RESUMO
BACKGROUND: Biomarkers for colorectal cancer (CRC) can complement population screening methods, but so far, few plasma proteins have been identified as biomarkers for CRC. This study aims to identify potential protein biomarkers and therapeutic targets for CRC within the proteome range. METHODS: We extracted summary-level data of circulating protein from 7 published genome-wide association studies (GWASs) of plasma proteome for Mendelian randomization (MR), summary-data-based MR (SMR), and co-localization analyses to screen and validate proteins with causal effects in CRC. In addition, we further conducted druggability evaluation, prognosis analysis at the transcriptional level, and enrichment expression at the single-cell level, highlighting the important role of these plasma protein biomarkers in CRC. RESULTS: We identified 117 plasma protein biomarkers associated with CRC risk, with 9 proteins showing stronger genetic correlations in Bayesian co-localization (PP.H4 > 0.70). Further, we found 26 protein-coding genes already used in targeted drug development and may potentially become therapeutic targets for CRC. In prognosis analysis, the encoding genes of plasma proteins exhibited consistent effects with MR analysis and can serve as prognostic biomarkers for CRC. Additionally, we also found that the differentially expressed proteins are mainly expressed in fibroblasts, endothelial cells, macrophages, and T cells. CONCLUSION: Our study has identified plasma protein biomarkers associated with CRC risk, which may complement population screening methods for CRC and achieve more precise treatment for patients.
RESUMO
PURPOSE: Metabolic reprogramming is a novel hallmark and therapeutic target of cancer. Our study aimed to establish fatty acid metabolism-associated scores based on gene signature and investigated its effects on immunotherapy in colon cancer. METHODS: Gene expression and clinical information were collected from Gene Expression Omnibus (GEO) database to identify a gene signature by non-negative matrix factorization (NMF) clustering and Cox regression analysis. Subsequently, we constructed the fatty acid metabolism score (FA-score) model by principal component analysis (PCA) and explored its relativity of prognosis and the response to immunotherapy in colon cancer. Finally, the Cancer Genome Atlas (TCGA) database was introduced and in vitro study was performed for verification. RESULTS: The FA-score-high group had a higher level of fatty acid metabolism and was associated with worse patient overall survival. Significantly, FA-score correlated closely with the biomarkers of immunotherapy, and the FA-score-high group had a poorer therapeutic efficacy of immune checkpoint blockade. In vitro experiments demonstrated that ACSL5 may be a critical metabolic regulatory target. CONCLUSIONS: Our study provided a comprehensive analysis of the heterogeneity of fatty acid metabolism in colon cancer. We highlighted the potential clinical utility of fatty acid metabolism-related genes to be biomarkers of colon cancer prognosis and targets to improve the effect of immunotherapy.