What are all the things that aspirin does?

BMJ 2003; 327 doi: http://dx.doi.org/10.1136/bmj.327.7415.572 (Published 11 September 2003) Cite this as: BMJ 2003;327:572
  1. Rod Flower, professor of biochemical pharmacology
  1. William Harvey Research Institute, London EC1M 6BQ

    This fascinating but simple and cheap drug has an assured future

    Ask any medical student and he or she will tell you that aspirin reduces fever, pain, and inflammation but may cause ulcers. Students may also recollect that it prolongs bleeding, and may prevent strokes and heart attacks, but would be unlikely to know of its use in cancer or Alzheimer's disease.

    A defining point in the history of aspirin was the discovery that it inhibited the prostaglandin forming cyclooxygenase.1 Prostaglandins cause inflammation, fever, and pain; have gastric cytoprotective actions; and are implicated in platelet aggregation, so this discovery provided a unified explanation for the effects of aspirin (and most other non-steroidal anti-inflammatory drugs). However, events took an even more interesting turn when a further isoform of cyclo-oxygenase, cyclooxygenase-2, was discovered.2 While similar in many ways to the original enzyme (COX 1) there were important differences, including the fact that COX 2 was induced in cells by inflammatory insults. COX 2 therefore seemed to be the most relevant target in inflammation, which led to the notion that the constitutive COX 1 generated prostaglandins required to maintain physiological functions (such as protection of the gastric mucosa, platelet aggregation) whereas COX 2 generated pro-inflammatory mediators.3 Aspirin inhibited both isoforms, as did most non-steroidal anti-inflammatory drugs, perhaps explaining why these compounds were not only effective therapeutically but also had characteristic side effects.

    The ensuing search by the pharmaceutical industry for selective COX 2 inhibitors culminated in the recent introduction of new, safer anti-inflammatory drugs as well as the rediscovery of older drugs that had COX 2 selective actions. But, as aspirin inhibits both isoforms, why does it continue to be used and why is there continuing interest in its pharmacology?

    The answer to the first part of this question is partly down to aspirin's unique mechanism of action that inhibits both COX 1 and COX 2 irreversibly. The effects of this are evident in platelets where cyclo-oxygenase cannot be replaced, explaining why a single aspirin can depress platelet aggregation for many days. The half life of aspirin in plasma is short; esterases remove the acetyl group leaving free salicylate, which may have a secondary pharmacological effect through cyclooxygenase inhibition or other mechanism, adding to the complexity of aspirin's action.

    The current interest in aspirin stems from the fact that many animal experiments and human epidemio-logical studies now link aspirin (and other non-steroidal anti-inflammatory drugs) with beneficial effects in various cancers, including breast, ovarian, oesophageal, and colorectal cancer. Recent meta-analyses supported the idea that the overall relative risk of colorectal cancer is reduced in people taking long term aspirin.4 Another meta-analysis of observational data confirmed a protective effect in oesophageal cancer and provided evidence of a relation with dose and duration of treatment, and other studies showed a beneficial effect in ovarian cancer.4 5 How aspirin or other non-steroidal anti-inflammatory drugs produce this effect is not entirely clear, but the synthesis or activity of COX 2 is increased in many tumours, and inhibition could activate apoptotic mechanisms or suppress angiogenesis.6 It has even been suggested that the link between diet and the prevention of colorectal cancer is attributable to the presence of salicylic acid in plant and vegetable foodstuffs.7

    Evidence from longitudinal studies of long term users of non-steroidal anti-inflammatory drugs originally pointed to a reduced risk of Alzheimer's disease,8 and these findings are supported by other, more recent data,9 where an inverse relation was found between taking aspirin (and other non-steroidal anti-inflammatory drugs) and Alzheimer's disease, but not other forms of dementia. The mechanism is uncertain—Alzheimer's has an inflammatory component and therefore COX 2 may be the target, although other mechanisms have been suggested.10

    Two questions bedevil what is otherwise an exciting therapeutic prospect. What is the minimum dose required to achieve these effects, and how can we assess the relative risk and benefit of a preventive treatment that will entail treating healthy people for many years with a drug known to have gastric and other side effects? It is here that aspirin's grandchildren may have a role. COX 2 seems to be the main culprit in both cancer and Alzheimer's, so the selective COX 2 inhibitors, which have reduced gastric side effects, are natural choices for such long term prophylactic treatment.

    What of the future of aspirin itself? Because of its profound effects on platelets it is unlikely to be supplanted as a cheap and effective prophylactic treatment for those patients at risk from excessive platelet aggregation, but in view of its venerable history, it is surprising that aspirin is still the subject of ongoing medicinal chemistry effort. Attaching a nitric oxide donor to the molecule seems to ameliorate the side effects of the drug while boosting its therapeutic effects.11 The discovery of a third form of cyclooxygenase,12 mainly confined to the central nervous system and heart, which is also inhibited by aspirin, will no doubt provide yet another twist to the continuing story of this fascinating but simple drug.


    • Competing interests RF is on the scientific advisory board of Nicox Spa and has received consulting fees and research support from this company. He has appeared as an expert witness for Searle-Monsanto, manufacturer of Celebrex.