Abstract B026: Multimodal spatial profiling uncovers cell state–associated receptor–ligand interactions in pancreatic ductal adenocarcinoma
Sibyl Drissler, Noor Shakfa, Elizabeth Sunnucks, Tiak Tan, Ferris Nowlan, Chengxin Yu, Cassandra Wong, Brendon Seale, Zhen-Yuan Lin, Michelle Chan-Seng-Yue, Amy Zhang, Sabiq Chaudhary, Ayelet Borgida, Sheng-Ben Liang, Klaudia Nowak, Miralem Mrkonjic, Anna Dodd, Julie Wilson, Kieran Campbell, Jennifer Gorman, Robert Grant, Jennifer Knox, Anne-Claude Gingras, Steven Gallinger, Grainne O’Kane, Barbara Grünwald, Faiyaz Notta, Hartland JacksonAbstract
Malignant cells do not exist in isolation. In pancreatic ductal adenocarcinoma (PDAC), malignant cells reside within distinct spatial niches shaped by stromal, immune, endothelial, and neural compartments, and this spatial context is linked to therapeutic resistance and clinical outcome. While genomic alterations drive malignant cell diversity, their spatial organization suggests that cell-extrinsic cues from the microenvironment further shape malignant phenotype. We previously showed that PDAC tumours harbour an expanded epithelial state spectrum organized into recurrent spatial ecosystems and their associated immune and stromal programs, many of which are prognostic of patient survival. To identify receptor-ligand interactions that may organize these ecosystems, we analyzed receptor-ligand expression across cell type pairs in 8 published PDAC scRNAseq datasets (n=160 patients), revealing substantial heterogeneity across malignant states in their potential for communication with myCAFs, lymphoid cells, and suppressive myeloid populations. Together with MDSC and neutrophil markers nominated by our prior laser-capture microdissection mass spectrometry (LCM-MS) analysis, these data informed the design of a custom 480-gene spatial transcriptomics (Xenium) panel targeting malignant, immune, fibroblast/mural, and endothelial cell states and their receptor-ligand interactions. We applied this panel, H&E histology, and 40-marker imaging mass cytometry (IMC) to a tissue microarray of 128 primary resected PDAC patients tumours with matched whole-genome sequencing (WGS) and clinical outcomes. The IMC panel profiled malignant cell states, extracellular matrix components, phosphorylated signalling proteins, cell death, and neural markers. Co-occurrence and anti-correlation analyses are now nominating activating and inhibitory partners within recurrent malignant-myCAF, malignant-lymphoid, and malignant-MDSC/neutrophil niches across this cohort. Together, this multimodal framework establishes a resource for defining the cellular and molecular architecture of PDAC and for nominating receptor-ligand interactions linked to therapeutic vulnerability and clinical outcome
Citation Format:
Sibyl Drissler, Noor Shakfa, Elizabeth Sunnucks, Tiak Tan, Ferris Nowlan, Chengxin Yu, Cassandra Wong, Brendon Seale, Zhen-Yuan Lin, Michelle Chan-Seng-Yue, Amy Zhang, Sabiq Chaudhary, Ayelet Borgida, Sheng-Ben Liang, Klaudia Nowak, Miralem Mrkonjic, Anna Dodd, Julie Wilson, Kieran Campbell, Jennifer Gorman, Robert Grant, Jennifer Knox, Anne-Claude Gingras, Steven Gallinger, Grainne O’Kane, Barbara Grünwald, Faiyaz Notta, Hartland Jackson. Multimodal spatial profiling uncovers cell state–associated receptor–ligand interactions in pancreatic ductal adenocarcinoma [abstract]. In: Proceedings of the AACR Conference on Pancreatic Cancer: New Frontiers in Biology and Therapeutic Development; 2026 Sep 25-28; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(18_Suppl_2):Abstract nr B026.