Abstract:

Objective

This study aims to identify miRNAs associated with overall survival （OS）in colorectal cancer patients using The Cancer Genome Atlas （TCGA） database，develop a miRNA-based nomogram prognostic model，and assess the model's predictive capability.

Methods

miRNA sequencing data and clinical information of colorectal cancer patients were downloaded and integrated from the TCGA database. The data were randomly divided into a training cohort and a validation cohort in a 3:1 ratio.Univariate Cox regression，LASSO regression，and multivariate Cox regression analyses were employed to identify a set of miRNAs related to colorectal cancer prognosis and to establish a prognostic risk score for colorectal cancer. A nomogram model was then constructed by combining the risk score with clinical indicators. The predictive performance of the model was evaluated using receiver operating characteristic （ROC）curves，concordance index （C-index），calibration curves，and decision curve analysis （DCA）.

Results

A total of 498 CRC patients were included in the study. Differential analysis identified 291 miRNAs. The risk score was calculated as follows: Risk Score=（0.05634381×miR-548u expression）+（0.03900542×miR-4665-5p expression）- （0.10097599×miR-887-3p expression）. A nomogram prognostic model was constructed incorporating the risk score，age，and TNM stage. In the validation cohort，the AUC values of the nomogram prognostic model，risk score，and TNM stage were 0.752，0.720，and 0.673，respectively. The C-index of the nomogram prognostic model in the training and validation cohorts were 0.743 and 0.761，respectively. The calibration curve demonstrated good agreement between the nomogram's predicted and actual 5-year OS.DCA showed that the nomogram prognostic model offered greater clinical benefit compared to the TNM staging system.

Conclusion

The nomogram prognostic model demonstrates strong predictive ability and may aid in clinical decision-making and prognosis assessment for colorectal cancer patients.

Key words: Colorectal neoplasms, miRNA, Prognostic model, TCGA database