JNCI Journal of the National Cancer Institute

The following are color versions of Figures 1–4 from the Article by Alaminos et al., J Natl Cancer Inst 2004;96:1208–1219:

Files in this Data Supplement:

• Fig. 1. - Methylation-specific PCR analysis of 45 genes with promoter-associated CpG islands in 10 neuroblastoma cell lines. A) Illustrative methylation-specific polymerase chain reaction (PCR) results for seven of the genes studied. U = PCR product with primers for unmethylated DNA; M = PCR product with primers specific for methylated DNA; NL = normal adrenal medulla DNA (used as a control of unmethylated genes); IVD = in vitro-methylated normal lymphocyte DNA (positive control for methylated genes). B) Unsupervised hierarchical cluster analysis of all 45 genes in the neuroblastoma cell lines. The analysis separated the lines in two main branches: low-proliferative and high-proliferative cells. Red boxes show methylated status, and green boxes show unmethylated status. White boxes correspond to homozygous deletions. The dendrogram shown at left represents the hierarchical relationship among the genes, arranged in a binary tree. C) Illustrative examples of RNA expression analysis in neuroblastoma cell lines as assessed by reverse transcription–PCR in untreated cell lines (C) and cells treated with the demethylating agent 5-aza-2'-deoxycytidine (5AZA). Expression of the GAPDH (glyceraldehyde-3-phosphate dehydrogenase) gene was used as a control of RNA loading. Gene full names and abbreviations described in Table 1.
• Fig. 2. - Methylation-specific PCR analysis of 145 neuroblastic tumors. (A) Representative results. U = polymerase chain reaction (PCR) product with primers for unmethylated DNA; M = PCR product with primers specific for methylated DNA. B) Histogram representing the percentage of tumors showing methylation for the 10 genes as indicated. C) Histogram representing groups of tumors according to the number of genes hypermethylated (0 = 0 of 10 genes; 1 = 1 of 10 genes, etc.). Gene full names and abbreviations described in Table 1.
• Fig. 3. - Survival curves for the different comparison described in the text. A) Top panel: Kaplan–Meier survival curves for the comparison of infant (<12 months old at diagnosis) versus noninfant patients (≥12 months old at diagnosis) (P = .006). Lower panel: Survival of cases with HOXA9 promoter methylation versus cases with unmethylated HOXA9 for noninfant patients (P =.040). B) Top panel: Kaplan–Meier analysis for comparison of MYCN-amplified versus non–MYCN-amplified tumors (P<.001). Lower panel: Comparison of cases with HOXA9 promoter hypermethylation versus tumors with unmethylated HOXA9 for non–MYCN-amplified cases (P = .023). C) Significance analysis for Microarrays (SAM) for all neuroblastic tumors and for the 10 genes selected. The three genes showing positive significant association with survival are shown (δ = 0.1710; q value = 33.3333). D) Kaplan–Meier survival curves of cases with promoter-associated CpG island hypermethylation of the gene RARβ2 versus cases with no methylation of this gene. U = unmethylated; M = methylated.
• Fig. 4. - Unsupervised hierarchical cluster analysis of all neuroblastic tumors for the 10 genes selected, showing the three main branches with their corresponding clinicopathological characteristics. Boxes with red shading represent methylated genes, and boxes with green shading represent unmethylated genes. International Neuroblastoma Staging System (INSS) stages are shown after the tumor identification number. A = alive; D = dead; I = infant; NI = noninfant.