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

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

JOURNAL=Frontiers in Cell and Developmental Biology

VOLUME=Volume 9 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2021.664415

DOI=10.3389/fcell.2021.664415

ISSN=2296-634X

ABSTRACT=Dear Editors,
  Enclosed please find the description of the colon cancer prognostic predictive model entitled “Accurate prediction of hierarchy prognosis integrated clinical and molecular characteristic in colon cancer”, wrote by Wang et al. To highlight the importance of the results, we are submitting it to Frontiers in Cell and Developmental Biology.The difference of prognosis for colon cancer is associated with complicated factors. This increases the difficulty of accurate prognosic assessment and treatment decision in clinical practice.In the study, a series of prediction models were exploited to screen potential factors for stratified survival probability  in colon cancer,based on SEER database and TCGA database by machine learning and statistical model. The 9 clinical characteristics of 161,694 postoperative patients with adenocarcinoma were obtained in colon cancer, based on SEER. We also collected clinical and four types molecular datasets from the TCGA. To better assess the prognosis of colon cancer using clinical features, we redefined the ratio of positive lymph nodes(LNR) by information gain method. LNR thresholds for one and three years survival time were 0, 0.2 and 0.6, for five years were 0, 0.3 and 0.7. . For the survival of one, three and five years, We obtained 8, 9 and 4 features consisting of combination clinical factors and a group of molecular features to predict the difference between high and low risk. In all feature sets, molecular risk score was the best factor for survival assessment, the top one of molecular features was DNA methylation.T4&poorly differentiated or undifferentiated was the most important clinical factor for one year, and M0 was for three and five years overall lifetime. According to the combination of clinical factors and molecular features, the classier AUC, survivalROC, nomogram, C-index and Kaplan-Meier survival curves were displayed the results of different follow-up time. 
  Our study is original, unpublished and is not being considered for publication elsewhere. All authors have contributed to this study and approved to submit to the journal. We would be grateful if the study could be reviewed and published in Frontiers in Cell and Developmental Biology
Best regards,
Yan Zhang