Statistics from Altmetric.com
See article on page 822
A series of remarkable clinical, epidemiological, and laboratory studies has shown that aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) such as sulindac, can prevent the development of colorectal cancer and cause regression of pre-existing adenomas.1 Aspirin is one of the most potent chemopreventive agents against colorectal cancer. However, its mechanism for causing a reduction in colorectal cancer risk and mortality is unknown.2 The anti-inflammatory properties of NSAIDs are thought to reside in their ability to inhibit prostaglandin production at sites of inflammation.
Cyclooxygenase (COX) is a key enzyme in the production of prostaglandins, and aspirin can inhibit its activity directly. Initially, COX was thought to be a single enzyme, but in 1991 a second COX isoform was discovered (COX-2) which is induced by cytokines, growth factors, and tumour promoters.3 COX-1 is expressed constitutively in many tissues and is inhibited by aspirin and other commonly used NSAIDs. Contrastingly, COX-2 can be induced and expressed at high levels in most colorectal cancers.4-6 Both isoforms catalyse the conversion of arachidonic acid to a variety of eicosanoid products, including prostaglandins. Inhibition of COX-1 by NSAIDs is a major cause of NSAID associated gastropathy.7However, COX-2 is undetectable or barely expressed in the normal stomach and its inhibition does not cause peptic ulceration.8 9 Recently, a number of highly specific COX-2 inhibitors has been developed which promise to prevent colorectal cancer without incurring serious side effects in the rest of the gastrointestinal tract.10 11
COX-2 plays a major role in the development of adenomas.Min and APCΔ716 mice, which have mutations in the APC gene analogous to those in patients with familial adenomatous polyposis (FAP), have significantly fewer adenomas if COX-2 is inhibited pharmacologically or removed by genetic manipulation.12 Furthermore, COX-2 contributes to the malignant phenotype as it promotes tissue invasion, angiogenesis, and may even create areas of immune privilege within the tumour.13-15
How does aspirin prevent the development of adenomas and cause the regression of pre-existing adenomas? In a landmark randomised controlled trial, Giardiello and colleagues have already shown that sulindac can cause the regression of pre-existing colonic adenomas in patients with familial adenomatous polyposis.16 In their current paper (see page 822) the authors confirmed that sulindac decreases polyp numbers in patients with FAP. They then evaluated apoptosis along the long axis of the crypt in biopsy samples of normal rectal mucosa from patients treated with sulindac for three months, finding that sulindac increases apoptosis at the surface of the crypt, but decreases apoptosis at the base. These changes in the distribution of apoptosis cannot be explained by changes in the expression of genes known to regulate apoptosis—for example, Bcl-2, Bax, p53, or the cell cycle inhibitor p21. Colonic epithelial cells arise from stem cells located at the base of the crypt which then migrate up to the intercrypt table at the surface, where they are shed. There is a constant low rate of apoptosis at the base of the crypt which may regulate the number of cells migrating towards the surface. Apoptosis may also contribute to the shedding process at the top of the crypt. However, it is unclear whether apoptosis is the regulated event which causes cell shedding or whether it is merely secondary to cleavage of cellular integrins from their ligands in the extracellular matrix. The significance of the change in distribution of apoptosis from the base of the crypt towards the surface during treatment with sulindac is unclear and further studies will be needed. However, Kelleret al suggest that this change in ratio of apoptosis between the crypt base and surface may be a useful intermediate biomarker for studying the efficacy of chemopreventive agents of colorectal tumorigenesis. This is a bold claim as the accurate assessment of apoptosis in human colonic biopsy samples is notoriously hazardous. The authors report that the mean number of crypts counted in each patient was 12.74. Apoptosis is so heterogeneous that to obtain reproducible data in the mouse colon, at least 200 perfectly orientated half crypt sections must be quantified. Nevertheless, the concept of NSAIDs increasing cell shedding at the mucosal surface is an appealing one and should stimulate future studies aimed at validating its usefulness as a biomarker in humans.
See article on page 822
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.