AUTHOR=He Zi-Xuan , Zhao Sheng-Bing , Fang Xue , E Ji-Fu , Fu Hong-Yu , Song Yi-Hang , Wu Jia-Yi , Pan Peng , Gu Lun , Xia Tian , Liu Yi-Long , Li Zhao-Shen , Wang Shu-Ling , Bai Yu TITLE=Prognostic and Predictive Value of BGN in Colon Cancer Outcomes and Response to Immunotherapy JOURNAL=Frontiers in Oncology VOLUME=11 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.761030 DOI=10.3389/fonc.2021.761030 ISSN=2234-943X ABSTRACT=Background

Colon cancer is one of the most frequent malignancies and causes high mortality worldwide. Exploring the tumor-immune interactions in the tumor microenvironment and identifying new prognostic and therapeutic biomarkers will assist in decoding the novel mechanism of tumor immunotherapy. BGN is a typical extracellular matrix protein that was previously validated as a signaling molecule regulating multiple processes of tumorigenesis. However, its role in tumor immunity requires further investigation.

Methods

The differentially expressed genes in three GEO datasets were analyzed, and BGN was identified as the target gene by intersection analysis of PPIs. The relevance between clinical outcomes and BGN expression levels was evaluated using data from the GEO database, TCGA and tissue microarray of colon cancer samples. Univariable and multivariable Cox regression models were conducted for identifying the risk factors correlated with clinical prognosis of colon cancer patients. Next, the association between BGN expression levels and the infiltration of immune cells as well as the process of the immune response was analyzed. Finally, we predicted the immunotherapeutic response rates in the subgroups of low and high BGN expression by TIS score, ImmuCellAI and TIDE algorithms.

Results

BGN expression demonstrated a statistically significant upregulation in colon cancer tissues than in normal tissues. Elevated BGN was associated with shorter overall survival as well as unfavorable clinicopathological features, including tumor size, serosa invasion and length of hospitalization. Mechanistically, pathway enrichment and functional analysis demonstrated that BGN was positively correlated with immune and stromal scores in the TME and primarily involved in the regulation of immune response. Further investigation revealed that BGN was strongly expressed in the immunosuppressive phenotype and tightly associated with the infiltration of multiple immune cells in colon cancer, especially M2 macrophages and induced Tregs. Finally, we demonstrated that high BGN expression presented a better immunotherapeutic response in colon cancer patients.

Conclusion

BGN is an encouraging predictor of diagnosis, prognosis and immunotherapeutic response in patients with colon cancer. Assessment of BGN expression represents a novel approach with great promise for identifying patients who may potentially benefit from immunotherapy.