American Association for Cancer Research
15357163mct190864-sup-228653_2_supp_6055074_q68p17.xlsx (20.49 kB)

Supplementary Table S3 from A Qualitative Transcriptional Signature for Predicting Prognosis and Response to Bevacizumab in Metastatic Colorectal Cancer

Download (20.49 kB)
posted on 2023-04-03, 18:27 authored by Jing Yang, Kai Song, Wenbing Guo, Hailong Zheng, Yelin Fu, Tianyi You, Kai Wang, Lishuang Qi, Wenyuan Zhao, Zheng Guo

Gene expression profile of genes in GPS-64


National Natural Science Foundation of China

Natural Science Foundation of Heilongjiang Province

Joint Scientific and Technology Innovation Found of Fujian Province



Bevacizumab is the molecular-targeted agent used for the antiangiogenic therapy of metastatic colorectal cancer. But some patients are resistant to bevacizumab, it needs an effective biomarker to predict the prognosis and responses of metastatic colorectal cancer (mCRC) to bevacizumab therapy. In this work, we developed a qualitative transcriptional signature to individually predict the response of bevacizumab in patients with mCRC. First, using mCRC samples treated with bevacizumab, we detected differentially expressed genes between response and nonresponse groups. Then, the gene pairs, consisting of at least one differentially expressed gene, with stable relative expression orderings in the response samples but reversal stable relative expression orderings in the nonresponse samples were identified, denoted as pairs-bevacizumab. Similarly, we screened the gene pairs significantly associated with primary tumor locations, donated as pairs-LR. Among the overlapped gene pairs between the pairs-bevacizumab and pairs-LR, we adopted a feature selection process to extract gene pairs that reached the highest F-score for predicting bevacizumab response status in mCRC as the final gene pair signature (GPS), denoted as 64-GPS. In two independent datasets, the predicted response group showed significantly better overall survival than the nonresponse group (P = 6.00e−4 in GSE72970; P = 0.04 in TCGA). Genomic analyses showed that the predicted response group was characterized by frequent copy number alternations, whereas the nonresponse group was characterized by hypermutation. In conclusion, 64-GPS was an objective and robust predictive signature for patients with mCRC treated with bevacizumab, which could effectively assist in the decision of clinical therapy.