Abstract Precision medicine approaches in pancreatic ductal adenocarcinoma (PDAC) are imperative for improving disease outcomes. However, the long-term fidelity of recently deployed ex vivo preclinical platforms, such as patient-derived organoids (PDOs) remains unknown. Through single-cell RNA sequencing (scRNA-seq), we identify substantial transcriptomic evolution of PDOs propagated from the parental tumor, which may alter predicted drug sensitivity. In contrast, scRNA-seq is readily applicable to limited biopsies from human primary and metastatic PDAC and identifies most cancers as being an admixture of previously described epithelial transcriptomic subtypes. Integrative analyses of our data provide an in-depth characterization of the heterogeneity within the tumor microenvironment, including cancer-associated fibroblast (CAF) subclasses, and predicts for a multitude of ligand-receptor interactions, revealing potential targets for immunotherapy approaches. While PDOs continue to enable prospective therapeutic prediction, our analysis also demonstrates the complementarity of using orthogonal de novo biopsies from PDAC patients paired with scRNA-seq to inform clinical decision-making. Statement of Significance The application of single-cell RNA sequencing to diagnostic pancreatic cancer biopsies provides in-depth transcriptomic characterization of the tumor epithelium and microenvironment, while minimizing potential artifacts introduced by an intervening ex vivo passaging step. Thus, this approach can complement the use of patient-derived organoids in implementing precision oncology.