Abstract
Backgrounds
Local advanced rectal cancer (LARC) is a common disease occurrence in clinical settings due to its unique anatomical location and treatment approach. The effectiveness of neoadjuvant chemoradiotherapy (nCRT) plays a crucial role in determining the appropriate treatment and prognosis for patients. Currently, there exists no universally acknowledged benchmark for prognosticating the effectiveness of neoadjuvant therapy. Our study obtained the GSE150082 dataset from the NCBI Geo Public Database, consisting of expression profile data for 39 groups of patients who either responded to nCRT or did not. Differential gene analysis was performed using the Limma package with a significance threshold of p < 0.05 and |log fold change| > 0.585. The functions and pathways associated with the differentially expressed genes were analyzed, and a protein interaction network was constructed using Cytoscape software. Additionally, the TCGA data was used to identify prognostic-related genes from the differential genes through Cox univariate regression and the lasso regression algorithm. Predictive models were then constructed and validated using both internal and external datasets.
Results
A total of 633 differentially expressed genes associated with nCRT were identified, comprising 238 up-regulated and 395 down-regulated genes. These genes are predominantly enriched in pathways related to innate immune response, regulation of biological stimulus-response, and cell activation. The results from gene screening and the construction of a predictive model demonstrate the model's efficacy in effectively distinguishing between high- and low-risk patients. This predictive capability was validated in both the training set and an external validation set. Additionally, an analysis of the relationship between the risk score and immune infiltration in the tumor microenvironment unveiled a potential molecular mechanism, suggesting that the risk score may impact patient prognosis by modulating specific immune cell populations and immune-related genes.
Conclusions
MCOLN3、CINP、HAND2 and CCDC85A might be the identified key genes and play a critical role in several pathways associated with response to nCRT. Furthermore, a risk score model was constructed based on prognosis-related genes, providing potential molecular markers and therapeutic targets for personalized strategies in nCRT. Our findings could potentially offer a fresh and innovative outlook for future treatment for patients with LARC.