Open AccessA nineteen gene‐based risk score classifier predicts prognosis of colorectal cancer patients
Author(s) -
Kim Seon-Kyu,
Kim Seon-Young,
Kim Jeong-Hwan,
Roh Seon Ae,
Cho Dong-Hyung,
Kim Yong Sung,
Kim Jin Cheon
Publication year - 2014
Publication title -
molecular oncology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.332
H-Index - 88
eISSN - 1878-0261
pISSN - 1574-7891
DOI - 10.1016/j.molonc.2014.06.016
Subject(s) - medicine , colorectal cancer , oncology , proportional hazards model , pathological , adjuvant chemotherapy , multivariate analysis , cancer , breast cancer
Colorectal cancer (CRC) patients frequently experience disease recurrence and distant metastasis. This study aimed to identify prognostic indicators, including individual responses to chemotherapy, in CRC patients. RNA‐seq data was generated using 54 samples (normal colon, primary CRC, and liver metastases) from 18 CRC patients and genes associated with CRC aggressiveness were identified. A risk score based on these genes was developed and validated in four independent CRC patient cohorts ( n = 1063). Diverse statistical methods were applied to validate the risk scoring system, including a generalized linear model likelihood ratio test, Kaplan–Meier curves, a log‐rank test, and the Cox model. TREM1 and CTGF were identified as two activated regulators associated with CRC aggressiveness. A risk score based on 19 genes regulated by TREM1 or CTGF activation (TCA19) was a significant prognostic indicator. In multivariate and subset analyses based on pathological staging, TCA19 was an independent risk factor (HR = 1.894, 95% CI = 1.227–2.809, P = 0.002). Subset stratification in stage III patients revealed that TCA19 had prognostic potential and identified patients who would benefit from adjuvant chemotherapy, regardless of age. The TCA19 predictor represents a novel diagnostic tool for identifying high‐risk CRC patients and possibly predicting the response to adjuvant chemotherapy.