An epigenetic signature of adhesion molecules predicts poor prognosis of ovarian cancer patients

Ping-Ying Chang, Yu-Ping Liao, Hui-Chen Wang, Yu-Chih Chen, Rui-Lan Huang, Yu-Chi Wang, Chiou-Chung Yuan, Hung-Cheng Lai

Research output: Contribution to journalArticle

Abstract

DNA methylation is a promising biomarker for cancer. The epigenetic effects of cell adhesion molecules may affect the therapeutic outcome and the present study examined their effects on survival in ovarian cancer. We integrated methylomics and genomics datasets in The Cancer Genome Atlas (n = 391) and identified 106 highly methylated adhesion-related genes in ovarian cancer tissues. Univariate analysis revealed the methylation status of eight genes related to progression-free survival. In multivariate Cox regression analysis, four highly methylated genes (CD97, CTNNA1, DLC1, HAPLN2) and three genes (LAMA4, LPP, MFAP4) with low methylation were significantly associated with poor progression-free survival. Low methylation of VTN was an independent poor prognostic factor for overall survival after adjustment for age and stage. Patients who carried any two of CTNNA1, DLC1 or MFAP4 were significantly associated with poor progression-free survival (hazard ratio: 1.59; 95% confidence interval: 1.23, 2.05). This prognostic methylation signature was validated in a methylomics dataset generated in our lab (n = 37, hazard ratio: 16.64; 95% confidence interval: 2.68, 103.14) and in another from the Australian Ovarian Cancer Study (n = 91, hazard ratio: 2.43; 95% confidence interval: 1.11, 5.36). Epigenetics of cell adhesion molecules is related to ovarian cancer prognosis. A more comprehensive methylomics of cell adhesion molecules is needed and may advance personalized treatment with adhesion molecule-related drugs.

Original languageEnglish
Pages (from-to)53432-53449
Number of pages18
JournalOncotarget
Volume8
Issue number32
DOIs
Publication statusPublished - Aug 8 2017

Fingerprint

Epigenomics
Ovarian Neoplasms
Methylation
Cell Adhesion Molecules
Disease-Free Survival
Confidence Intervals
Genes
Survival
Atlases
DNA Methylation
Tumor Biomarkers
Genomics
Regression Analysis
Outcome Assessment (Health Care)
Genome
Therapeutics
Pharmaceutical Preparations
Neoplasms
Datasets

Keywords

  • Journal Article

Cite this

An epigenetic signature of adhesion molecules predicts poor prognosis of ovarian cancer patients. / Chang, Ping-Ying; Liao, Yu-Ping; Wang, Hui-Chen; Chen, Yu-Chih; Huang, Rui-Lan; Wang, Yu-Chi; Yuan, Chiou-Chung; Lai, Hung-Cheng.

In: Oncotarget, Vol. 8, No. 32, 08.08.2017, p. 53432-53449.

Research output: Contribution to journalArticle

Chang, Ping-Ying ; Liao, Yu-Ping ; Wang, Hui-Chen ; Chen, Yu-Chih ; Huang, Rui-Lan ; Wang, Yu-Chi ; Yuan, Chiou-Chung ; Lai, Hung-Cheng. / An epigenetic signature of adhesion molecules predicts poor prognosis of ovarian cancer patients. In: Oncotarget. 2017 ; Vol. 8, No. 32. pp. 53432-53449.
@article{c84a8bb43b954a0da5a606482b8d9f27,
title = "An epigenetic signature of adhesion molecules predicts poor prognosis of ovarian cancer patients",
abstract = "DNA methylation is a promising biomarker for cancer. The epigenetic effects of cell adhesion molecules may affect the therapeutic outcome and the present study examined their effects on survival in ovarian cancer. We integrated methylomics and genomics datasets in The Cancer Genome Atlas (n = 391) and identified 106 highly methylated adhesion-related genes in ovarian cancer tissues. Univariate analysis revealed the methylation status of eight genes related to progression-free survival. In multivariate Cox regression analysis, four highly methylated genes (CD97, CTNNA1, DLC1, HAPLN2) and three genes (LAMA4, LPP, MFAP4) with low methylation were significantly associated with poor progression-free survival. Low methylation of VTN was an independent poor prognostic factor for overall survival after adjustment for age and stage. Patients who carried any two of CTNNA1, DLC1 or MFAP4 were significantly associated with poor progression-free survival (hazard ratio: 1.59; 95{\%} confidence interval: 1.23, 2.05). This prognostic methylation signature was validated in a methylomics dataset generated in our lab (n = 37, hazard ratio: 16.64; 95{\%} confidence interval: 2.68, 103.14) and in another from the Australian Ovarian Cancer Study (n = 91, hazard ratio: 2.43; 95{\%} confidence interval: 1.11, 5.36). Epigenetics of cell adhesion molecules is related to ovarian cancer prognosis. A more comprehensive methylomics of cell adhesion molecules is needed and may advance personalized treatment with adhesion molecule-related drugs.",
keywords = "Journal Article",
author = "Ping-Ying Chang and Yu-Ping Liao and Hui-Chen Wang and Yu-Chih Chen and Rui-Lan Huang and Yu-Chi Wang and Chiou-Chung Yuan and Hung-Cheng Lai",
year = "2017",
month = "8",
day = "8",
doi = "10.18632/oncotarget.18515",
language = "English",
volume = "8",
pages = "53432--53449",
journal = "Oncotarget",
issn = "1949-2553",
publisher = "Impact Journals LLC",
number = "32",

