Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive cancers, with less than 10% of patients surviving more than five years. A major challenge of PDAC is understanding the complexity of the tumor architecture and identifying targetable cellular phenotypes that can inform clinical decision-making. Hwang et al. recently harnessed spatially-resolved transcriptomics to characterize gene expression profiles across PDAC tumor slices and identified recurrent expression patterns in malignant and fibroblast cellular neighborhoods that correlated with treatment outcomes.