Accurate Prediction of Prognosis by Integrating Clinical and Molecular Characteristics in Colon Cancer

Liru Wang; Mu Su; Mengyan Zhang; Hongyan Zhao; Hongli Wang; Jie Xing; Chenyu Guo; Dianshuang Zhou; Wenhui Xue; Haibo Lu; Yan Zhang

doi:10.3389/fcell.2021.664415

Accurate Prediction of Prognosis by Integrating Clinical and Molecular Characteristics in Colon Cancer

Front Cell Dev Biol. 2021 May 21:9:664415. doi: 10.3389/fcell.2021.664415. eCollection 2021.

Authors

Liru Wang^{1

2}, Mu Su¹, Mengyan Zhang¹, Hongyan Zhao³, Hongli Wang¹, Jie Xing¹, Chenyu Guo¹, Dianshuang Zhou¹, Wenhui Xue¹, Haibo Lu⁴, Yan Zhang^{1

5}

Affiliations

¹ School of Life Science and Technology, Computational Biology Research Center, Harbin Institute of Technology, Harbin, China.
² Department of Medical Oncology, Heilongjiang Provincial Hospital, Harbin, China.
³ Department of Gastroenterology, The Fourth Affiliated Hospital of Harbin Medical University, Harbin, China.
⁴ Department of Gastrointestinal Medical Oncology, Harbin Medical University Cancer Hospital, Harbin, Heilongjiang, China.
⁵ Guangzhou Institute of Respiratory Health, State Key Laboratory of Respiratory Disease, Guangzhou Medical University, Guangzhou, China.

Abstract

Various factors affect the prognosis of patients with colon cancer. Complicated factors are found to be conducive to accurate assessment of prognosis. In this study, we developed a series of prognostic prediction models for survival time of colon cancer patients after surgery. Analysis of nine clinical characteristics showed that the most important factor was the positive lymph node ratio (LNR). High LNR was the most important clinical factor affecting 1- and 3-year survival; M0&age < 70 was the most important feature for 5 years. The performance of the model was improved through the integration of clinical characteristics and four types of molecule features (mRNA, lncRNA, miRNA, DNA methylation). The model provides guidance for clinical practice. According to the high-risk molecular features combined with age ≥ 70&T3, poorly differentiated or undifferentiated, M0&well differentiated, M0&T2, LNR high, T4&poorly differentiated, or undifferentiated, the survival time may be less than 1 year; for patients with high risk of molecular features combined with M0&T2, M0&T4, LNR 0& M0, LNR median &T3, and LNR high, the survival is predicted less than 3 years; and the survival of patients with M1&T3, M0 and high risk molecular features is less than 5 years. Using multidimensional and complex patient information, this study establishes potential criteria for clinicians to evaluate the survival of patients for colon cancer.

Keywords: clinical feature; combination; conlon cancer; molecular characteristic; prognostic model.