Epigenetic profiling reveals etiologically distinct patterns of DNA methylation in head and neck squamous cell carcinoma

Thirteen-gene DNA methylation analysis of oral brushing samples: A potential surveillance tool for periodic monitoring of treated patients with oral cancer

BACKGROUND

We evaluated the prognostic role of 13-gene DNA methylation analysis by oral brushing repeatedly performed during the follow-up of patients surgically treated for oral cancer.

METHODS

This is a nested case-control study including 61 patients for a total of 64 outcomes (2/61 patients experienced multiple relapses). Samples were collected at baseline (4-10 months after OSCC resection) and repeatedly every 4-10 months until relapse or death. DNA methylation scores were classified as persistently positive, persistently negative, or mixed.

RESULTS

Twenty cases who had persistently positive scores and 30 cases with mixed scores had, respectively, an almost 42-fold (p < 0.001) and 32-fold (p = 0.006) higher likelihood of relapse, compared to 14 patients with persistently negative scores. The last score before reoccurrence was positive in 18/19 secondary events.

CONCLUSIONS

The 13-gene DNA methylation analysis may be considered for the surveillance of patients treated for oral carcinoma.

A pilot study to evaluate the expression of microRNA‑let‑7a in patients with intestinal‑type sinonasal adenocarcinoma

Despite its histological resemblance to colorectal adenocarcinoma, there is little information about the molecular events involved in the pathogenesis of intestinal-type sinonasal adenocarcinoma (ITAC). The present study investigated the possible role and clinical value of microRNA (miR)-let-7a, a head and neck squamous cell carcinoma-related miR, in a well-characterized and homogeneous cohort of patients with ethmoidal ITAC associated with occupational exposure, treated by primary surgery. miR-let-7a expression levels were analyzed in 23 pairs of ethmoidal ITAC and adjacent normal formalin-fixed paraffin-embedded tissues by reverse transcription-quantitative PCR. The expression was evaluated in tumor and healthy tissues according to: Tumor grade (G) of differentiation and extension, and pTNM stage, and presence/absence of recurrence. Comparisons within and between groups were performed using two-tailed Student's paired t-test and one-way ANOVA with Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference. miR-let-7a expression in ethmoidal ITAC tissues was significantly lower than that in adjacent normal tissues (P<0.05; mean expression level ± SD, 1.452707±1.4367189 vs. 4.094017±2.7465375). miR expression varied with pT stage. miR-let-7a was downregulated (P<0.05) in advanced stages (pT3-pT4) compared with earlier stages (pT1-pT2). Furthermore, downregulation of miR-let-7a in ITAC was associated with poorly-differentiated (G3) cancer (P<0.05). No other associations were observed between miR-let-7a expression and the other clinicopathological parameters, including disease-free survival. In conclusion, downregulation of miR-let-7a in ITAC was associated with advanced-stage (pT3 and pT4) and poorly-differentiated (G3) disease, suggesting that the mutation of this gene, combined with additional genetic events, could serve a role in ITAC pathogenesis.

Shared epigenetic alterations between oral cancer and periodontitis: a preliminary study

INTRODUCTION

We recently developed a non-invasive sampling procedure for oral squamous cell carcinoma (OSCC) detection based on DNA methylation analysis of a panel of 13 genes. Oral cancer, as well as acute and chronic inflammatory diseases, may influence the methylation level of several genes in the oral cavity. In the present study, we evaluated the presence of periodontal disease(PD) and the methylation status using our 13-gene panel.

METHODS

Oral brushing specimens were collected from three different patient groups: 23 gingival OSCC patients, 15 patients affected by PD, and 15 healthy volunteers lacking evidence of PD. DNA methylation analysis was performed and each sample was determined to be positive or negative based on a predefined cut-off value.

RESULTS

Positive results were found for 23/23 OSCC patients, 3/15 PD patients and 0/15 samples from healthy volunteers. The GP1BB and MIR193 genes in the PD group exhibited mean methylation levels similar to OSCC patients. ZAP70 showed different methylation levels among three groups.

CONCLUSION

Preliminary data identified shared epigenetic alterations between PD and OSCC patients in two inflammatory genes(GP1BB and MIR193). This study may help identify potential links between the two diseases and serve as a starting point for future research focused on pathogenesis.

Alcohol and Head and Neck Cancer: Updates on the Role of Oxidative Stress, Genetic, Epigenetics, Oral Microbiota, Antioxidants, and Alkylating Agents

Head and neck cancer (HNC) concerns more than 890,000 patients worldwide annually and is associated with the advanced stage at presentation and heavy outcomes. Alcohol drinking, together with tobacco smoking, and human papillomavirus infection are the main recognized risk factors. The tumorigenesis of HNC represents an intricate sequential process that implicates a gradual acquisition of genetic and epigenetics alterations targeting crucial pathways regulating cell growth, motility, and stromal interactions. Tumor microenvironment and growth factors also play a major role in HNC. Alcohol toxicity is caused both directly by ethanol and indirectly by its metabolic products, with the involvement of the oral microbiota and oxidative stress; alcohol might enhance the exposure of epithelial cells to carcinogens, causing epigenetic modifications, DNA damage, and inaccurate DNA repair with the formation of DNA adducts. Long-term markers of alcohol consumption, especially those detected in the hair, may provide crucial information on the real alcohol drinking of HNC patients. Strategies for prevention could include food supplements as polyphenols, and alkylating drugs as therapy that play a key role in HNC management. Indeed, polyphenols throughout their antioxidant and anti-inflammatory actions may counteract or limit the toxic effect of alcohol whereas alkylating agents inhibiting cancer cells’ growth could reduce the carcinogenic damage induced by alcohol. Despite the established association between alcohol and HNC, a concerning pattern of alcohol consumption in survivors of HNC has been shown. It is of primary importance to increase the awareness of cancer risks associated with alcohol consumption, both in oncologic patients and the general population, to provide advice for reducing HNC prevalence and complications.

Association Between Smoking and Molecular Subtypes of Colorectal Cancer

Background

Smoking is associated with colorectal cancer (CRC) risk. Previous studies suggested this association may be restricted to certain molecular subtypes of CRC, but large-scale comprehensive analysis is lacking.

Methods

A total of 9789 CRC cases and 11 231 controls of European ancestry from 11 observational studies were included. We harmonized smoking variables across studies and derived sex study-specific quartiles of pack-years of smoking for analysis. Four somatic colorectal tumor markers were assessed individually and in combination, including BRAF mutation, KRAS mutation, CpG island methylator phenotype (CIMP), and microsatellite instability (MSI) status. A multinomial logistic regression analysis was used to assess the association between smoking and risk of CRC subtypes by molecular characteristics, adjusting for age, sex, and study. All statistical tests were 2-sided and adjusted for Bonferroni correction.

Results

Heavier smoking was associated with higher risk of CRC overall and stratified by individual markers (P trend < .001). The associations differed statistically significantly between all molecular subtypes, which was the most statistically significant for CIMP and BRAF. Compared with never-smokers, smokers in the fourth quartile of pack-years had a 90% higher risk of CIMP-positive CRC (odds ratio = 1.90, 95% confidence interval = 1.60 to 2.26) but only 35% higher risk for CIMP-negative CRC (odds ratio = 1.35, 95% confidence interval = 1.22 to 1.49; P difference = 2.1 x 10-6). The association was also stronger in tumors that were CIMP positive, MSI high, or KRAS wild type when combined (P difference < .001).

Conclusion

Smoking was associated with differential risk of CRC subtypes defined by molecular characteristics. Heavier smokers had particularly higher risk of CRC subtypes that were CIMP positive and MSI high in combination, suggesting that smoking may be involved in the development of colorectal tumors via the serrated pathway.

Validity of a method for identifying disease subtypes that are etiologically heterogeneous

A focus of cancer epidemiologic research has become the identification of risk factors that influence specific subtypes of disease, a phenomenon known as etiologic heterogeneity. In previous work we developed a novel strategy to cluster tumor markers and identify disease subtypes that differ maximally with respect to known risk factors for use in the context of case-control studies. The method relies on the premise that unsupervised k-means clustering will find candidate solutions that are closely aligned with the sought-after etiologically distinct clusters, which may not be true in the presence of clusters of tumor markers that are not related to risk of disease. In this article, we investigate in detail the ability of the method to identify the "true" clusters in the presence of clusters that are unrelated to risk factors, what we term "counterfeit" clusters. We find that our method works when the strength of structure is larger in the clusters that truly represent etiologic heterogeneity than in the counterfeit clusters, but when this condition is not met, or when there are many tumor markers that simply represent noise, the method will not find the correct solution without first performing variable selection to identify the tumor markers most strongly related to the risk factors. We illustrate the results using data from a breast cancer case-control study.

