Methylation of MYLK3 gene promoter region: a biomarker to stratify surgical care in ovarian cancer in a multicentre study.

BACKGROUND: Survival benefit from surgical debulking of ovarian cancer (OC) is well established, but some women, despite total macroscopic clearance of disease, still have poor prognosis. We aimed to identify biomarkers to predict benefit from conventional surgery. METHODS: Clinical data from women debulked for high-stage OC were analysed (Hammersmith Hospital, London, UK; 2001-2014). Infinium's HumanMethylation27 array interrogated tumour DNA for differentially methylated CpG sites, correlated to survival, in patients with the least residual disease (RD; Hammersmith Array). Validation was performed using bisulphite pyrosequencing (Charité Hospital, Berlin, Germany cohort) and The Cancer Genome Atlas' (TCGA) methylation data set. Kaplan-Meier curves and Cox models tested survival. RESULTS: Altogether 803 women with serous OC were studied. No RD was associated with significantly improved overall survival (OS; hazard ratio (HR) 1.25, 95% CI 1.06-1.47; P=0.0076) and progression-free survival (PFS; HR 1.23, 95% CI 1.05-1.43; P=0.012; Hammersmith database n=430). Differentially methylated loci within FGF4, FGF21, MYLK2, MYLK3, MYL7, and ITGAE associated with survival. Patients with the least RD had significantly better OS with higher methylation of MYLK3 (Hammersmith (HR 0.51, 95% CI 0.31-0.84; P=0.01), Charité (HR 0.46, 95% CI 0.21-1.01; P=0.05), and TCGA (HR 0.64, 95% CI 0.44-0.93; P=0.02)). CONCLUSIONS: MYLK3 methylation is associated with improved OS in patients with the least RD, which could potentially be used to determine response to surgery.

Transcriptional implications of intragenic DNA methylation in the oestrogen receptor alpha gene in breast cancer cells and tissues.

BACKGROUND: DNA methylation variability regions (MVRs) across the oestrogen receptor alpha (ESR1) gene have been identified in peripheral blood cells from breast cancer patients and healthy individuals. In contrast to promoter methylation, gene body methylation may be important in maintaining active transcription. This study aimed to assess MVRs in ESR1 in breast cancer cell lines, tumour biopsies and exfoliated epithelial cells from expressed breast milk (EBM), to determine their significance for ESR1 transcription. METHODS: DNA methylation levels in eight MVRs across ESR1 were assessed by pyrosequencing bisulphite-converted DNA from three oestrogen receptor (ER)-positive and three ER-negative breast cancer cell lines. DNA methylation and expression were assessed following treatment with DAC (1 µM), or DMSO (controls). ESR1 methylation levels were also assayed in DNA from 155 invasive ductal carcinoma biopsies provided by the Breast Cancer Campaign Tissue Bank, and validated with DNA methylation profiles from the TCGA breast tumours (n = 356 ER-pos, n = 109 ER-neg). DNA methylation was profiled in exfoliated breast epithelial cells from EBM using the Illumina 450 K (n = 36) and pyrosequencing in a further 53 donor samples. ESR1 mRNA levels were measured by qRT-PCR. RESULTS: We show that ER-positive cell lines had unmethylated ESR1 promoter regions and highly methylated intragenic regions (median, 80.45%) while ER-negative cells had methylated promoters and lower intragenic methylation levels (median, 38.62%). DAC treatment increased ESR1 expression in ER-negative cells, but significantly reduced methylation and expression of ESR1 in ER-positive cells. The ESR1 promoter was unmethylated in breast tumour biopsies with high levels of intragenic methylation, independent of ER status. However, ESR1 methylation in the strongly ER-positive EBM DNA samples were very similar to ER-positive tumour cell lines. CONCLUSION: DAC treatment inhibited ESR1 transcription in cells with an unmethylated ESR1 promoter and reduced intragenic DNA methylation. Intragenic methylation levels correlated with ESR1 expression in homogenous cell populations (cell lines and exfoliated primary breast epithelial cells), but not in heterogeneous tumour biopsies, highlighting the significant differences between the in vivo tumour microenvironment and individual homogenous cell types. These findings emphasise the need for care when choosing material for epigenetic research and highlights the presence of aberrant intragenic methylation levels in tumour tissue.

Is there a Role for Epigenetic Enhancement of Immunomodulatory Approaches to Cancer Treatment?

