Research Paper Volume 11, Issue 24 pp 11880—11892

Molecular subtypes based on DNA methylation predict prognosis in colon adenocarcinoma patients

Changshun Yang1, *, , Yu Zhang2, *, , Xiaoqin Xu3, , Weihua Li1, *, ,

  • 1 Department of Surgical Oncology, Fujian Provincial Hospital, Fuzhou 350001, China
  • 2 Department of Pathology, The First Affiliated Hospital of Fujian Medical University, Fuzhou 350001, China
  • 3 School of Public Health, Fujian Medical University, Fuzhou 350001, China
* Equal contribution

Received: May 16, 2019       Accepted: November 17, 2019       Published: December 18, 2019
How to Cite

Copyright © 2019 Yang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Tumor heterogeneity makes early diagnosis and effective treatment of colon adenocarcinoma difficult. As an important regulator of gene expression, DNA methylation can influence tumor heterogeneity. In this study, we explored the prognostic value of subtypes based on DNA methylation status in 424 colon adenocarcinoma samples from the Cancer Genome Atlas database. Differences in DNA methylation levels were associated with differences in T, N, and M category, age, stage, and prognosis. Seven subgroups were identified based on consensus clustering using 356 CpG sites that significantly influenced survival. Finally, a prognostic model was constructed and used to classify samples in a testing dataset into seven DNA methylation subgroups based on the classification results of a training dataset. These specific classifications based on DNA methylation may help account for heterogeneity within previously established molecular subgroups of colon adenocarcinoma and could potentially aid in the development of more effective personalized treatments.


TSG: tumor suppressor genes; SAM: S-adenosyl-methionine; 5-mC: 5-methylcytosine; DNMT: DNA methyltransferase; TSS: transcription start site; MSI: microsatellite instability; CDF: Cumulative Distribution Function; AUC: area under curve; FDR: false discovery rate; ROC curve: receiver operating characteristic curve; FC: fold change; KNN: K-nearest neighbor; COAD: colon adenocarcinoma; T category: primary tumor; N category: regional lymph nodes; M category: distant metastasis.