DNA Methylation Markers from Negative Surgical Margins Can Predict Recurrence of Oral Squamous Cell Carcinoma

The identification of molecular markers in negative surgical margins of oral squamous cell carcinoma (OSCC) might help in identifying residual molecular aberrations, and potentially improve the prediction of prognosis. We performed an Infinium MethylationEPIC BeadChip array on 32 negative surgical margins stratified based on the status of tumor recurrence in order to identify recurrence-specific aberrant DNA methylation (DNAme) markers. We identified 2512 recurrence-associated Differentially Methylated Positions (DMPs) and 392 Differentially Methylated Regions (DMRs) which were enriched in cell signaling and cancer-related pathways. A set of 14-CpG markers was able to discriminate recurrent and non-recurrent cases with high specificity and sensitivity rates (AUC 0.98, p = 3 × 10-6; CI: 0.95-1). A risk score based on the 14-CpG marker panel was applied, with cases classified within higher risk scores exhibiting poorer survival. The results were replicated using tumor-adjacent normal HNSCC samples from The Cancer Genome Atlas (TCGA). We identified residual DNAme aberrations in the negative surgical margins of OSCC patients, which could be informative for patient management by improving therapeutic intervention. This study proposes a novel DNAme-based 14-CpG marker panel as a promising predictor for tumor recurrence, which might contribute to improved decision-making for the personalized treatment of OSCC cases.

Portrait of DNA methylated genes predictive of poor prognosis in head and neck cancer and the implication for targeted therapy

In addition to chronic infection with human papilloma virus (HPV) and exposure to environmental carcinogens, genetic and epigenetic factors act as major risk factors for head and neck cancer (HNC) development and progression. Here, we conducted a systematic review in order to assess whether DNA hypermethylated genes are predictive of high risk of developing HNC and/or impact on survival and outcomes in non-HPV/non-tobacco/non-alcohol associated HNC. We identified 85 studies covering 32,187 subjects where the relationship between DNA methylation, risk factors and survival outcomes were addressed. Changes in DNA hypermethylation were identified for 120 genes. Interactome analysis revealed enrichment in complex regulatory pathways that coordinate cell cycle progression (CCNA1, SFN, ATM, GADD45A, CDK2NA, TP53, RB1 and RASSF1). However, not all these genes showed significant statistical association with alcohol consumption, tobacco and/or HPV infection in the multivariate analysis. Genes with the most robust HNC risk association included TIMP3, DCC, DAPK, CDH1, CCNA1, MGMT, P16, MINT31, CD44, RARβ. From these candidates, we further validated CD44 at translational level in an independent cohort of 100 patients with tongue cancer followed-up beyond 10 years. CD44 expression was associated with high-risk of tumor recurrence and metastasis (P = 0.01) in HPV-cases. In summary, genes regulated by methylation play a modulatory function in HNC susceptibility and it represent a critical therapeutic target to manage patients with advanced disease.

Emerging noninvasive methylation biomarkers of cancer prognosis and drug response prediction

Cancer is the second leading cause of death worldwide being responsible for 9.6 million deaths in 2018. Epigenetic alterations are key in directing the aberrant expression of tumor-associated genes that drive cellular malignant transformation and cancer progression. Among epigenetic alterations, DNA methylation is the most deeply studied one in relation to environmental exposure. Tissue biopsies have traditionally been the main procedure by which a small sample of body tissue is excised to confirm cancer diagnosis or to indicate the primary site when cancer has spread. In contrast, the analysis of circulating tumor-derived material, or tumor circulome, by means of liquid biopsy of peripheral blood, urine, saliva or sputum is a noninvasive, fast and reproducible alternative to tissue biopsy. Recently, the assessment of epigenetic alterations such as DNA methylation and hydroxymethylation in circulating free DNA has been proved possible. These marks can be associated to prognosis and response to a variety of treatments including chemotherapy, hormonotherapy or immunotherapy. Epigenetic biomarkers may offer some advantages over RNA or genetic biomarkers given their stability in bodily fluids and their high tissue-specificity. While many challenges are still ahead, the unique advantages of these types of biomarkers is urging the scientific community to persevere in their clinical validation and integration into reliable prediction models. This review aims at recapitulating the emerging noninvasive DNA methylated biomarkers of importance for prediction of prognosis and drug response in cancer.

Validation of oral brushing as a non-invasive technique for the identification of feline oral squamous cell carcinoma by DNA methylation and TP53 mutation analysis

Feline oral squamous cell carcinoma (FOSCC) is a frequent and progressively invasive tumour. Early lesions are difficult to recognize based on the sole clinical examination and may be misinterpreted as non-neoplastic. Mutations of TP53 and epigenetic alterations of specific genes are present in FOSCC and may be early detected. Aim of this prospective study was to investigate the DNA methylation pattern of a 17-gene panel and TP53 mutational status of FOSCC cytological samples obtained by oral brushing. Results were compared with a control group, in order to validate this non-invasive procedure for the screening of FOSCC. In FOSCC, the same analyses were carried out on the corresponding histological sample, if available. Thirty-five FOSCC and 60 controls were included. Mutations of TP53 were detected in 17 FOSCC brushings (48%) and in none of the controls (P < .001). Six genes (ZAP70, FLI1, MiR124-1, KIF1A, MAGEC2 and MiR363) were differentially methylated in FOSCC and were included in a methylation score. An algorithm based on TP53 mutational status and methylation score allowed to differentiate FOSCC from controls with a 69% sensitivity and a 97% specificity (accuracy, 86%). In 19 FOSCC histological samples, TP53 mutational status was fully concordant with brushings and a positive methylation score was observed in all cases. These results are promising for the identification of FOSCC by oral brushing, although some factors may limit the accuracy of this technique and further studies are required to assess its reproducibility in clinical practice.

Pre-Operative Evaluation of DNA Methylation Profile in Oral Squamous Cell Carcinoma Can Predict Tumor Aggressive Potential

BACKGROUND

Prognosis of oral squamous cell carcinoma (OSCC) is difficult to exactly assess on pre-operative biopsies. Since OSCC DNA methylation profile has proved to be a useful pre-operative diagnostic tool, the aim of the present study was to evaluate the prognostic impact of DNA methylation profile to discriminate OSCC with high and low aggressive potential.

METHODS

36 OSCC cases underwent neoplastic cells collection by gentle brushing of the lesion, before performing a pre-operative biopsy. The CpG islands methylation status of 13 gene (ZAP70, ITGA4, KIF1A, PARP15, EPHX3, NTM, LRRTM1, FLI1, MiR193, LINC00599, MiR296, TERT, GP1BB) was studied by bisulfite Next Generation Sequencing (NGS). A Cox proportional hazards model via likelihood-based component-wise boosting was used to evaluate the prognostic power of the CpG sites.

RESULTS

The boosting estimation identified five CpGs with prognostic significance: EPHX3-24, EPHX3-26, ITGA4-3, ITGA4-4, and MiR193-3. The combination of significant CpGs provided promising results for adverse events prediction (Brier score = 0.080, C-index = 0.802 and AUC = 0.850). ITGA4 had a strong prognostic power in patients with early OSCC.

CONCLUSIONS

These data confirm that the study of methylation profile provides new insights into the molecular mechanisms of OSCC and can allow a better OSCC prognostic stratification even before surgery.

Analysis of DNA methylation and TP53 mutational status for differentiating feline oral squamous cell carcinoma from non-neoplastic mucosa: A preliminary study

Feline oral squamous cell carcinoma (FOSCC) is characterized by high local invasiveness and early bone lysis. The late diagnosis largely limits the efficacy of therapy and increases treatment-related morbidity. The aim of this exploratory study was to assess the methylation pattern of 10 candidate genes and TP53 mutational status in histologic samples of FOSCC. Results were compared with normal oral mucosa and oral inflammatory lesions, in order to establish a gene panel for FOSCC detection. For 10 cats, the above analyses were also performed on oral brushing samples, in order to explore the utility of these methods for screening purposes. Thirty-one FOSCC, 25 chronic inflammatory lesions and 12 controls were included. TP53 mutations were significantly more frequent in the FOSCC (68%) than in the non-neoplastic oral mucosa (3%; P <.001). Based on lasso regression analysis, a step-wise algorithm including TP53, FLI1, MiR124-1, KIF1A and MAGEC2 was proposed. The algorithm allowed to differentiate FOSCC with 94% sensitivity and 100% specificity (accuracy, 97%). When applying the proposed algorithm on 10 brushing samples, accuracy decreased to 80%. These results indicate that the altered DNA methylation of specific genes is present in FOSCC, together with a significant proportion of TP53 mutations. Such alterations are infrequent in normal oral mucosa and chronic stomatitis in cats, suggesting their involvement in feline oral carcinogenesis and their utility as diagnostic biomarkers. Further studies on a high number of brushing samples will be needed to assess the utility of a screening test for the early detection of FOSCC.

