Abstract 5091: Exploration of conversion therapies target in pancreatic cancer with liver metastasis in single cell spatial transcriptomic levels

Abstract Objectives: Multidisciplinary treatment drives conversion surgery of pancreatic cancer with liver metastatic clinical cases with unresectable features. The adjuvant regime included AG and FOLFIRINOX and partial cases exhibited no poor response for the treatment. We analyzed single cell spatial transcriptome data to explore biomarkers of therapy. Materials and Methods: The GSE202051 and GSE199102 cohorts derived from GEO database. The quality control, analysis, and plotting of single cell spatial transcriptome rely on the R packages including R packages ISCHIA, robustbase, data.table, ggplot2, Seurat, dplyr, factoextra, cluster, showtext, gridExtra, and pdftools. Results: Malignant biomarkers in liver metastatic samples We drew dim plots to display spatial distribution of various types cells in primary and liver metastatic pancreatic cancer samples. We selected subset of tumor epithelial cells to analyze clusters and displayed c0-c26 tumor epithelial clusters in primary and liver metastatic two groups with dim plots and bar plot. The clusters among c0(59.2%), c5(32.8%), c6(32.4%), and c10(21.9%) possessed more enrichment distribution in liver metastatic region. We used the function FindAllMarkers to analyze top differentiation expression genes in clusters and found three prominent genes including TNFSF14, CYP2E1, and LCN2(Fig1).Stromal microenvironment evolution driven liver metastatic biomarkers gene We calculated and settled each cell type proportion data in each sample based on spatial transcriptome information. Subsequently, we drew the correlation heatmap to discover the positive correlation of TNFSF14 expression levels in tumor epithelial cells with endothelial cells(R=0.64, pvalue=0.032) and negative correlation of CYP2E1 with CAF(R=-0.91, pvalue<0.001) in primary cell type proportion. We identify CYP2E1 driving myCAF changes in stromal microenvironment. For the purpose of exploring extinct myCAF subtypes in volution, we also displayed the correlation heatmap to discover the correlation results of c0-c14 myCAF clusters with CYP2E1 gene expression. Interestingly, only positive result demonstrated positive correlation between c5 cluster and CYP2E1 gene(R=0.76, pvalue<001). The heatmap displayed that CYP2E1 expressed prominently in c5 cluster myCAF. Simultaneously, we identified the expression tendency in samples with neoadjuvant therapy and displayed the positive correlation between CYP2E1 and poor response(Fig2). Conclusion: Our study used single cell spatial transcriptome data with in-depth explaining spatial distribution and expression levels and explored vital biomarker CYP2E1 in pancreatic cancer with liver metastasis. A series of experiments and clinical trials would be designed and conducted by our team in future. Citation Format: Zijin Xu, Qiang Zhou, Ying Ying, Zeyuan Cao, Wenquan Wang, Huaxiang Xu, Liang Liu. Exploration of conversion therapies target in pancreatic cancer with liver metastasis in single cell spatial transcriptomic levels [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 5091.