The anti-angiogenic potential of targeting Prostaglandin E2 receptors in tumours

Tumour hypoxia refers to the gradual decrease in partial pressure of oxygen as tumours grow. Cells more than 1-2 cm distant from blood vessels therefore have reduced access to oxygen and nutrients. Tumour cells, however, display a coordinated response to hypoxia and release a number of factors which are involved in processes such as metabolic reprogramming and angiogenesis which promote their growth further and provide a route for metastasis.

Angiogenesis can broadly be defined as the growth of blood vessels from an existing vascular bed, which is dependent on endothelial cell proliferation, migration and tube formation. In healthy individuals endothelial cells are normally quiescent and only undergo proliferation in a highly regulated manner during processes such as wound healing and repair of the uterine lining. This quiescence is due to the presence of a broad group of endogenous anti-angiogenic factors which override pro-angiogenic stimuli. In tumours, however, the balance is shifted in favour of pro-angiogenic factors which mobilise endothelial cells; a phenomenon referred to as the 'angiogenic switch'.

Tumour cells are known to release a number pro-angiogenic factors during hypoxia, the best characterised of which is VEGF. There is considerable evidence, however, that prostaglandin E2 (PGE2) is also upregulated during hypoxia in a number of solid tumours, and this in itself is a potent angiogenic factor. PGE2 is synthesised downstream of COX-2, an enzyme which is well recognised therapeutic target for inflammation. While COX-2 inhibition has displayed success in preventing tumour growth, its prolonged inhibition in vivo is associated with a number of complications that are discussed in more detail below.


The mechanisms via which PGE2 promotes angiogenesis are not well understood. This is an area I will be researching from April to August as part of my Master's dissertation. In the meantime, I am conducting a literature review to consolidate the current understanding of PGE2 in angiogenesis and the tumour microenvironment.

PGE2 binds to a series of cognate G-protein coupled receptors termed EP1-4. These receptors are coupled to diverse intracellular signalling cascades and display heterogeneous expression across different tissues. While endothelial cells express all four receptors, the expression pattern on other cells within the tumour microenvironment, including cancer associated fibroblasts (CAFs), myeloid-derived suppressor cells (MDSCs) and tumour associated macrophages (TAMs) remains to be elucidated. It is known, however, that  tumour-derived PGE2 can promote angiogenesis directly through endothelial cells, but can also influence stromal cells which themselves release pro-angiogenic molecules. I have attempted to clarify this complex relationship in the diagram below:

It is clear PGE2 is a key player at a number of steps. It can polarize TAMs to an M2 phenotype to promote inflammation and release pro-angiogenic factors; it is involved in induction of MDSCs, which secrete a number of immunosuppressive and pro-angiogenic factors and can also act on CAFs. Moreover, PGE2 and VEGF may promote endothelial cell mobilisation via an autocrine mechanism.

Non-steroidal antiflammatory drugs (NSAIDs) have long been recognised to attenuate inflammation and reduce cancer risk through COX-1 and COX-2 inhibition. Prolonged use of NSAIDs is however associated with adverse events, most common of which are gastric ulcers, owing to the important role of COX-1- derived prostanoids in homeostasis across different tissues. The advent of COX-2 specific inhibitors such as celecoxib attempted to rectify this problem. While initially successful, a number of longitudinal studies have linked high long-term doses of these compounds to adverse cardiovascular events. This may reflect reduced biosynthesis of COX-2 derived anti-thrombotic PGI2 with sustained production of COX-1-derived pro-thrombotic Thromboxane A2. More specific compounds which modulate levels of individual prostanoids are clearly needed.

The focus of my research over the four months will be to understand which EP receptors are important to angiogenesis and the intracellular cascades mediating their effects. Targeting the terminal prostaglandin E2 synthase enzymes  may be a viable strategy to specifically impair angiogenesis while leaving levels of other prostanoids unaltered. Similarly, specific targeting of EP receptors on tumour and stromal cells may one day prove useful as monotherapy or in conjunction with current anti-angiogenic therapies such as bevacizumab.