Analysis of methylation datasets identified significantly changed genes and functional pathways in osteoarthritis

BACKGROUND

Researches indicate that epigenetics was involved in osteoarthritis (OA). The purpose of this study was to describe the alterations of DNA methylation in hip and knee OA by comparing DNA methylome of OA cartilage and non-OA samples and to identify novel genes and pathways associated with OA.

METHODS

We gained two expression profiling datasets (GSE73626 and GSE63695) from the GEO dataset. The RnBeads in R package was used to identify differentially methylated CpG sites. Genes that showed significant differences in DNA methylation between OA and normal control groups underwent functional annotation analysis using the online tool of GeneCodis. Furthermore, we used the Sequenom MassARRAY platform (CapitalBio, Beijing, China) to perform the quantitative methylation analysis.

RESULTS

A total of 249 hypermethylated sites and 96 hypomethylated sites were obtained from OA samples compared with normal control samples. Functional analysis of differentially methylated genes obtained that embryonic skeletal system morphogenesis, cartilage development, and skeletal system development may be involved in the pathogenesis of OA. Eight genes including HOXB3, HOXB4, HOXB6, HOXC4, HOXC10, HOXD3, TBX3, and TBX5 were identified as potential novel biomarkers for OA.

CONCLUSION

Taken together, our study found different molecular characteristics between OA patients and normal controls. This may provide new clues to elucidate the pathogenesis of OA.Key Points• Embryonic skeletal system morphogenesis, cartilage development, skeletal system development may be involved in the pathogenesis of OA.• Eight genes are identified as potential novel markers for OA.• Our future in vivo molecular intervention experiments will extend our current findings.

Comprehensive mapping of the methylation landscape of 16 CpG-dense regions in oral and pharyngeal squamous cell carcinoma

Aim: The goal of this study was to comprehensively interrogate and map DNA methylation across 16 CpG-dense regions previously associated with oral and pharyngeal squamous cell carcinoma (OPSCC). Materials & methods: Targeted multiplex bisulfite amplicon sequencing was performed on four OPSCC cell lines and primary non-neoplastic oral epithelial cells. Real-time quantitative polymerase chain reaction (RT-qPCR) was performed for a subset of associated genes. Results: There was clear differential methylation between one or more OPSCC cell lines and control cells for the majority of CpG-dense regions. Conclusion: Targeted multiplex bisulfite amplicon sequencing allowed us to efficiently map methylation across the entire region of interest with a high degree of sensitivity and helps shed light on novel differentially methylated regions that may have value as biomarkers of OPSCC.

DNA methylation biomarkers for head and neck squamous cell carcinoma

DNA methylation plays an important role in the etiology and pathogenesis of head and neck squamous cell carcinoma (HNSCC). The current study aimed to identify aberrantly methylated-differentially expressed genes (DEGs) by a comprehensive bioinformatics analysis. In addition, we screened for DEGs affected by DNA methylation modification and further investigated their prognostic values for HNSCC. We included microarray data of DNA methylation (GSE25093 and GSE33202) and gene expression (GSE23036 and GSE58911) from Gene Expression Omnibus. Aberrantly methylated-DEGs were analyzed with R software. The Cancer Genome Atlas (TCGA) RNA sequencing and DNA methylation (Illumina HumanMethylation450) databases were utilized for validation. In total, 27 aberrantly methylated genes accompanied by altered expression were identified. After confirmation by The Cancer Genome Atlas (TCGA) database, 2 hypermethylated-low-expression genes (FAM135B and ZNF610) and 2 hypomethylated-high-expression genes (HOXA9 and DCC) were identified. A receiver operating characteristic (ROC) curve confirmed the diagnostic value of these four methylated genes for HNSCC. Multivariate Cox proportional hazards analysis showed that FAM135B methylation was a favorable independent prognostic biomarker for overall survival of HNSCC patients.

New variable selection strategy for analysis of high-dimensional DNA methylation data

In genetic association studies, regularization methods are often used due to their computational efficiency for analysis of high-dimensional genomic data. DNA methylation data generated from Infinium HumanMethylation450 BeadChip Kit have a group structure where an individual gene consists of multiple Cytosine-phosphate-Guanine (CpG) sites. Consequently, group-based regularization can precisely detect outcome-related CpG sites. Representative examples are sparse group lasso (SGL) and network-based regularization. The former is powerful when most of the CpG sites within the same gene are associated with a phenotype outcome. In contrast, the latter is preferred when only a few of the CpG sites within the same gene are related to the outcome. In this paper, we propose new variable selection strategy based on a selection probability that measures selection frequency of individual variables selected by both SGL and network-based regularization. In extensive simulation study, we demonstrated that the proposed strategy can show relatively outstanding selection performance under any situation, compared with both SGL and network-based regularization. Also, we applied the proposed strategy to identify differentially methylated CpG sites and their corresponding genes from ovarian cancer data.

Use of integrative epigenetic and mRNA expression analyses to identify significantly changed genes and functional pathways in osteoarthritic cartilage

Aim

Osteoarthritis (OA) is caused by complex interactions between genetic and environmental factors. Epigenetic mechanisms control the expression of genes and are likely to regulate the OA transcriptome. We performed integrative genomic analyses to define methylation-gene expression relationships in osteoarthritic cartilage.

Patients and Methods

Genome-wide DNA methylation profiling of articular cartilage from five patients with OA of the knee and five healthy controls was conducted using the Illumina Infinium HumanMethylation450 BeadChip (Illumina, San Diego, California). Other independent genome-wide mRNA expression profiles of articular cartilage from three patients with OA and three healthy controls were obtained from the Gene Expression Omnibus (GEO) database. Integrative pathway enrichment analysis of DNA methylation and mRNA expression profiles was performed using integrated analysis of cross-platform microarray and pathway software. Gene ontology (GO) analysis was conducted using the Database for Annotation, Visualization and Integrated Discovery (DAVID).

Results

We identified 1265 differentially methylated genes, of which 145 are associated with significant changes in gene expression, such as DLX5, NCOR2 and AXIN2 (all p-values of both DNA methylation and mRNA expression < 0.05). Pathway enrichment analysis identified 26 OA-associated pathways, such as mitogen-activated protein kinase (MAPK) signalling pathway (p = 6.25 × 10-4), phosphatidylinositol (PI) signalling system (p = 4.38 × 10-3), hypoxia-inducible factor 1 (HIF-1) signalling pathway (p = 8.63 × 10-3 pantothenate and coenzyme A (CoA) biosynthesis (p = 0.017), ErbB signalling pathway (p = 0.024), inositol phosphate (IP) metabolism (p = 0.025), and calcium signalling pathway (p = 0.032).

Conclusion

We identified a group of genes and biological pathwayswhich were significantly different in both DNA methylation and mRNA expression profiles between patients with OA and controls. These results may provide new clues for clarifying the mechanisms involved in the development of OA.

Cite this article: A. He, Y. Ning, Y. Wen, Y. Cai, K. Xu, Y. Cai, J. Han, L. Liu, Y. Du, X. Liang, P. Li, Q. Fan, J. Hao, X. Wang, X. Guo, T. Ma, F. Zhang. Use of integrative epigenetic and mRNA expression analyses to identify significantly changed genes and functional pathways in osteoarthritic cartilage. Bone Joint Res 2018;7:343–350. DOI: 10.1302/2046-3758.75.BJR-2017-0284.R1.

pETM: a penalized Exponential Tilt Model for analysis of correlated high-dimensional DNA methylation data

Motivation

DNA methylation plays an important role in many biological processes and cancer progression. Recent studies have found that there are also differences in methylation variations in different groups other than differences in methylation means. Several methods have been developed that consider both mean and variance signals in order to improve statistical power of detecting differentially methylated loci. Moreover, as methylation levels of neighboring CpG sites are known to be strongly correlated, methods that incorporate correlations have also been developed. We previously developed a network-based penalized logistic regression for correlated methylation data, but only focusing on mean signals. We have also developed a generalized exponential tilt model that captures both mean and variance signals but only examining one CpG site at a time.

Results

In this article, we proposed a penalized Exponential Tilt Model (pETM) using network-based regularization that captures both mean and variance signals in DNA methylation data and takes into account the correlations among nearby CpG sites. By combining the strength of the two models we previously developed, we demonstrated the superior power and better performance of the pETM method through simulations and the applications to the 450K DNA methylation array data of the four breast invasive carcinoma cancer subtypes from The Cancer Genome Atlas (TCGA) project. The developed pETM method identifies many cancer-related methylation loci that were missed by our previously developed method that considers correlations among nearby methylation loci but not variance signals.

Availability and Implementation

The R package 'pETM' is publicly available through CRAN: http://cran.r-project.org .

Contact

sw2206@columbia.edu.

Supplementary information

Supplementary data are available at Bioinformatics online.

