A prognostic model based on seven cell adhesion molecule-related genes to predict the prognosis of colorectal cancer

Colorectal cancer (CRC) remains a major contributor to cancer mortality. Cell adhesion molecules (CAMs) hold an essential role in CRC malignant progression. However, how to leverage CAMs-related genes to predict CRC prognosis and guide its treatment is still unclear. Gene expression data and clinical information of CRC patients were obtained from TCGA and GEO databases. Subsequently, a series of analyses were utilized to obtain 7 CAM-related feature genes and set up a prognostic risk model. The riskscore was obtained, and patients were stratified into high-risk (HR) and low-risk (LR) groups. Significant survival differences were observed (P < 0.001). According to the clinicopathological features and risk score, the ROC curve demonstrated that the riskscore showed good prognostic performance. Significant differences were in the immune infiltration of different groups. Patients in the LR group had a higher rate of gene mutation. Through enrichment analysis, the differential genes of the two groups were mainly involved in the Ras signaling pathway, signaling receptor activator activity, and growth factor activity. In addition, CGP-60,474, CMK, Dasatinib, and Imatinib may be targeted drugs in the HR group, while the targeted drugs in the LR group may be Erlotinib, Gemcitabine, JW-7-52-1, and Lapatinib. Expression validation of the 7 model genes was performed using both TCGA transcriptomic data and in vitro qRT-PCR, with consistent trends observed between the two approaches. In summary, the prognostic model can be applied to predict CRC prognosis.

The online version contains supplementary material available at https://doi.org/10.1007/s10616-026-01086-5.