BACKGROUND: Recently, oxidative stress and inflammatory responses have been shown to directly impact tumor growth and the tumor microenvironment (TME). However, more research is necessary to fully understand the relationship between oxidative stress and inflammatory responses and colorectal cancer (CRC). METHODS: The FindCluster algorithm was used to extract CRC Single-cell RNA sequencing (scRNA-seq) data and identify tumor cell groupings. From the MSigDB database, genes associated with oxidative stress and the inflammatory response were taken. We identified molecular subtypes and built a predictive risk model with the LASSO-Cox method using the ConsensusClusterPlus software suite. We incorporated the prognostic risk model and other clinicopathological parameters into a column-line chart. Finally, we used Quantitative Polymerase Chain Reaction (qPCR) and immunohistochemistry to check the expression of the unreported hub model genes. Cell proliferation was assessed using EDU and colony formation assays. Reactive Oxygen Species (ROS) tests were used to quantitatively determine the ROS content in CRC cells. The ability of CRC cells to invade and migrate was examined using transwell experiments. The regulatory functions of hub model genes were discovered in vivo using a xenograft model tumor assay. RESULTS: Oxidative stress and inflammatory response factors in monocytic/macrophages of CRC were significantly upregulated, and their oxidative stress and inflammatory response functions were significantly higher than those of other cell subgroups, as indicated by the enrichment score. These factors showed significant synergistic overexpression and enrichment in this cell population. We constructed a prognostic risk model consisting of seven signatures. The good and stable prognostic evaluation efficacy of the model was confirmed, and risk scores were determined to be independent prognostic factors for CRC. We explored the relationship between the risk score model and malignant progression of tumor cells, tumor immune microenvironment, genomic variation, chemotherapy resistance, and immune response. Further qPCR and immunohistochemistry analysis showed that the expression of ZNF385A was high in CRC tissues. The functional experiment results indicated that interfering with the expression of ZNF385A could suppress the proliferation, ROS, migration and invasion of SW620 cells in vitro and the growth of xenograft tumors in vivo. CONCLUSION: In this study, we investigated the critical expression patterns of oxidative stress- and inflammatory response-related genes in CRC, which may contribute to the prognosis and immunotherapy of CRC. Additionally, we discovered ZNF385A to be a novel oncogene in CRC. These findings imply that this model may be applied to assess prognostic risk and identify potential therapeutic targets for CRC patients.