Expression Levels and Clinical Significance of miR-21-5p, miR-let-7a, and miR-34c-5p in Laryngeal Squamous Cell Carcinoma

Objective

Altered microRNAs (miRNAs) expression has been found in many cancer types, including laryngeal squamous cell carcinoma (LSCC). The aim of this study was to determine the role and clinical value of three LSCC-related miRs, such as miR-21-5p, miR-let-7a, and miR-34c-5p in a homogeneous cohort of patients with primary LSCC treated by primary surgery.

Methods

Expression levels of miR-21-5p, miR-let-7a, and miR-34c-5p were detected in 43 pairs of LSCC and adjacent normal tissues by reverse-transcription quantitative PCR. Overall survival and disease-free survival were evaluated using the Kaplan–Meier method, and multivariate analysis was performed using the Cox proportional hazard analysis.

Results

miR-21-5p is significantly upregulated, while miR-let-7a is significantly downregulated in LSCC tumor tissues compared with the corresponding adjacent normal tissues. The downregulation of miR-34c-5p expression significantly correlated with a shorter disease-free survival and, in the multivariate analysis, low miR-34c-5p expression was associated with an increased risk of recurrence.

Conclusions

miR-21-5p, miR-let-7a, and miR-34c-5p seem to play a critical role in LSCC carcinogenesis and might have a diagnostic and prognostic clinical value. The miR-let-7a levels could have a predictive role for lymph node metastases and miR-34c-5p might be a promising biomarker of patient outcome.

CpG location and methylation level are crucial factors for the early detection of oral squamous cell carcinoma in brushing samples using bisulfite sequencing of a 13-gene panel

Background

Oral squamous cell carcinoma (OSCC) is usually diagnosed at an advanced stage and is commonly preceded by oral premalignant lesions. The mortality rates have remained unchanged (50% within 5 years after diagnosis), and it is related to tobacco smoking and alcohol intake. Novel molecular markers for early diagnosis are urgently needed. The purpose of this study was to evaluate the diagnostic value of methylation level in a set of 18 genes by bisulfite next-generation sequencing.

Methods

With minimally invasive oral brushing, 28 consecutive OSCC, one squamous cell carcinoma with sarcomatoid features, six high-grade squamous intraepithelial lesions (HGSIL), 30 normal contralateral mucosa from the same patients, and 65 healthy donors were evaluated for DNA methylation analyzing 18 target genes by quantitative bisulfite next-generation sequencing. We further evaluated an independent cohort (validation dataset) made of 20 normal donors, one oral fibroma, 14 oral lichen planus (OLP), three proliferative verrucous leukoplakia (PVL), and two OSCC.

Results

Comparing OSCC with normal healthy donors and contralateral mucosa in 355 CpGs, we identified the following epigenetically altered genes: ZAP70, ITGA4, KIF1A, PARP15, EPHX3, NTM, LRRTM1, FLI1, MIR193, LINC00599, PAX1, and MIR137HG showing hypermethylation and MIR296, TERT, and GP1BB showing hypomethylation. The behavior of ZAP70, GP1BB, H19, EPHX3, and MIR193 fluctuated among different interrogated CpGs. The gap between normal and OSCC samples remained mostly the same (Kruskal-Wallis P values < 0.05), but the absolute values changed conspicuously. ROC curve analysis identified the most informative CpGs, and we correctly stratified OSCC and HGSIL from normal donors using a multiclass linear discriminant analysis in a 13-gene panel (AUC 0.981). Only the OSCC with sarcomatoid features was negative. Three contralateral mucosa were positive, a sign of a possible field cancerization. Among imprinted genes, only MIR296 showed loss of imprinting. DNMT1, TERC, and H19 together with the global methylation of long interspersed element 1 were unchanged. In the validation dataset, values over the threshold were detected in 2/2 OSCC, in 3/3 PVL, and in 2/14 OLP.

Conclusions

Our data highlight the importance of CpG location and correct estimation of DNA methylation level for highly accurate early diagnosis of OSCC.

Electronic supplementary material

The online version of this article (doi:10.1186/s13148-017-0386-7) contains supplementary material, which is available to authorized users.

High-Risk Human Papillomavirus E7 Alters Host DNA Methylome and Represses HLA-E Expression in Human Keratinocytes

Human papillomavirus (HPV) infection distinctly alters methylation patterns in HPV-associated cancer. We have recently reported that HPV E7-dependent promoter hypermethylation leads to downregulation of the chemokine CXCL14 and suppression of antitumor immune responses. To investigate the extent of gene expression dysregulated by HPV E7-induced DNA methylation, we analyzed parallel global gene expression and DNA methylation using normal immortalized keratinocyte lines, NIKS, NIKS-16, NIKS-18, and NIKS-16∆E7. We show that expression of the MHC class I genes is downregulated in HPV-positive keratinocytes in an E7-dependent manner. Methylome analysis revealed hypermethylation at a distal CpG island (CGI) near the HLA-E gene in NIKS-16 cells compared to either NIKS cells or NIKS-16∆E7 cells, which lack E7 expression. The HLA-E CGI functions as an active promoter element which is dramatically repressed by DNA methylation. HLA-E protein expression on cell surface is downregulated by high-risk HPV16 and HPV18 E7 expression, but not by low-risk HPV6 and HPV11 E7 expression. Conversely, demethylation at the HLA-E CGI restores HLA-E protein expression in HPV-positive keratinocytes. Because HLA-E plays an important role in antiviral immunity by regulating natural killer and CD8+ T cells, epigenetic downregulation of HLA-E by high-risk HPV E7 may contribute to virus-induced immune evasion during HPV persistence.

Environmental Enrichment Modified Epigenetic Mechanisms in SAMP8 Mouse Hippocampus by Reducing Oxidative Stress and Inflammaging and Achieving Neuroprotection

With the increase in life expectancy, aging and age-related cognitive impairments are becoming one of the most important issues for human health. At the same time, it has been shown that epigenetic mechanisms are emerging as universally important factors in life expectancy. The Senescence Accelerated Mouse P8 (SAMP8) strain exhibits age-related deterioration evidenced in learning and memory abilities and is a useful model of neurodegenerative disease. In SAMP8, Environmental Enrichment (EE) increased DNA-methylation levels (5-mC) and reduced hydroxymethylation levels (5-hmC), as well as increased histone H3 and H4 acetylation levels. Likewise, we found changes in the hippocampal gene expression of some chromatin-modifying enzyme genes, such as Dnmt3b. Hdac1. Hdac2. Sirt2, and Sirt6. Subsequently, we assessed the effects of EE on neuroprotection-related transcription factors, such as the Nuclear regulatory factor 2 (Nrf2)-Antioxidant Response Element pathway and Nuclear Factor kappa Beta (NF-κB), which play critical roles in inflammation. We found that EE produces an increased expression of antioxidant genes, such as Hmox1. Aox1, and Cox2, and reduced the expression of inflammatory genes such as IL-6 and Cxcl10, all of this within the epigenetic context modified by EE. In conclusion, EE prevents epigenetic changes that promote or drive oxidative stress and inflammaging.

Methods for genome-wide DNA methylation analysis in human cancer

Aberrant DNA methylation is considered to be one of the most common hallmarks of cancer. Several recent advances in assessing the DNA methylome provide great promise for deciphering the cancer-specific DNA methylation patterns. Herein, we present the current key technologies used to detect high-throughput genome-wide DNA methylation, and the available cancer-associated methylation databases. Additionally, we focus on the computational methods for preprocessing, analyzing and interpreting the cancer methylome data. It not only discusses the challenges of the differentially methylated region calling and the prediction model construction but also highlights the biomarker investigation for cancer diagnosis, prognosis and response to treatment. Finally, some emerging challenges in the computational analysis of cancer methylome data are summarized.

Prognostic Value of Drinking Status and Aldehyde Dehydrogenase 2 Polymorphism in Patients With Head and Neck Squamous Cell Carcinoma

Background

The association between alcohol drinking, aldehyde dehydrogenase 2 (ALDH2) polymorphism, and survival in patients with head and neck squamous cell carcinoma (HNSCC) remains unclear.

Methods

We performed a retrospective cohort study of 267 HNSCC patients at Aichi Cancer Center. Of these, 65 patients (24%) were non-drinkers, 104 (39%) were light drinkers (ethanol <46 g or <5 days/week), 46 (17%) were moderate drinkers (ethanol intake 46–68 g/day and ≥5 days/week), and 52 (20%) were heavy drinkers (ethanol intake ≥69 g and ≥5 days/week). The prognostic value of pre-treatment drinking status and ALDH2 polymorphism was investigated using multivariate proportional hazard models.