The efficacy of cancer immunotherapy relies on the ability of the host immune system to recognise the cancer as non-self and eliminate it from the body. Whilst this is an extremely fertile area of medical research, with positive clinical trials showing durable responses, attention must be paid to the subset of patients that do not respond to these treatments. Immune surveillance and immunoediting by the host could itself select for immune-evasive tumour cells during tumour development leading to immunotherapy resistance. One such mechanism of non-efficacy or resistance is the epigenetic silencing of a specific gene required in the immunotherapy response pathway. Epigenetics is the study of the control of expression patterns in a cell via mechanisms not involving a change in DNA sequence. All tumour types show aberrant epigenetic regulation of genes involved in all the hallmarks of cancer, including immunomodulation. Inhibition of key enzymes involved in maintenance of epigenetic states is another important area of research for new treatment strategies for cancer. Could epigenetic therapies be used to successfully enhance the action of immunomodulatory agents in cancer, and are they acting in the way we imagine? An understanding of the effects of epigenetic therapies on immunological pathways in both the tumour and host cells, especially the tumour microenvironment, will be essential to further develop such combination approaches.

Clinical Use of Programmed Cell Death-1 and Its Ligand Expression as Discriminatory and Predictive Markers in Ovarian Cancer.

Purpose: We aimed to establish whether programmed cell death-1 (PD-1) and programmed cell death ligand 1 (PD-L1) expression, in ovarian cancer tumor tissue and blood, could be used as biomarkers for discrimination of tumor histology and prognosis of ovarian cancer.Experimental Design: Immune cells were separated from blood, ascites, and tumor tissue obtained from women with suspected ovarian cancer and studied for the differential expression of possible immune biomarkers using flow cytometry. PD-L1 expression on tumor-associated inflammatory cells was assessed by immunohistochemistry and tissue microarray. Plasma soluble PD-L1 was measured using sandwich ELISA. The relationships among immune markers were explored using hierarchical cluster analyses.Results: Biomarkers from the discovery cohort that associated with PD-L1+ cells were found. PD-L1+ CD14+ cells and PD-L1+ CD11c+ cells in the monocyte gate showed a distinct expression pattern when comparing benign tumors and epithelial ovarian cancers (EOCs)-confirmed in the validation cohort. Receiver operating characteristic curves showed PD-L1+ and PD-L1+ CD14+ cells in the monocyte gate performed better than the well-established tumor marker CA-125 alone. Plasma soluble PD-L1 was elevated in patients with EOC compared with healthy women and patients with benign ovarian tumors. Low total PD-1+ expression on lymphocytes was associated with improved survival.Conclusions: Differential expression of immunological markers relating to the PD-1/PD-L1 pathway in blood can be used as potential diagnostic and prognostic markers in EOC. These data have implications for the development and trial of anti-PD-1/PD-L1 therapy in ovarian cancer. Clin Cancer Res; 23(13); 3453-60. ©2016 AACR.

DNA methylation profiling to assess pathogenicity of BRCA1 unclassified variants in breast cancer.

Germline pathogenic mutations in BRCA1 increase risk of developing breast cancer. Screening for mutations in BRCA1 frequently identifies sequence variants of unknown pathogenicity and recent work has aimed to develop methods for determining pathogenicity. We previously observed that tumor DNA methylation can differentiate BRCA1-mutated from BRCA1-wild type tumors. We hypothesized that we could predict pathogenicity of variants based on DNA methylation profiles of tumors that had arisen in carriers of unclassified variants. We selected 150 FFPE breast tumor DNA samples [47 BRCA1 pathogenic mutation carriers, 65 BRCAx (BRCA1-wild type), 38 BRCA1 test variants] and analyzed a subset (n=54) using the Illumina 450K methylation platform, using the remaining samples for bisulphite pyrosequencing validation. Three validated markers (BACH2, C8orf31, and LOC654342) were combined with sequence bioinformatics in a model to predict pathogenicity of 27 variants (independent test set). Predictions were compared with standard multifactorial likelihood analysis. Prediction was consistent for c.5194-12G>A (IVS 19-12 G>A) (P>0.99); 13 variants were considered not pathogenic or likely not pathogenic using both approaches. We conclude that tumor DNA methylation data alone has potential to be used in prediction of BRCA1 variant pathogenicity but is not independent of estrogen receptor status and grade, which are used in current multifactorial models to predict pathogenicity.