Purpose: Resolution of aberrant epigenetic changes leading to altered gene expression during transformation and tumor progression is pertinent for mechanistic understanding of disrupted pathways in cancer. Such changes provide for biomarkers that can be applied in drug screening and improved disease management. Experimental Design: Genome-wide profiling and analyses of promoter DNA methylation, histone modifications, and gene expression of an in vitro progression model of serous ovarian adenocarcinoma were carried out. Similar in silico analyses and comparison of methylation and gene expression of early- and late-grade ovarian cancer samples in The Cancer Genome Atlas assigned a clinical relevance to our study. Candidate biomarkers were evaluated for epigenetic drug treatments in experimental animal models on a background of differing tumor cell responses arising from intratumor heterogeneity. Results: Differentially regulated genes during tumor progression were identified through the previously mentioned analyses as candidate biomarkers. In examining the tumor suppressor PTGIS as a potential biomarker for treatment with either 5-Aza-dC or TSA, 5-Aza-dC effectively stabilized cell cycling, restricted genetic instability, and derepressed PTGIS expression, while TSA led to emergence of drug-resistant progenitors lacking PTGIS expression. Profiling MEST and RXRγ for curcumin and CBB1007, respectively, indicated an inability of curcumin and CBB1007 in restricting residual tumor regenerative capabilities. Conclusions: Our study provides novel insights into epigenetic regulation in ovarian cancer progression and potential biomarkers for evaluating efficacy of epigenetic drugs in restricting residual tumor regeneration. Such approaches may assign a new functional interpretation of drug efficacy and cell tumor responses in ovarian cancer. Clin Cancer Res; 21(22); 5151–63. ©2015 AACR.

<p>Supplementary Table 1. Differentially methylated genes in correlation with their expression pattern in the A4 progression model Supplementary Table 2. Genes enriched for different histone methylation marks in A4 progression model identified through ChIP-on-chip Supplementary Table 3. Expression analyses of genes enriched with mono-, bi- and trivalent K4, K9 and K27 marks in A4 progression model Supplementary Table 4. Differentially regulated genes associated exclusively with histone marks</p>

<div>Abstract<p><b>Purpose:</b> Resolution of aberrant epigenetic changes leading to altered gene expression during transformation and tumor progression is pertinent for mechanistic understanding of disrupted pathways in cancer. Such changes provide for biomarkers that can be applied in drug screening and improved disease management.</p><p><b>Experimental Design:</b> Genome-wide profiling and analyses of promoter DNA methylation, histone modifications, and gene expression of an <i>in vitro</i> progression model of serous ovarian adenocarcinoma were carried out. Similar <i>in silico</i> analyses and comparison of methylation and gene expression of early- and late-grade ovarian cancer samples in The Cancer Genome Atlas assigned a clinical relevance to our study. Candidate biomarkers were evaluated for epigenetic drug treatments in experimental animal models on a background of differing tumor cell responses arising from intratumor heterogeneity.</p><p><b>Results:</b> Differentially regulated genes during tumor progression were identified through the previously mentioned analyses as candidate biomarkers. In examining the tumor suppressor PTGIS as a potential biomarker for treatment with either 5-Aza-dC or TSA, 5-Aza-dC effectively stabilized cell cycling, restricted genetic instability, and derepressed PTGIS expression, while TSA led to emergence of drug-resistant progenitors lacking PTGIS expression. Profiling MEST and RXRγ for curcumin and CBB1007, respectively, indicated an inability of curcumin and CBB1007 in restricting residual tumor regenerative capabilities.</p><p><b>Conclusions:</b> Our study provides novel insights into epigenetic regulation in ovarian cancer progression and potential biomarkers for evaluating efficacy of epigenetic drugs in restricting residual tumor regeneration. Such approaches may assign a new functional interpretation of drug efficacy and cell tumor responses in ovarian cancer. <i>Clin Cancer Res; 21(22); 5151–63. ©2015 AACR</i>.</p></div>