Results

Drinking status was associated with disease-free survival (DFS) in HNSCC patients, with marginal statistical significance (5-year DFS: 67.9% [95% confidence interval {CI}, 53.8–78.4%] for non-drinkers, 57.6% [95% CI, 47.4–66.6%] for light drinkers, 46.1% [95% CI, 30.8–60.1%] for moderate drinkers, and 43.5% [95% CI, 29.3–56.9%] for heavy drinkers; P = 0.088). However, this association lost significance when multivariate analyses were adjusted for established prognostic factors. ALDH2 genotype was not significantly associated with DFS in HNSCC patients (5-year DFS: 85.7% [95% CI, 53.9–96.2%] for Lys/Lys, 56.2% [95% CI, 47.4–64.1%] for Glu/Lys, and 50.5% [95% CI, 40.3–59.7%] for Glu/Glu; P = 0.154). After stratification by ALDH2 genotype, we observed a significant positive dose-response relationship between drinking status and DFS in HNSCC patients with ALDH2 Glu/Glu (P_trend = 0.029).

Conclusions

In this study, we identified a significant positive dose-response relationship between pre-treatment drinking status and DFS in HNSCC patients with ALDH2 Glu/Glu. To confirm this association, further study is warranted.

New DNA methylation markers and global DNA hypomethylation are associated with oral cancer development

DNA promoter methylation of tumor suppressor genes and global DNA hypomethylation are common features of head and neck cancers. Our goal was to identify early DNA methylation changes in oral premalignant lesions (OPL) that may serve as predictive markers of developing oral squamous cell carcinoma (OSCC). Using high-throughput DNA methylation profiles of 24 OPLs, we found that the top 86 genes differentially methylated between patients who did or did not develop OSCC were simultaneously hypermethylated, suggesting that a CpG island methylation phenotype may occur early during OSCC development. The vast majority of the 86 genes were nonmethylated in normal tissues and hypermethylated in OSCC versus normal mucosa. We used pyrosequencing in a validation cohort of 44 patients to evaluate the degree of methylation of AGTR1, FOXI2, and PENK promoters CpG sites that were included in the top 86 genes and of LINE1 repetitive element methylation, a surrogate of global DNA methylation. A methylation index was developed by averaging the percent methylation of AGTR1, FOXI2, and PENK promoters; patients with a high methylation index had a worse oral cancer-free survival (P = 0.0030). On the other hand, patients with low levels of LINE1 methylation had a significantly worse oral cancer-free survival (P = 0.0153). In conclusion, AGTR1, FOXI2, and PENK promoter methylation and LINE1 hypomethylation may be associated with an increased risk of OSCC development in patients with OPLs.

MicroRNA-34c-5p is related to recurrence in laryngeal squamous cell carcinoma

OBJECTIVES/HYPOTHESIS

Altered microRNA expression has been found in many cancer types, including laryngeal squamous cell carcinoma (LSCC). We investigated the association of LSCC-related miR-34c-5p with disease-free survival and overall survival.

STUDY DESIGN

Retrospective cohort study.

METHODS

Expression levels of miR-34c-5p were detected in 90 LSCC formalin-fixed paraffin-embedded tissues by reverse-transcription quantitative polymerase chain reaction. Overall survival and disease-free survival were evaluated using the Kaplan-Meier method, and multivariate analysis was performed using Cox proportional hazard analysis.

RESULTS

A downregulation of miR-34c-5p expression significantly correlated with worse disease-free and overall survival. In the multivariate analysis, low miR-34c-5p expression was associated with an increased risk of recurrence.

CONCLUSIONS

A downregulation of miR-34c-5p in LSCC is independently associated with unfavorable disease-free survival, suggesting that miR-34c-5p might be a promising marker for evaluating the risk of recurrences.

LEVEL OF EVIDENCE

NA.

DNA methylation as a promising landscape: A simple blood test for breast cancer prediction

Breast cancer is the most common malignancy among women worldwide. Risk assessment is one of the main services delivered by cancer clinics. Biomarker analysis on different tissues including the peripheral blood can provide crucial information. One of the potential epigenetic biomarkers (epimarkers) is introduced as the peripheral blood DNA methylation pattern. This study was conducted to evaluate the potential value of peripheral blood epimarkers as an accessible tool to predict the risk of breast cancer development. WBC's DNA was the focus of several case-control studies at both genome wide and candidate gene levels to reveal epigenetic changes accounting for predisposition to breast cancer, leading to suggest that ATM, TITF1, SFRP1, NUP155, NEUROD1, ZNF217, DBC2, DOK7 and ESR1 genes and the LINE1, Alu and Sat2 DNA elements could be considered as the potential epimarkers. To address that by which mechanisms WBC's DNA methylation patterns could be linked to the propensity to breast cancer, several contemplations have been offered. Constitutional epimutation during embryonic life, and methylation changes secondary to either environmental exposures or tumor-mediated immune response, are the two main mechanisms. One can deduce that epimarkers based on their potential properties or regulatory impacts on cancer-related genes may be employed for risk prediction, prognosis, and survival inferences that are highly required for breast cancer management toward personalized medicine.

Identifying Etiologically Distinct Sub-Types of Cancer: A Demonstration Project Involving Breast Cancer

With the advent of increasingly detailed molecular portraits of tumor specimens, much attention has been directed toward identifying clinically distinct subtypes of cancer. Subtyping of tumors can also be accomplished with the goal of identifying distinct etiologies. We demonstrate the use of new methodologies to identify genes that distinguish etiologically heterogeneous subtypes of breast cancer using data from the case-control Cancer and Steroid Hormone Study. Tumor specimens were evaluated using a breast cancer expression panel of 196 genes. Using a statistical measure that distinguishes the degree of etiologic heterogeneity in tumor subtypes, each gene is ranked on the basis of its ability to distinguish etiologically distinct subtypes. This is accomplished independently using case-control comparisons and by examining the concordance odds ratios in double primaries. The estrogen receptor gene, and others in this pathway with expression levels that correlated strongly with estrogen receptor levels, demonstrate high degrees of etiologic heterogeneity in both methods. Our results are consistent with a growing literature that confirms the distinct etiologies of breast cancers classified on the basis of estrogen receptor expression levels. This proof-of-principle project demonstrates the viability of new strategies to identify genomic features that distinguish subtypes of cancer from an etiologic perspective.

Genome methylation patterns in male breast cancer - Identification of an epitype with hypermethylation of polycomb target genes

Male breast cancer (MBC) is a rare disease that shares both similarities and differences with female breast cancer (FBC). The aim of this study was to assess genome-wide DNA methylation profiles in MBC and compare them with the previously identified transcriptional subgroups of MBC, luminal M1 and M2, as well as the intrinsic subtypes of FBC. Illumina's 450K Infinium arrays were applied to 47 MBC and 188 FBC tumors. Unsupervised clustering of the most variable CpGs among MBC tumors revealed two stable epitypes, designated ME1 and ME2. The methylation patterns differed significantly between the groups and were closely associated with the transcriptional subgroups luminal M1 and M2. Tumors in the ME1 group were more proliferative and aggressive than ME2 tumors, and showed a tendency toward inferior survival. ME1 tumors also displayed hypermethylation of PRC2 target genes and high expression of EZH2, one of the core components of PRC2. Upon combined analysis of MBC and FBC tumors, ME1 MBCs clustered among luminal B FBC tumors and ME2 MBCs clustered within the predominantly luminal A FBC cluster. The majority of the MBC tumors remained grouped together within the clusters rather than being interspersed among the FBC tumors. Differences in the genomic location of methylated CpGs, as well as in the regulation of central canonical pathways may explain the separation between MBC and FBC tumors in the respective clusters. These findings further suggest that MBC is not readily defined using conventional criteria applied to FBC.

DNA methylation profiles in primary cutaneous melanomas are associated with clinically significant pathologic features

DNA methylation studies have elucidated a methylation signature distinguishing primary melanomas from benign nevi and provided new insights about genes that may be important in melanoma development. However, it is unclear whether methylation differences among primary melanomas are related to tumor pathologic features with known clinical significance. We utilized the Illumina GoldenGate Cancer Panel array to investigate the methylation profiles of 47 primary cutaneous melanomas. Arraywide methylation patterns revealed a positive association of methylation with Breslow thickness and mutated BRAF, a negative association with mitotic rate, and a weak association with ulceration. Hierarchical clustering on CpG sites exhibiting the most variable methylation (n = 235) divided the melanoma samples into three clusters, including a highly methylated cluster that was positively associated with Breslow thickness and an intermediately methylated cluster associated with Breslow thickness and mitotic rate. Our findings provide support for the existence of methylation-defined subsets in melanomas with increased methylation associated with Breslow thickness.

Covariate adjusted differential variability analysis of DNA methylation with propensity score method

It has been proposed recently that differentially variable CpG methylation (DVC) may contribute to transcriptional aberrations in human diseases. In large scale epigenetic studies, potential confounders could affect the observed methylation variabilities and need to be accounted for. In this paper, we develop a robust statistical model for differential variability DVC analysis that accounts for potential confounding covariates by utilizing the propensity score method. Our method is based on a weighted score test on strata generated propensity score stratification. To the best of our knowledge, this is the first proposed statistical method for detecting DVCs that adjusts for confounding covariates. We show that this method is robust against model misspecification and achieves good operating characteristics based on extensive simulations and a case study.

