 Corn PG, Kuerbitz SJ, van Noesel MM, Esteller M, Compitello N, azacytidine have recently been shown to be effective in patients Baylin SB et al. Transcriptional silencing of the p73 gene in acute with acute leukemia, myelodysplastic syndrome and chronic lymphoblastic leukemia and Burkitt’s lymphoma is associated with myelogenous leukemia.40–43 The increasing evidence for the 5’ CpG island methylation. Cancer Res 1999; 59: 3352–3356. 19 Graff JR, Herman JG, Lapidus RG, Chopra H, Xu R, Jarrard DF et al. important role of epigenetic aberrations in malignant plasma E-cadherin expression is silenced by DNA hypermethylation in cell disorders may serve as a basis for the use of demethylating human breast and prostate carcinomas. Cancer Res 1995; 55: therapeutic approaches in MM in the future. 5195–5199.

Leukemia DNA methylation in multiple myeloma O Galm et al 1692 20 Yoshikawa H, Matsubara K, Qian GS, Jackson P, Groopman JD, associated CpG islands in adult acute lymphocytic leukemia. Clin Manning JE et al. SOCS-1, a negative regulator of the JAK/STAT Cancer Res 2002; 8: 2217–2224. pathway, is silenced by methylation in human hepatocellular 33 Chen CY, Tsay W, Tang JL, Shen HL, Lin SW, Huang SY et al. carcinoma and shows growth-suppression activity. Nat Genet SOCS1 methylation in patients with newly diagnosed acute 2001; 28: 29–35. myeloid leukemia. Genes Chromosomes Cancer 2003; 37: 21 Esteller M, Hamilton SR, Burger PC, Baylin SB, Herman JG. 300–305. Inactivation of the DNA repair gene O6-methylguanine-DNA 34 Rossi D, Capello D, Gloghini A, Franceschetti S, Paulli M, Bhatia K methyltransferase by promoter hypermethylation is a common et al. Aberrant promoter methylation of multiple genes throughout event in primary human neoplasia. Cancer Res 1999; 59: the clinico-pathologic spectrum of B-cell neoplasia. Haematolo- 793–797. gica 2004; 89: 154–164. 22 Cameron EE, Baylin SB, Herman JG. p15(INK4B) CpG island 35 Guillerm G, Gyan E, Wolowiec D, Facon T, Avet-Loiseau H, methylation in primary acute leukemia is heterogeneous and Kuliczkowski K et al. p16(INK4a) and p15(INK4b) gene methyla- suggests density as a critical factor for transcriptional silencing. tions in plasma cells from monoclonal gammopathy of undeter- Blood 1999; 94: 2445–2451. mined signiﬁcance. Blood 2001; 98: 244–246. 23 Herman JG, Umar A, Polyak K, Graff JR, Ahuja N, Issa JP et al. 36 Guillerm G, Depil S, Wolowiec D, Quesnel B. Different prognostic Incidence and functional consequences of hMLH1 promoter values of p15INK4b and p16INK4a gene methylations in multiple hypermethylation in colorectal carcinoma. Proc Natl Acad Sci myeloma. Haematologica 2003; 88: 476–478. USA 1998; 95: 6870–6875. 37 Depil S, Saudemont A, Quesnel B. SOCS-1 gene methy- 24 Katzenellenbogen RA, Baylin SB, Herman JG. Hypermethylation lation is frequent but does not appear to have prognostic of the DAP-kinase CpG island is a common alteration in B-cell value in patients with multiple myeloma. Leukemia 2003; 17: malignancies. Blood 1999; 93: 4347–4353. 1678–1679. 25 Burbee DG, Forgacs E, Zochbauer-Muller S, Shivakumar L, Fong 38 Kawano S, Miller CW, Gombart AF, Bartram CR, Matsuo Y, Asou K, Gao B et al. Epigenetic inactivation of RASSF1A in lung and H et al. Loss of p73 gene expression in leukemias/lymphomas due breast cancers and malignant phenotype suppression. J Natl to hypermethylation. Blood 1999; 94: 1113–1120. Cancer Inst 2001; 93: 691–699. 39 Martinez-Delgado B, Melendez B, Cuadros M, Jose Garcia M, 26 Merlo A, Herman JG, Mao L, Lee DJ, Gabrielson E, Burger PC et al. Nomdedeu J, Rivas C et al. Frequent inactivation of the p73 gene 50 CpG island methylation is associated with transcriptional by abnormal methylation or LOH in non-Hodgkin’s lymphomas. silencing of the tumour suppressor p16/CDKN2/MTS1 in human Int J Cancer 2002; 102: 15–19. cancers. Nat Med 1995; 1: 686–692. 40 Willemze R, Suciu S, Archimbaud E, Muus P, Stryckmans P, 27 Virmani AK, Rathi A, Zochbauer-Muller S, Sacchi N, Fukuyama Y, Louwagie EA et al. A randomized phase II study on the effects of 5- Bryant D et al. Promoter methylation and silencing of the retinoic aza-2’-deoxycytidine combined with either amsacrine or idarubi- acid receptor-beta gene in lung carcinomas. J Natl Cancer Inst cin in patients with relapsed acute leukemia: an EORTC Leukemia 2000; 92: 1303–1307. Cooperative Group phase II study (06893). Leukemia 1997; 11 28 Durie BG. Staging and kinetics of multiple myeloma. Semin Oncol (Suppl 1): S24–S27. 1986; 13: 300–309. 41 Wijermans P, Lubbert M, Verhoef G, Bosly A, Ravoet C, Andre M 29 Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. et al. Low-dose 5-aza-20-deoxycytidine, a DNA hypomethylating Methylation-speciﬁc PCR: a novel PCR assay for methylation agent, for the treatment of high-risk myelodysplastic syndrome: a status of CpG islands. Proc Natl Acad Sci USA 1996; 93: multicenter phase II study in elderly patients. J Clin Oncol 2000; 9821–9826. 18: 956–962. 30 Smiraglia DJ, Rush LJ, Fruhwald MC, Dai Z, Held WA, Costello JF 42 Silverman LR, Demakos EP, Peterson BL, Kornblith AB, Holland et al. Excessive CpG island hypermethylation in cancer cell lines JC, Odchimar-Reissig R et al. Randomized controlled trial of vs primary human malignancies. Hum Mol Genet 2001; 10: azacitidine in patients with the myelodysplastic syndrome: a study 1413–1419. of the cancer and leukemia group B. J Clin Oncol 2002; 20: 31 Melki JR, Vincent PC, Clark SJ. Concurrent DNA hypermethylation 2429–2440. of multiple genes in acute myeloid leukemia. Cancer Res 1999; 43 Kantarjian HM, O’Brien S, Cortes J, Giles FJ, Faderl S, Issa JP et al. 59: 3730–3740. Results of decitabine (5-aza-20deoxycytidine) therapy in 130 32 Garcia-Manero G, Daniel J, Smith TL, Kornblau SM, Lee MS, patients with chronic myelogenous leukemia. Cancer 2003; 98: Kantarjian HM et al. DNA methylation of multiple promoter- 522–528.

Leukemia