}

TY - JOUR

T1 - An epigenetic signature of adhesion molecules predicts poor prognosis of ovarian cancer patients

AU - Chang, Ping-Ying

AU - Liao, Yu-Ping

AU - Wang, Hui-Chen

AU - Chen, Yu-Chih

AU - Huang, Rui-Lan

AU - Wang, Yu-Chi

AU - Yuan, Chiou-Chung

AU - Lai, Hung-Cheng

PY - 2017/8/8

Y1 - 2017/8/8

N2 - DNA methylation is a promising biomarker for cancer. The epigenetic effects of cell adhesion molecules may affect the therapeutic outcome and the present study examined their effects on survival in ovarian cancer. We integrated methylomics and genomics datasets in The Cancer Genome Atlas (n = 391) and identified 106 highly methylated adhesion-related genes in ovarian cancer tissues. Univariate analysis revealed the methylation status of eight genes related to progression-free survival. In multivariate Cox regression analysis, four highly methylated genes (CD97, CTNNA1, DLC1, HAPLN2) and three genes (LAMA4, LPP, MFAP4) with low methylation were significantly associated with poor progression-free survival. Low methylation of VTN was an independent poor prognostic factor for overall survival after adjustment for age and stage. Patients who carried any two of CTNNA1, DLC1 or MFAP4 were significantly associated with poor progression-free survival (hazard ratio: 1.59; 95% confidence interval: 1.23, 2.05). This prognostic methylation signature was validated in a methylomics dataset generated in our lab (n = 37, hazard ratio: 16.64; 95% confidence interval: 2.68, 103.14) and in another from the Australian Ovarian Cancer Study (n = 91, hazard ratio: 2.43; 95% confidence interval: 1.11, 5.36). Epigenetics of cell adhesion molecules is related to ovarian cancer prognosis. A more comprehensive methylomics of cell adhesion molecules is needed and may advance personalized treatment with adhesion molecule-related drugs.

AB - DNA methylation is a promising biomarker for cancer. The epigenetic effects of cell adhesion molecules may affect the therapeutic outcome and the present study examined their effects on survival in ovarian cancer. We integrated methylomics and genomics datasets in The Cancer Genome Atlas (n = 391) and identified 106 highly methylated adhesion-related genes in ovarian cancer tissues. Univariate analysis revealed the methylation status of eight genes related to progression-free survival. In multivariate Cox regression analysis, four highly methylated genes (CD97, CTNNA1, DLC1, HAPLN2) and three genes (LAMA4, LPP, MFAP4) with low methylation were significantly associated with poor progression-free survival. Low methylation of VTN was an independent poor prognostic factor for overall survival after adjustment for age and stage. Patients who carried any two of CTNNA1, DLC1 or MFAP4 were significantly associated with poor progression-free survival (hazard ratio: 1.59; 95% confidence interval: 1.23, 2.05). This prognostic methylation signature was validated in a methylomics dataset generated in our lab (n = 37, hazard ratio: 16.64; 95% confidence interval: 2.68, 103.14) and in another from the Australian Ovarian Cancer Study (n = 91, hazard ratio: 2.43; 95% confidence interval: 1.11, 5.36). Epigenetics of cell adhesion molecules is related to ovarian cancer prognosis. A more comprehensive methylomics of cell adhesion molecules is needed and may advance personalized treatment with adhesion molecule-related drugs.

KW - Journal Article

U2 - 10.18632/oncotarget.18515

DO - 10.18632/oncotarget.18515

M3 - Article

C2 - 28881822

VL - 8

SP - 53432

EP - 53449

JO - Oncotarget

JF - Oncotarget

SN - 1949-2553

IS - 32

ER -