DNA methylation profiling in the Carolina Breast Cancer Study defines cancer subclasses differing in clinicopathologic characteristics and survival

INTRODUCTION

Breast cancer is a heterogeneous disease, with several intrinsic subtypes differing by hormone receptor (HR) status, molecular profiles, and prognosis. However, the role of DNA methylation in breast cancer development and progression and its relationship with the intrinsic tumor subtypes are not fully understood.

METHODS

A microarray targeting promoters of cancer-related genes was used to evaluate DNA methylation at 935 CpG sites in 517 breast tumors from the Carolina Breast Cancer Study, a population-based study of invasive breast cancer.

RESULTS

Consensus clustering using methylation (β) values for the 167 most variant CpG loci defined four clusters differing most distinctly in HR status, intrinsic subtype (luminal versus basal-like), and p53 mutation status. Supervised analyses for HR status, subtype, and p53 status identified 266 differentially methylated CpG loci with considerable overlap. Genes relatively hypermethylated in HR+, luminal A, or p53 wild-type breast cancers included FABP3, FGF2, FZD9, GAS7, HDAC9, HOXA11, MME, PAX6, POMC, PTGS2, RASSF1, RBP1, and SCGB3A1, whereas those more highly methylated in HR-, basal-like, or p53 mutant tumors included BCR, C4B, DAB2IP, MEST, RARA, SEPT5, TFF1, THY1, and SERPINA5. Clustering also defined a hypermethylated luminal-enriched tumor cluster 3 that gene ontology analysis revealed to be enriched for homeobox and other developmental genes (ASCL2, DLK1, EYA4, GAS7, HOXA5, HOXA9, HOXB13, IHH, IPF1, ISL1, PAX6, TBX1, SOX1, and SOX17). Although basal-enriched cluster 2 showed worse short-term survival, the luminal-enriched cluster 3 showed worse long-term survival but was not independently prognostic in multivariate Cox proportional hazard analysis, likely due to the mostly early stage cases in this dataset.

CONCLUSIONS

This study demonstrates that epigenetic patterns are strongly associated with HR status, subtype, and p53 mutation status and may show heterogeneity within tumor subclass. Among HR+ breast tumors, a subset exhibiting a gene signature characterized by hypermethylation of developmental genes and poorer clinicopathologic features may have prognostic value and requires further study. Genes differentially methylated between clinically important tumor subsets have roles in differentiation, development, and tumor growth and may be critical to establishing and maintaining tumor phenotypes and clinical outcomes.

Environmental enrichment modulates 5-hydroxymethylcytosine dynamics in hippocampus

Gene-environment interactions mediated at the epigenetic level may provide an initial step in delivering an appropriate response to environmental changes. 5-Hydroxymethylcytosine (5hmC), a DNA base derived from 5-methylcytosine (5mC), accounts for ~40% of modified cytosine in the brain and has been implicated in DNA methylation-related plasticity. To identify the role of 5hmC in gene-environment interactions, we exposed both young (6-week-old) and aged (18-month-old) mice to both an enriched environment and a standard environment. Exposure to EE significantly improves learning and memory in aged mice and reduces 5hmC abundance in mouse hippocampus. Furthermore, we mapped the genome-wide distribution of 5hmC and found that the alteration of 5hmC modification occurred mainly at gene bodies. In particular, genes involved in axon guidance are enriched among the genes with altered 5hmC modification. These results together suggest that environmental enrichment could modulate the dynamics of 5hmC in hippocampus, which could potentially contribute to improved learning and memory in aged animals.

Epigenetics of human myometrium: DNA methylation of genes encoding contraction-associated proteins in term and preterm labor

Preterm birth involves the interaction of societal and environmental factors potentially modulating the length of gestation via the epigenome. An established form of epigenetic regulation is DNA methylation where promoter hypermethylation is associated with gene repression. We hypothesized we would find differences in DNA methylation in the myometrium of women with preterm labor of different phenotypes versus normal term labor. Myometrial tissue was obtained at cesarean section at term with or without labor, preterm without labor, idiopathic preterm labor, and twin gestations with labor. Genomic DNA was isolated, and samples in each group were combined and analyzed on a NimbleGen 2.1M human DNA methylation array. Differences in methylation from -8 to +3 kb of transcription start sites of 22 contraction-associated genes were determined. Cytosine methylation was not present in CpG islands of any gene but was present outside of CpG islands in shores and shelves in 19 genes. No differential methylation was found across the tissue groups for six genes (PTGES3L, PTGER2, PTGER4, PTGFRN, ESR2, and GJA1). For 13 genes, differential methylation occurred in several patterns between tissue groups. We find a correlation between hypomethylation and increased mRNA expression of PTGES/mPGES-1, indicating potential functional relevance of methylation, but no such correlation for PTGS2/COX-2, suggesting other regulatory mechanisms for PTGS2 at labor. The majority of differential DNA methylation of myometrial contraction-associated genes with different labor phenotypes occurs outside of CpG islands in gene promoters, suggesting that the entirety of DNA methylation across the genome should be considered.

DNA methylation subgroups and the CpG island methylator phenotype in gastric cancer: a comprehensive profiling approach

BACKGROUND

Methylation-induced silencing of promoter CpG islands in tumor suppressor genes plays an important role in human carcinogenesis. In colorectal cancer, the CpG island methylator phenotype (CIMP) is defined as widespread and elevated levels of DNA methylation and CIMP+ tumors have distinctive clinicopathological and molecular features. In contrast, the existence of a comparable CIMP subtype in gastric cancer (GC) has not been clearly established. To further investigate this issue, in the present study we performed comprehensive DNA methylation profiling of a well-characterised series of primary GC.

METHODS

The methylation status of 1,421 autosomal CpG sites located within 768 cancer-related genes was investigated using the Illumina GoldenGate Methylation Panel I assay on DNA extracted from 60 gastric tumors and matched tumor-adjacent gastric tissue pairs. Methylation data was analysed using a recursively partitioned mixture model and investigated for associations with clinicopathological and molecular features including age, Helicobacter pylori status, tumor site, patient survival, microsatellite instability and BRAF and KRAS mutations.

RESULTS

A total of 147 genes were differentially methylated between tumor and matched tumor-adjacent gastric tissue, with HOXA5 and hedgehog signalling being the top-ranked gene and signalling pathway, respectively. Unsupervised clustering of methylation data revealed the existence of 6 subgroups under two main clusters, referred to as L (low methylation; 28% of cases) and H (high methylation; 72%). Female patients were over-represented in the H tumor group compared to L group (36% vs 6%; P = 0.024), however no other significant differences in clinicopathological or molecular features were apparent. CpG sites that were hypermethylated in group H were more frequently located in CpG islands and marked for polycomb occupancy.

CONCLUSIONS

High-throughput methylation analysis implicates genes involved in embryonic development and hedgehog signaling in gastric tumorigenesis. GC is comprised of two major methylation subtypes, with the highly methylated group showing some features consistent with a CpG island methylator phenotype.

Distinct patterns of DNA methylation in conventional adenomas involving the right and left colon

Recent studies have shown two distinct non-CIMP methylation clusters in colorectal cancer, raising the possibility that DNA methylation, involving non-CIMP genes, may play a role in the conventional adenoma-carcinoma pathway. A total of 135 adenomas (65 left colon and 70 right colon) were profiled for epigenome-wide DNA methylation using the Illumina HumanMethylation450 BeadChip. A principal components analysis was performed to examine the association between variability in DNA methylation and adenoma location. Linear regression and linear mixed effects models were used to identify locus-specific differential DNA methylation in adenomas of right and left colon. A significant association was present between the first principal component and adenoma location (P=0.007), even after adjustment for subject age and gender (P=0.009). A total of 168 CpG sites were differentially methylated between right- and left-colon adenomas and these loci demonstrated enrichment of homeobox genes (P=3.0 × 10(-12)). None of the 168 probes were associated with CIMP genes. Among CpG loci with the largest difference in methylation between right- and left-colon adenomas, probes associated with PRAC (prostate cancer susceptibility candidate) gene showed hypermethylation in right-colon adenomas whereas those associated with CDX2 (caudal type homeobox transcription factor 2) showed hypermethylation in left-colon adenomas. A subgroup of left-colon adenomas enriched for current smokers (OR=6.1, P=0.004) exhibited a methylation profile similar to right-colon adenomas. In summary, our results indicate distinct patterns of DNA methylation, independent of CIMP genes, in adenomas of the right and left colon.

