Search

The prognostic impact of the aberrant hypermethylation in response to azacytidine (AZA) remains to be determined. Therefore, we have analyzed the influence of the methylation status prior to AZA treatment on the overall survival and clinical response of myeloid malignancies. DNA methylation status of 24 tumor suppressor genes was analyzed by methylation-specific multiplex ligation-dependent probe amplification in 63 patients with myelodysplastic syndromes and acute myeloid leukemia treated with azacytidine. Most patients (73 %) showed methylation of at least one gene, but only 12 % of patients displayed ≥3 methylated genes. The multivariate analysis demonstrated that the presence of a high number (≥2) of methylated genes (P = 0.022), a high WBC count (P = 0.033), or anemia (P = 0.029) were independent prognostic factors associated with shorter overall survival. The aberrant methylation status did not correlate with the response to AZA, although four of the five patients with ≥3 methylated genes did not respond. By contrast, favorable cytogenetics independently influenced the clinical response to AZA as 64.7 % of patients with good-risk cytogenetic abnormalities responded (P = 0.03). Aberrant methylation status influences the survival of patients treated with AZA, being shorter in those patients with a high number of methylated genes.

Several studies have evaluated the prognostic value of the individual expression of certain genes in patients with myelodysplastic syndromes (MDS). However, none of them includes their simultaneous analysis by quantitative polymerase chain reaction (PCR). We evaluated relative expression levels of 14 molecular markers in 193 peripheral blood samples from untreated MDS patients using real-time PCR. Detectable WT1 expression levels, low TET2, and low IER3 gene expression were the only markers showing in univariate analysis a poor prognostic value for all treatment-free (TFS), progression-free (PFS), and overall survival (OS). In multivariate analysis, molecular parameters associated with a shorter TFS were: WT1 detection (p = 0.014), low TET2 (p = 0.002), and low IER3 expression (p = 0.025). WT1 detection (p = 0.006) and low TET2 (p = 0.006) expression were associated with a shorter PFS when multivariate analysis was carried out by including only molecular markers. Molecular values with an independent value in OS were: WT1 detection (p = 0.003), high EVI1 expression (p = 0.001), and undetectatable p15-CDKN2B (p = 0.037). WT1 expressers were associated with adverse clinical-biological features, high IPSS and WPSS scoring, and unfavorable molecular expression profile. In summary, detectable WT1 expression levels, and low TET2 and low IER3 expression in peripheral blood showed a strong association with adverse prognosis in MDS patients at diagnosis. However, WT1 was the only molecular marker displaying an independent prognostic value in both OS and TFS.