Epigenetic Modulation Leads to Re-Expression of Relapse Specific Silenced Genes and Induces Chemosensitivity in Childhood Acute Lymphoblastic Leukemia (ALL),

The poor outcome of patients with ALL who relapse and the frequent failure of conventional salvage strategies mandate novel treatment approaches. Previous work from our laboratory led to the discovery that the histone deacetylase inhibitor (HDACi) vorinostat could effectively reverse the expression of a relapse-specific gene signature and restore blast chemosensitivity in vitro (Bhatla et al ASH 2010 #3630, ASPHO 2011 # 302D). We now have also performed genome wide DNA methylation profiling in 33 matched diagnosis/relapsed paired marrow samples by Infinium Human Methylation27 Beadchip array obtained from patients enrolled on Children's Oncology Group (COG) protocols and observed that blasts at relapse are significantly more methylated compared those at diagnosis (p<0.00001) with concordant downregulation of gene expression. Based on these data, we sought to determine whether the demethylating agent 5-aza-2'-deoxycytidine decitabine could restore the expression of silenced genes and improve chemosensitivity of leukemic blasts and whether combination epigenetic therapy with decitabine and vorinostat could enhance further cell kill in vitro. Methods: B-lineage leukemia cell lines (Reh and UOCB1) and Ficoll enriched bone marrow patient samples (n=5) were treated with 1 μM of decitabine for 48–72 hours and quantitative RT-PCR (qRT-PCR) was performed on a panel of relapse specific, differentially methylated genes. To determine whether re-expression of genes was linked to reversal of promoter hypermethylation and not a secondary effect, genomic DNA was extracted from cell lines before and after treatment and methylation specific PCR (MSP) was performed. Unmethylated to Methylated amplicon ratios (U: M) were calculated using ImageJ software. To examine the effect of decitabine on chemosensitivity, cells were seeded in 96-well plate and treated with varying concentrations of decitabine in combination with vorinostat, prednisone, etoposide and doxorubicin. Cytotoxicity assays were performed using the Celltitre Glo Luminescent Cell Viability assay (Promega). Results were analyzed using the Calcusyn software (Biosoft) for combination drug effect which calculates a combination index (CI) where CI>1.1=antagonism, 0.9–1.1=additive and <0.90=synergy. Results: Following treatment with decitabine, we observed a 2–200 fold increase in expression by qRT-PCR, of the genes which were differentially hypermethylated and downregulated at relapse (CDKN2A, PTPRO, COL6A2, GATA4, WT1, APC and HOXA9) in primary samples and the Reh and UOCB1 leukemia cell lines. Further, MSP confirmed the enrichment of U: M amplicon ratio by 1.5 to 3 fold in these genes after decitabine treatment in cell lines. Cytotoxicity assays revealed additive or synergistic effects when primary samples and B-lineage cell lines were treated sequentially with decitabine followed by conventional chemotherapeutic agents (CI=0.5–1.05). Moreover, we observed that pretreatment with vorinostat and decitabine followed by chemotherapeutic agents had the most robust cell kill compared to any other combination (CI=0.2–0.5). Conclusions: Decitabine leads to re-expression of genes shown to be preferentially methylated at relapse and is synergistic when applied prior to conventional chemotherapy. Combination therapy with vorinostat results in maximal cell kill in cell lines and primary patient samples. Targeted epigenetic therapy with agents such as decitabine and vorinostat may be a promising approach to the treatment of relapsed ALL.

No relevant conflicts of interest to declare.