Abstract 5548: Single-cell transcriptome analysis reveals the role of myofibroblastic cancer-associated fibroblasts in chemotherapy resistance in pancreatic cancer

Abstract Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal cancer types often diagnosed late due to lack of specific biomarkers and its high capacity to metastasize early. In PDAC, tumor microenvironment (TME) is particularly complex, representing up to 80% of the tumor mass. Thus, better understanding of the cellular states and their controlling transcriptional networks within both malignant and stromal cells is pertinent to improve the survival of PDAC patients. The goal of this project is to elucidate the dynamic non-genetic regulatory mechanisms that contribute to PDAC progression and development of chemoresistance. We have performed in-silico analysis of single-cell RNA-sequencing data from >200,000 cells derived from pancreatic cancer patients with and without neoadjuvant treatment. Based on their gene expression profiles, we identified 12 distinct cell types including three major fibroblast subtypes: regular cancer-associated fibroblasts (CAF), inflammatory CAFs (iCAF), and myofibroblastic CAFs (myCAF). We found that tumors from patients with poor treatment response as well as from untreated patients exhibited a higher proportion of myCAFs compared to treatment-responsive patients. Using specific markers derived from each fibroblast subtype, we applied these signatures to the PAAD cohort from TCGA and found a significant association between the myCAF signature and worse prognosis as well as advanced TNM staging. Functional analyses further demonstrated elevated levels of Wnt/β-Catenin and Notch signaling pathways in the untreated and poor-response groups. Second, we utilized state-of-the-art machine learning methods to model robust relations between transcription factors (TF) and gene regulatory networks from treatment-naïve and neoadjuvant-treated PDAC patients. Our analysis revealed both well-known and new TFs related to pancreatic cell identity and PDAC progression. In-silico perturbation of these TFs was performed to study their effect on gene expression and cell state dynamics. The predictions reported a strong activity from both well-known and newly predicted TFs regarding the cell state transitions and their regulatory role. Currently, we are further characterizing the identified cell states and associated transcriptional programs, and the findings will be presented in the meeting. Citation Format: Jiahao Guo, Samuele Cancellieri, Liangru Fei, Tero Aittokallio, Biswajyoti Sahu. Single-cell transcriptome analysis reveals the role of myofibroblastic cancer-associated fibroblasts in chemotherapy resistance in pancreatic cancer [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 5548.