A conceptual and methodological framework for investigating etiologic heterogeneity

Cancer has traditionally been studied using the disease site of origin as the organizing framework. However, recent advances in molecular genetics have begun to challenge this taxonomy, as detailed molecular profiling of tumors has led to discoveries of subsets of tumors that have profiles that possess distinct clinical and biological characteristics. This is increasingly leading to research that seeks to investigate whether these subtypes of tumors have distinct etiologies. However, research in this field has been opportunistic and anecdotal, typically involving the comparison of distributions of individual risk factors between tumors classified on the basis of candidate tumor characteristics. The purpose of this article is to place this area of investigation within a more general conceptual and analytic framework, with a view to providing more efficient and practical strategies for designing and analyzing epidemiologic studies to investigate etiologic heterogeneity. We propose a formal definition of etiologic heterogeneity and show how classifications of tumor subtypes with larger etiologic heterogeneities inevitably possess greater disease risk predictability overall. We outline analytic strategies for estimating the degree of etiologic heterogeneity among a set of subtypes and for choosing subtypes that optimize the heterogeneity, and we discuss technical challenges that require further methodologic research. We illustrate the ideas by using a pooled case-control study of breast cancer classified by expression patterns of genes known to define distinct tumor subtypes.

DNA methylation profiling identifies EYA4 gene as a prognostic molecular marker in hepatocellular carcinoma

BACKGROUND

DNA hypermethylation plays important roles in carcinogenesis by silencing key genes. This study aims to identify pivotal genes in hepatocellular carcinoma (HCC) by DNA methylation microarray and to assess their prognostic values.

MATERIALS AND METHODS

DNA methylation microarray was performed in 45 pairs of HCC and adjacent nontumorous tissues and six normal liver tissues to identify hypermethylated genes in HCC. Potential prognosis-related genes were selected among hypermethylated genes by analyzing influences of methylation levels on disease-free survival (DFS) and overall survival (OS) in 45 patients. Their prognostic values were validated in 154 patients with HCC (including the initial 45 patients) to determine the independent prognostic gene.

RESULTS

Altogether, 54 CpG islands in 44 genes were hypermethylated in HCC compared with liver tissues. Among them, methylation levels of ERG and HOXA11 were inversely associated with DFS (both P < 0.050), and methylation levels of EYA4 were inversely related to DFS and OS (both P < 0.050). EYA4 expression was inversely related to tumor size (P < 0.050). Lower EYA4 expression and larger tumor size were independent predictors of both shorter DFS and OS, and higher Barcelona Clinic Liver Cancer (BCLC) staging was an independent predictor of shorter OS (all P < 0.050).

CONCLUSIONS

EYA4 functions as a prognostic molecular marker in HCC. Its aberrant hypermethylation and subsequent down-regulation may promote tumor progression.

Genome stability pathways in head and neck cancers

Genomic instability underlies the transformation of host cells toward malignancy, promotes development of invasion and metastasis and shapes the response of established cancer to treatment. In this review, we discuss recent advances in our understanding of genomic stability in squamous cell carcinoma of the head and neck (HNSCC), with an emphasis on DNA repair pathways. HNSCC is characterized by distinct profiles in genome stability between similarly staged cancers that are reflected in risk, treatment response and outcomes. Defective DNA repair generates chromosomal derangement that can cause subsequent alterations in gene expression, and is a hallmark of progression toward carcinoma. Variable functionality of an increasing spectrum of repair gene polymorphisms is associated with increased cancer risk, while aetiological factors such as human papillomavirus, tobacco and alcohol induce significantly different behaviour in induced malignancy, underpinned by differences in genomic stability. Targeted inhibition of signalling receptors has proven to be a clinically-validated therapy, and protein expression of other DNA repair and signalling molecules associated with cancer behaviour could potentially provide a more refined clinical model for prognosis and treatment prediction. Development and expansion of current genomic stability models is furthering our understanding of HNSCC pathophysiology and uncovering new, promising treatment strategies.

Mechanisms of oral carcinogenesis induced by dibenzo[a,l]pyrene: an environmental pollutant and a tobacco smoke constituent

We previously reported that dibenzo[a,l]pyrene (DB[a,l]P), the most potent known environmental carcinogen among polycyclic aromatic hydrocarbons (PAH) congeners, is carcinogenic in the oral tissues of mice. We have now developed a new mouse model which employs the oral application of the fjord region diol epoxide, (±)-anti-11,12-dihydroxy-13,14-epoxy-11,12,13,14-tetrahydrodibenzo[a,l]pyrene (DB[a,l]PDE), a metabolite of the tobacco smoke constituent DB[a,l]P, and we show its specific induction of oral squamous cell carcinoma (OSCC) in both tongue and other oral tissues. Groups of B6C3F1 mice (20/group) received 6 or 3 nmol of (±)-anti-DB[a,l]PDE administered into the oral cavity; 3 times per week for 38 weeks. Additional groups received the vehicle alone or were left untreated. Mice were sacrificed 42 weeks after the first carcinogen administration. The high dose induced 74 and 100% OSCC in the tongue and other oral tissues, respectively; the corresponding values at the lower dose were 45 and 89%. Using immunohistochemistry, we showed that DB[a,l]PDE resulted in overexpression of p53 and COX-2 proteins in malignant tissues when compared to normal oral tissues and tongues. Consistent with the carcinogenicity, we demonstrated powerful mutagenicity in cII gene in B6C3F1 (Big Blue) mouse tongue. The mutational profile in lacI reporter gene is similar to those detected in human head and neck cancer, and p53 mutations were observed in mouse oral tumor tissues. Taken together, we conclude that the formation of diol epoxides plays a major role among the mechanisms by which DB[a,l]P exerts its oral mutagenicity and tumorigenicity.

Signaling Networks of Activated Oncogenic and Altered Tumor Suppressor Genes in Head and Neck Cancer

Head and neck squamous cell carcinoma (HNSCC) arises from the upper aerodigestive tract and is the six most common cancers worldwide. HNSCC is associated with high morbidity and mortality, as standard surgery, radiation, and chemotherapy can cause significant disfigurement and only provide 5-year survival rates of ~50-60%. The heterogeneity of HNSCC subsets with different potentials for recurrence and metastasis challenges the traditional pathological classification system, thereby increasing demand for the development of new diagnostic, prognostic, and therapeutic tools based on global molecular signatures of HNSCC. Historically, using classical biological techniques, it has been extremely difficult and time-consuming to survey hundreds or thousands of genes in a given disease. However, the development of high throughput technologies and high-powered computation throughout the last two decades has enabled us to investigate hundreds or thousands of genes simultaneously. Using high throughput technologies, our laboratory has identified the gene signatures and protein networks, which significantly affect HNSCC malignant phenotypes, including TP53/p63/p73 family members, IL-1/TNF-β/NF-κB, PI3K/AKT/mTOR, IL-6/IL-6R/JAK/STAT3, EGFR/MAPK/AP1, HGF/cMET/EGR1, and TGFβ/TGFβR/TAK1/SMAD pathways. This review summarizes the results from high-throughput technological assays conducted on HNSCC samples, including microarray, DNA methylation, miRNA profiling, and protein array, using primarily experimental data and conclusions generated in our own laboratory. The use of bioinformatics and integrated analyses of data sets from different platforms, as well as meta-analysis of large datasets pulled from multiple publicly available studies, provided significantly higher statistical power to extract biologically relevant information. The data suggested that the heterogeneity of HNSCC genotype and phenotype are much more complex than we previously thought. Understanding of global molecular signatures and disease classification for specific subsets of HNSCC will be essential to provide accurate diagnoses for targeted therapy and personalized treatment, which is an important effort toward improving patient outcomes.

Differential DNA methylation in umbilical cord blood of infants exposed to low levels of arsenic in utero

BACKGROUND

There is increasing epidemiologic evidence that arsenic exposure in utero, even at low levels found throughout much of the world, is associated with adverse reproductive outcomes and may contribute to long-term health effects. Animal models, in vitro studies, and human cancer data suggest that arsenic may induce epigenetic alterations, specifically by altering patterns of DNA methylation.

OBJECTIVES

In this study we aimed to identify differences in DNA methylation in cord blood samples of infants with in utero, low-level arsenic exposure.

METHODS

DNA methylation of cord-blood derived DNA from 134 infants involved in a prospective birth cohort in New Hampshire was profiled using the Illumina Infinium Methylation450K array. In utero arsenic exposure was estimated using maternal urine samples collected at 24-28 weeks gestation. We used a novel cell mixture deconvolution methodology for examining the association between inferred white blood cell mixtures in infant cord blood and in utero arsenic exposure; we also examined the association between methylation at individual CpG loci and arsenic exposure levels.

RESULTS

We found an association between urinary inorganic arsenic concentration and the estimated proportion of CD8+ T lymphocytes (1.18; 95% CI: 0.12, 2.23). Among the top 100 CpG loci with the lowest p-values based on their association with urinary arsenic levels, there was a statistically significant enrichment of these loci in CpG islands (p = 0.009). Of those in CpG islands (n = 44), most (75%) exhibited higher methylation levels in the highest exposed group compared with the lowest exposed group. Also, several CpG loci exhibited a linear dose-dependent relationship between methylation and arsenic exposure.

