Tumor cell–derived prostaglandin E2 (PGE2) is a tumor cell–intrinsic factor that supports immunosuppression in the tumor microenvironment (TME) by acting on the immune cells, but the impact of PGE2 signaling in tumor cells on the immunosuppressive TME is unclear. We demonstrate that deleting the PGE2 synthesis enzyme or disrupting autocrine PGE2 signaling through EP4 receptors on tumor cells reverses the T cell–low, myeloid cell–rich TME, activates T cells, and suppresses tumor growth. Knockout (KO) of Ptges (the gene encoding the PGE2 synthesis enzyme mPGES-1) or the EP4 receptor gene (Ptger4) in KPCY (KrasG12D P53R172H Yfp CrePdx) pancreatic tumor cells abolished growth of implanted tumors in a T cell–dependent manner. Blockade of the EP4 receptor in combination with immunotherapy, but not immunotherapy alone, induced complete tumor regressions and immunological memory. Mechanistically, Ptges- and Ptger4-KO tumor cells exhibited altered T and myeloid cell attractant chemokines, became more susceptible to TNF-α–induced killing, and exhibited reduced adenosine synthesis. In hosts treated with an adenosine deaminase inhibitor, Ptger4-KO tumor cells accumulated adenosine and gave rise to tumors. These studies reveal an unexpected finding — a nonredundant role for the autocrine mPGES-1/PGE2/EP4 signaling axis in pancreatic cancer cells, further nominating mPGES-1 inhibition and EP4 blockade as immune-sensitizing therapy in cancer.
Nune Markosyan, Il-Kyu Kim, Charu Arora, Liz Quinones-Ware, Nikhil Joshi, Noah Cheng, Emma Y. Schechter, John W. Tobias, Joseph E. Hochberg, Emily Corse, Kang Liu, Varenka Rodriguez DiBlasi, Li-Chuan (Eric) Chan, Emer M. Smyth, Garret A. FitzGerald, Ben Z. Stanger, Robert H. Vonderheide
Usage data is cumulative from July 2025 through July 2026.
| Usage | JCI | PMC |
|---|---|---|
| Text version | 2,525 | 308 |
| 298 | 74 | |
| Figure | 603 | 0 |
| Supplemental data | 372 | 23 |
| Citation downloads | 261 | 0 |
| Totals | 4,059 | 405 |
| Total Views | 4,464 | |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.