Pancreatic tumor organoids for modeling in vivo drug response and discovering clinically-actionable biomarkers

Abstract

Patient-derived models are transforming translational cancer research. It is not clear if the emergence of patient-derived organoid (PDO) models can extend the utility of the widely used patient-derived xenograft (PDX). In addition, the utility of PDO models for serum biomarker discovery is not known. Here, we demonstrate that PDO models recapitulate the genomics, cell biology, glycomics and drug responses observed in PDX models. Furthermore, we demonstrate the applicability of PDO models for identification of N-glycans that are enriched in the glycome of pancreatic ductal adenocarcinoma (PDAC). Surprisingly, among all the glycans observed in PDX and PDOs, a core set of 57 N-glycans represent 50-94% of the relative abundance of all N-glycans detected, suggesting that only a subset of glycans dominate the cell surface landscape in PDAC. In addition, we outline a tumor organoid-based pipeline to identify surface proteins in extracellular vesicles (EV) from media supernatant of PDO cultures. When combined with the affinity-based validation platform, the EV surface proteins discovered in PDOs are effective in differentiating patients with PDAC from those with benign pancreatitis in the clinic, identifying PDO as powerful discovery platform for serum biomarkers. Thus, PDOs extend the utility of the archival collections of PDX models for translational research and function as a powerful platform for identification of clinically-actionable biomarkers in patients blood. Significance statementTumor organoids extend the utility of PDX models as platforms for investigating drug response, glycosylation changes and function as new platforms for discovering blood-based biomarkers