Table of Contents  

Tuveson: Organoids to study and stop pancreatic cancer

Pancreatic cancer remains the most deadly of the common malignancies, as a result of the lack of a viable early detection method and the limited activity of most drugs. To increase our basic understanding of this disease and advance medical approaches, we have recently co-developed, with Hans Clevers, mouse and human pancreatic cancer organoid model systems.1 Organoid cultures can be robustly prepared from either normal pancreatic ducts or any stage of neoplastic pancreatic tissue, including preinvasive carcinomas. Additionally, orthotopically transplanted neoplastic organoids remarkably demonstrate cellular plasticity and model the progression of pancreatic cancer from a low-grade preinvasive carcinoma to an invasive and metastatic carcinoma over several months. Thus, the in vitro and in vivo organoid models provide a platform to explore new diagnostic and therapeutic strategies, facilitate the systematic analysis of neoplastic biochemical cascades and also enable the dissection of tumour microenvironment interactions.

Having a panel of organoids that represent all stages of pancreatic cancer development has allowed us to develop very sensitive protein-based methods to identify early pancreatic cancers. A current method for following pancreatic cancer patients during treatment is to measure circulating CA19-9. CA19-9 is a carbohydrate epitope affixed to many proteins and occurs as a result of incomplete glycosylation processing in carcinoma cells. Although CA19-9 is a useful measurement for predicting the effectiveness of therapeutic interventions, because CA19-9 is elevated during pancreatic inflammation (pancreatitis) and hepatic or biliary dysfunction, it is neither specific nor sensitive enough to screen patients for pancreatic cancer. We sought to use the organoid series to identify protein carriers of CA19-9, incorporating a cross-species approach with both murine and human organoids to establish a more robust list of potential carriers. However, because several enzymes required to establish the CA19-9 glycan are not expressed in mice, we first humanized the mouse glycome such that it could produce CA19-9. Immune precipitation of CA19-9 carriers was carried out in the murine and human organoid series, followed by trypsinization and mass spectrometry identification. By subtracting the carriers present in wild-type proliferating pancreatic ductal cells, we can more clearly propose potential biomarkers of pancreatic cancer. We have found that many previously described CA19-9 protein carriers are expressed and released by proliferating normal ductal cells, deprioritizing them for future work. Importantly, we have also identified dozens of novel protein carriers of CA19-9 in pancreatic cancer organoids, and our early work has now revealed that many are present in the blood of pancreatic cancer patients but not in patients with benign pancreatic or liver pathologies. We are now exploring whether or not these proteins may serve as biomarkers of early pancreatic cancer in patients.

The organoid series has also allowed us to query the function of important mediators of pancreatic cancer oncogenesis. Although reactive oxygen species (ROS) potently affect cancer cell proliferation and influence therapeutic responsiveness, the critical biochemical pathways that are deterministic towards cell proliferation as opposed to cell death are still unclear. Using mouse models and organoids we found that oncogenic Kras induces the Nrf2/Nfe2l2 transcription factor to stimulate pancreatic cancer proliferation, and to maintain the viability of malignant cells. Nrf2 is the master regulator of the antioxidant response and works by directing the transcription of genes that promote a reduction in intracellular ROS levels. By establishing a sensitive redox proteomics method, we identified translational regulatory proteins to be specifically oxidized when Nrf2 was deleted from pancreatic cancer cells. Both cap-dependent and -independent messenger ribonucleic acid (mRNA) translation is impaired in Nrf2-deficient pancreatic cancer cells, as a result of both the direct oxidation of the protein translation machinery and the impairment of mitogenic signalling, and both antioxidants and exogenous mitogens can mitigate this defect. Mutational analysis revealed that multiple translation pathway proteins serve as redox switches. Pro-oxidants partially phenocopy the redox properties of Nrf2 by lowering the glutathione pools, and they potently synergize with inhibitors of PI3K/Akt to promote pancreatic cancer cell death in vitro and in vivo. Thus, mRNA translation is particularly impaired by elevated reactive oxygen species in pancreatic cancer, presenting a synthetic lethal strategy. We are now optimizing the therapeutic approaches that target protein translation in pancreatic ductal adenocarcinoma (PDA) patients for consideration of imminent evaluation in patients.2

Finally, the co-culture of organoids with pancreatic stellate cells (PSCs) promotes the activation of each cell type and results in the production of extracellular matrix. PSCs are proposed to differentiate to cancer-associated fibroblasts (CAFs) and thereby produce a prominent desmoplastic stroma that modulates disease progression and therapeutic responses in PDA. However, it is unknown whether CAFs uniformly carry out these putative tasks or if different types of CAFs with distinct phenotypes in PDA collectively participate. Using a panel of markers indicative of CAFs, we found a subpopulation located immediately adjacent to neoplastic cells in PDA tumours that had an elevated expression of alpha-smooth muscle actin (aSMA). We recapitulated this finding in cocultures of murine-naive PSCs and PDA organoids, and showed co-operative interactions between neoplastic cells and CAFs. Furthermore, the co-cultures revealed another distinct subpopulation of CAFs located distant from neoplastic cells that lacked aSMA expression and instead secreted high levels of interleukin 6 (IL-6) and additional inflammatory mediators. These findings were corroborated in PDA tissues to demarcate two mutually exclusive populations of CAFs: aSMAhighIL-6low myofibroblastic CAFs (‘myCAFs’) and aSMAlowIL-6high inflammatory CAFs (‘iCAFs’). These two CAF populations are functionally distinct and can interconvert in vitro, providing direct evidence for CAF heterogeneity in PDA tumour biology, with implications for disease aetiology and therapeutic development.



Boj SF, Hwang CI, Baker LA, et al. Organoid models of human and mouse ductal pancreatic cancer. Cell 2015; 160:324–38.


Chio II, Jafarnejad SM, Ponz-Sarvise M, et al. NRF2 promotes tumor maintenance by modulating mRNA translation in pancreatic cancer. Cell 2016; 166:963–76.

Add comment 

Home  Editorial Board  Search  Current Issue  Archive Issues  Announcements  Aims & Scope  About the Journal  How to Submit  Contact Us
Find out how to become a part of the HMJ  |   CLICK HERE >>
© Copyright 2012 - 2013 HMJ - HAMDAN Medical Journal. All Rights Reserved         Website Developed By Cedar Solutions INDIA