This is a rather beautiful study linking mouse model work and cancer cell lines with their colorectal cancer biobank, making the link between inflammatory colorectal carcinogenesis and how it might be mediated by anti-inflammatory drugs.

We know that inflammation is one of the hallmarks of cancer. To me, that opens up a whole new world of therapeutic potential. We have been dabbling with aspirin for decades, and we have been looking at the newer generations of COX-2 inhibitors. However, Dubois and colleagues, who have made a very distinguished contribution to the understanding of the molecular events that link inflammation together, have given us some interesting new insights.

The study showed that prostaglandin E₂ (PGE₂), which is one of the key lipid mediators of inflammation, is linked to progression of colorectal cancer through methylation. Very simply, what they seem to have shown is that PGE₂ induces the synthesis of DNA methyltransferase enzymes. These are then responsible for methylating and transcriptionally silencing some tumor suppressor genes, and that could lead to progression. If you switch the tumor suppressor off, that could lead to tumor progression.

In a series of rather elegant and quite compelling experiments, they have shown the link. They have taken it from cell lines to elegant mouse model systems and made some correlations with human tumors. They then went on to show in the murine tumor models that, if we use celecoxib, a selective important inhibitor of cyclooxygenase-2, you can reduce the effect very markedly.

If you use azacitidine, which is a demethylating agent, you can also reduce this effect markedly. If you put the 2 drugs together, however, they seem to be truly pharmacologically synergistic. So, we have some fascinating insights into the mechanism of colorectal carcinogenesis. A really nice molecular study of what pathways might be involved sequentially. For those of us who are translational clinicians, as Dubois and colleagues pointed out in the article, it may give us some potential new therapeutic options.

I'm very keen on exploring new literature. It looks as if we might be able to identify patients with inflammatory tumors (ie, tumors in which inflammation contributes to the aggressive nature of the cancer and a worse phenotype) just by looking at some simple markers of inflammation. C-reactive protein may identify a subgroup of patients in whom the tumor is driven by this inflammatory process. A fantastic contribution has been made by Professor Donald McMillan, a friend of mine working at the University of Glasgow.

So again, if we apply this to personalized medicine, we might be able to identify patients in whom inflammation is important and patients in whom we might wish to use such drugs as celecoxib and aspirin, and potentially in combination with azacitidine with demethylating agents, to see what we can do to suppress inflammation. Through this, we might see what can be done to alter the natural history of established cancer and, perhaps in the future, see what we might be able to do to prevent colorectal cancer.

These were some beautiful laboratory insights. Dubois is contributing at the top of his game. I can see how these ideas could be taken into the clinic to test the hypotheses pretty quickly. Thanks for listening, and as always, we would be delighted to receive any of your comments or feedback. Thank you.

# posted by roodbont @ 1:12 AM