CONCLUSIONS

Our findings suggest that in utero exposure to low levels of arsenic may affect the epigenome. Long-term follow-up is planned to determine whether the observed changes are associated with health outcomes.

Associations between methylation of paternally expressed gene 3 (PEG3), cervical intraepithelial neoplasia and invasive cervical cancer

Cytology-based screening for invasive cervical cancer (ICC) lacks sensitivity and specificity to discriminate between cervical intraepithelial neoplasia (CIN) likely to persist or progress from cases likely to resolve. Genome-wide approaches have been used to identify DNA methylation marks associated with CIN persistence or progression. However, associations between DNA methylation marks and CIN or ICC remain weak and inconsistent. Between 2008-2009, we conducted a hospital-based, case-control study among 213 Tanzania women with CIN 1/2/3 or ICC. We collected questionnaire data, biopsies, peripheral blood, cervical scrapes, Human papillomavirus (HPV) and HIV-1 infection status. We assessed PEG3 methylation status by bisulfite pyrosequencing. Multinomial logistic regression was used to estimate odds ratios (OR) and confidence intervals (CI 95%) for associations between PEG3 methylation status and CIN or ICC. After adjusting for age, gravidity, hormonal contraceptive use and HPV infection, a 5% increase in PEG3 DNA methylation was associated with increased risk for ICC (OR = 1.6; 95% CI 1.2-2.1). HPV infection was associated with a higher risk of CIN1-3 (OR = 15.7; 95% CI 5.7-48.6) and ICC (OR = 29.5, 95% CI 6.3-38.4). Infection with high risk HPV was correlated with mean PEG3 differentially methylated regions (DMRs) methylation (r = 0.34 p<0.0001), while the correlation with low risk HPV infection was weaker (r = 0.16 p = 0.047). Although small sample size limits inference, these data support that PEG3 methylation status has potential as a molecular target for inclusion in CIN screening to improve prediction of progression. Impact statement: We present the first evidence that aberrant methylation of the PEG3 DMR is an important co-factor in the development of Invasive cervical carcinoma (ICC), especially among women infected with high risk HPV. Our results show that a five percent increase in DNA methylation of PEG3 is associated with a 1.6-fold increase ICC risk. Suggesting PEG3 methylation status may be useful as a molecular marker for CIN screening to improve prediction of cases likely to progress.

Comprehensive analysis of DNA methylation in head and neck squamous cell carcinoma indicates differences by survival and clinicopathologic characteristics

Head and neck squamous cell carcinoma (HNSCC) is the eighth most commonly diagnosed cancer in the United States. The risk of developing HNSCC increases with exposure to tobacco, alcohol and infection with human papilloma virus (HPV). HPV-associated HNSCCs have a distinct risk profile and improved prognosis compared to cancers associated with tobacco and alcohol exposure. Epigenetic changes are an important mechanism in carcinogenic progression, but how these changes differ between viral- and chemical-induced cancers remains unknown. CpG methylation at 1505 CpG sites across 807 genes in 68 well-annotated HNSCC tumor samples from the University of Michigan Head and Neck SPORE patient population were quantified using the Illumina Goldengate Methylation Cancer Panel. Unsupervised hierarchical clustering based on methylation identified 6 distinct tumor clusters, which significantly differed by age, HPV status, and three year survival. Weighted linear modeling was used to identify differentially methylated genes based on epidemiological characteristics. Consistent with previous in vitro findings by our group, methylation of sites in the CCNA1 promoter was found to be higher in HPV(+) tumors, which was validated in an additional sample set of 128 tumors. After adjusting for cancer site, stage, age, gender, alcohol consumption, and smoking status, HPV status was found to be a significant predictor for DNA methylation at an additional 11 genes, including CASP8 and SYBL1. These findings provide insight into the epigenetic regulation of viral vs. chemical carcinogenesis and could provide novel targets for development of individualized therapeutic and prevention regimens based on environmental exposures.

Alcohol, DNA methylation, and cancer

Cancer is one of the most significant diseases associated with chronic alcohol consumption, and chronic drinking is a strong risk factor for cancer, particularly of the upper aerodigestive tract, liver, colorectum, and breast. Several factors contribute to alcohol-induced cancer development (i.e., carcinogenesis), including the actions of acetaldehyde, the first and primary metabolite of ethanol, and oxidative stress. However, increasing evidence suggests that aberrant patterns of DNA methylation, an important epigenetic mechanism of transcriptional control, also could be part of the pathogenetic mechanisms that lead to alcohol-induced cancer development. The effects of alcohol on global and local DNA methylation patterns likely are mediated by its ability to interfere with the availability of the principal biological methyl donor, S-adenosylmethionine (SAMe), as well as pathways related to it. Several mechanisms may mediate the effects of alcohol on DNA methylation, including reduced folate levels and inhibition of key enzymes in one-carbon metabolism that ultimately lead to lower SAMe levels, as well as inhibition of activity and expression of enzymes involved in DNA methylation (i.e., DNA methyltransferases). Finally, variations (i.e., polymorphisms) of several genes involved in one-carbon metabolism also modulate the risk of alcohol-associated carcinogenesis.

Peripheral blood immune cell methylation profiles are associated with nonhematopoietic cancers

BACKGROUND

Blood leukocytes from patients with solid tumors exhibit complex and distinct cancer-associated patterns of DNA methylation. However, the biologic mechanisms underlying these patterns remain poorly understood. Because epigenetic biomarkers offer significant clinical potential for cancer detection, we sought to address a mechanistic gap in recently published works, hypothesizing that blood-based epigenetic variation may be due to shifts in leukocyte populations.

METHODS

We identified differentially methylated regions (DMR) among leukocyte subtypes using epigenome-wide DNA methylation profiling of purified peripheral blood leukocyte subtypes from healthy donors. These leukocyte-tagging DMRs were then evaluated using epigenome-wide blood methylation data from three independent case-control studies of different cancers.

RESULTS

A substantial proportion of the top 50 leukocyte DMRs were significantly differentially methylated among head and neck squamous cell carcinoma (HNSCC) cases and ovarian cancer cases compared with cancer-free controls (48 and 47 of 50, respectively). Methylation classes derived from leukocyte DMRs were significantly associated cancer case status (P < 0.001, P < 0.03, and P < 0.001) for all three cancer types: HNSCC, bladder cancer, and ovarian cancer, respectively and predicted cancer status with a high degree of accuracy (area under the curve [AUC] = 0.82, 0.83, and 0.67).

CONCLUSIONS

These results suggest that shifts in leukocyte subpopulations may account for a considerable proportion of variability in peripheral blood DNA methylation patterns of solid tumors.

IMPACT

This illustrates the potential use of DNA methylation profiles for identifying shifts in leukocyte populations representative of disease, and that such profiles may represent powerful new diagnostic tools, applicable to a range of solid tumors.

Identification of an epigenetic profile classifier that is associated with survival in head and neck cancer

Panels of prognostic biomarkers selected using candidate approaches often do not validate in independent populations, so additional strategies are needed to identify reliable classifiers. In this study, we used an array-based approach to measure DNA methylation and applied a novel method for grouping CpG dinucleotides according to well-characterized genomic sequence features. A hypermethylation profile among 13 CpG loci, characterized by polycomb group target genes, mammalian interspersed repeats, and transcription factor-binding sites (PcG/MIR/TFBS), was associated with reduced survival (HR, 3.98; P = 0.001) in patients with head and neck squamous cell carcinoma. This association was driven by CpGs associated with the TAP1 and ALDH3A1 genes, findings that were validated in an independent patient group (HR, 2.86; P = 0.04). Together, the data not only elucidate new potential targets for therapeutic intervention in head and neck cancer but also may aid in the identification of poor prognosis patients who may require more aggressive treatment regimens.

New Concepts for Translational Head and Neck Oncology: Lessons from HPV-Related Oropharyngeal Squamous Cell Carcinomas

Human papillomavirus (HPV) infection is well established as an etiological agent responsible for a number of pathologies affecting the stratified epithelia of skin and anogenital sites. More recently, the infection by (mucosal) high-risk HPV types has also been found to be causally associated with squamous cell carcinoma in the head and neck region (HNSCC), especially in the oropharynx. Intriguingly, HPV-related oropharyngeal squamous cell carcinomas (OPSCC) represent a distinct clinical entity compared to HPV-negative tumors with particular regard to treatment–response and survival outcome. The association between HPV infection and OPSCC may therefore have important implications for the prevention and/or treatment of OPSCC. The improved survival of patients with HPV-related tumors also raises the question, as to whether a better understanding of the underlying differences may help to identify new therapeutic concepts that could be used in targeted therapy for HPV-negative and improved therapy for HPV-positive cancers. This review summarizes the most recent advances in our understanding of the molecular principles of HPV-related OPSCC, mainly based on functional genomic approaches, but also emphasizes the significant role played by the tumor microenvironment, especially the immune system, for improved clinical outcome and differential sensitivity of HPV-related tumors to current treatment options.