<div>Abstract<p><b>Purpose:</b> Resolution of aberrant epigenetic changes leading to altered gene expression during transformation and tumor progression is pertinent for mechanistic understanding of disrupted pathways in cancer. Such changes provide for biomarkers that can be applied in drug screening and improved disease management.</p><p><b>Experimental Design:</b> Genome-wide profiling and analyses of promoter DNA methylation, histone modifications, and gene expression of an <i>in vitro</i> progression model of serous ovarian adenocarcinoma were carried out. Similar <i>in silico</i> analyses and comparison of methylation and gene expression of early- and late-grade ovarian cancer samples in The Cancer Genome Atlas assigned a clinical relevance to our study. Candidate biomarkers were evaluated for epigenetic drug treatments in experimental animal models on a background of differing tumor cell responses arising from intratumor heterogeneity.</p><p><b>Results:</b> Differentially regulated genes during tumor progression were identified through the previously mentioned analyses as candidate biomarkers. In examining the tumor suppressor PTGIS as a potential biomarker for treatment with either 5-Aza-dC or TSA, 5-Aza-dC effectively stabilized cell cycling, restricted genetic instability, and derepressed PTGIS expression, while TSA led to emergence of drug-resistant progenitors lacking PTGIS expression. Profiling MEST and RXRγ for curcumin and CBB1007, respectively, indicated an inability of curcumin and CBB1007 in restricting residual tumor regenerative capabilities.</p><p><b>Conclusions:</b> Our study provides novel insights into epigenetic regulation in ovarian cancer progression and potential biomarkers for evaluating efficacy of epigenetic drugs in restricting residual tumor regeneration. Such approaches may assign a new functional interpretation of drug efficacy and cell tumor responses in ovarian cancer. <i>Clin Cancer Res; 21(22); 5151–63. ©2015 AACR</i>.</p></div>

<p>Supplementary Fig.1. Flow chart of approach for identification of progression associated biomarkers based on differential promoter methylation Supplementary Fig.2a. Venn diagram representing hypo- and hyper-methylated genes in the A4 progression model; b. Heat map representation of differentially expressed genes during A4 progression; c. Heat map representation of differentially expressed genes between Group1 (Grades I & II) and Group2 (Grade III) TCGA ovarian cancer tumors; d. Commonly regulated 25 genes during A4 and TCGA tumor progression through methylation and expression analyses; e. Functional annotation of differential progression associated genes Supplementary Fig.3a. Heat map representing mRNA expression pattern of genes enriched exclusively with histone marks during A4 progression model (values are in ratios of A4T/A4P because of comparative 2-color microarray); b. DAVID assigned different functional annotation to these genes. DAVID assigned different functional annotation to these genes. Supplementary Fig.4a. Quantitation of differential apoptosis mediated by epigenetic drugs during A4 progression (early apoptosis + late apoptosis); b-i. Spheroid formation capability of A4T cells (Day 12) following exposure to epigenetic drugs at their respective IC50 concentration and 3 days recovery; b-ii. Representative images of such A4T cell spheroids (20x, scale-100μm). c-i. Epigenetic drug regimes in studying effects of epigenetic drugs on A4T xenografts; c-ii. Tumor regression following drug exposure - controls (left) and treated tumors (right); d. Schematic for deriving the representative FACS profiles and quantification of subsets at various levels (Levels 1,2,3 represent PKH hierarchy, ploidy and cell cycle respectively). This resolution was used in analyzing the effects of epigenetic drugs on the background of tumor cell heterogeneity; (*p<0.05, **p< 0.01, ***p<0.001) Supplementary Fig.5. Phase contrast images representing functionality of different A4 xenograft sub-fractions following 5-Aza-dC treatment Supplementary Fig.6a. Expression modulation of epigenetic biomarkers following treatment with a.i.5-Aza-dC, a.ii.TSA, a.iii.Curcumin, a.iv.CBB1007, data generated as fold-change relative to control after normalization with beta-actin, fold-change > 1.5 (up-regulation) identified PTGIS as a biomarker for 5-Aza-dC and TSA, RXR-gamma for CBB1007 treatments while fold-change < 0.5 (down-regulation) identified MEST as a biomarker for curcumin treatment. b. Immunofluorscence imaging of associated biomarkers of each epigenetic drug in control and treated A4T cells. c. Immunofluorescence staining of PTGIS in spheroids formed by different tumor fractions based on the regenerative (PKH) hierarchy following 5-Aza-dC treatment. d. Frequency of cells in G0 phase after d.i. 5-Aza-dc/TSA d.ii. Curcumin d.iii. CBB1007 treatments in A4-T xenografts, *p<0.05, **p< 0.01, ***p<0.001.</p>