AUTHOR=Wang Hao , Liu Dong , Liang Hanyang , Ba Zhengqing , Ma Yue , Xu Haobo , Wang Juan , Wang Tianjie , Tian Tao , Yang Jingang , Gao Xiaojin , Qiao Shubin , Qu Yanling , Yang Zhuoxuan , Guo Wei , Zhao Min , Ao Huiping , Zheng Xiaodong , Yuan Jiansong , Yang Weixian 

TITLE=A Nomogram for Predicting Survival in Patients With Colorectal Cancer Incorporating Cardiovascular Comorbidities

JOURNAL=Frontiers in Cardiovascular Medicine

VOLUME=Volume 9 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2022.875560

DOI=10.3389/fcvm.2022.875560

ISSN=2297-055X

ABSTRACT=Background: Cardiovascular comorbidities (CVCs) affect the overall survival (OS) of patients with colorectal cancer (CRC), however, a prognostic evaluation system for these patients is currently lacking.
Objectives: This study aimed to develop and validate a nomogram, which takes CVCs into account, for predicting the survival of patients with CRC.
Methods: 21432 patients with CRC were recruited from 4 centers in China between January 2011 and December 2017, the nomogram was constructed, based on Cox regression, using training cohort (19102 patients), and validated using validation cohort (2330 patients). The discrimination and calibration of the model were assessed by concordance index and calibration curve. The clinical utility of the model was measured by decision curve analysis. Based on the nomogram, we divided patients into three groups: low, middle and high risk.
Results: Independent risk factors selected into our nomogram for OS including age, metastasis, malignant ascites, heart failure and venous thromboembolism, whereas dyslipidemia was found to be a protective factor. The c-index of our nomogram was 0.714 (95%CI: 0.708-0.720) in the training cohort and 0.742 (95%CI: 0.725-0.759) in the validation cohort. Calibration curve and decision curve analysis showed the reliability of the model. The cutoff values of the three groups were 68.19 and 145.44, which were also significant in the validation cohort (p<0.001).
Conclusion: Taken CVCs into account, an easy-to-use nomogram was provided to estimate OS for patients with CRC, improving the prognostic evaluation ability.