Abstract 5077: Single-cell and bulk RNA sequencing reveal SPINK1 and TIMP1 as epithelial cell marker genes linked to colorectal cancer survival

Abstract Introduction: Colorectal cancer (CRC) is a leading cause of cancer-related mortality, with tumor microenvironment and gene expression playing critical roles in disease progression and prognosis. Identifying specific epithelial cell marker genes linked to survival may enhance prognosis and guide targeted therapies. Methods: Single-cell RNA (scRNA) data for CRC patients was obtained from TISCH database (n=23176 cells), in which we identified EMGs using differential expression clustering analysis. Bulk RNA data was downloaded from The Cancer Genome Atlas (TCGA-COAD, n = 375). Using TCGA-CRC data, EMGs were analyzed using univariate Cox model, followed by Least Absolute Shrinkage and Selection Operator (LASSO) to reduce overfitting. Significant genes from LASSO were then fitted in a multivariate Cox model to screen independent predictors of survival. A survival-related EMG signature was developed to calculate a risk score, classifying patients into low- and high-risk groups based on the median risk score. We validated the risk score using GSE17536 CRC dataset (n=177). xCELL was used to perform immune deconvolution analysis. Results: 142 EMGs were obtained from scRNA and analyzed for prognostic significance in the TCGA bulk-RNA dataset. Univariate Cox analysis retrieved 34 survival-associated genes and was screened using LASSO-cox resulting in 6 genes: CA2, CLCA1, ITLN1, SPINK1, SPINK4, and TIMP1. Multivariate Cox analysis showed that SPINK1 and TIMP1 were independent predictors of survival with SPINK1 associated with better outcomes (HR: 0.87, p-value: 0.009), and TIMP1 with worse outcomes (HR: 1.81, p-value <0.001). The 2 survival-related EMG risk score was calculated, and patients were grouped into high (n = 187) and low-risk (n = 188) groups. The low-risk group showed significantly higher (overall survival) OS and (disease free survival) DFS compared to the high-risk group (Median OS: 23 vs. 21.8 months, p-value = 0.0027; Median DFS: 20.5 vs. 20.3 months, p-value = 0.044). The immune microenvironment analysis showed significant enrichment of dendritic cells, CD8+ central-memory T-cells, class-switched memory B-cells, and macrophages in the high-risk group, while the low-risk group was significantly enriched in gamma delta T-cells, T-helper 1 and T-helper 2 cells, plasma cells, and natural killer cells. In a validation cohort, the low-risk group (n=89) showed better survival outcomes compared to the high-risk group (n=88) (Median OS: 50.2 vs. 35.2 months, Median DFS: 43.8 vs. 24.5 months, p-value = <0.001). Conclusion: Our results showed a significant role of EMGs in CRC prognosis. Among them, SPINK1 and TIMP1 emerged as independent predictors of survival. Differential immune enrichment patterns between risk groups show the potential of integrating EMGs with immune landscape analysis to refine risk stratification and inform personalized therapeutic strategies. Citation Format: Noor Nedal Al-Bzour, Zaid Nassar Abu-Rjai’, Ayah Nedal Al-Bzour, Abdelrahman Qasaymeh, Lujia Chen, Azhar Saeed, Anwaar Saeed. Single-cell and bulk RNA sequencing reveal SPINK1 and TIMP1 as epithelial cell marker genes linked to colorectal cancer survival [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5077.