Aspirin Foundation Meeting 2013

Current research on treating cancer with aspirin

Prof Peter Rothwell - University of Oxford, UK

I was asked to talk about the effects of aspirin on cancer as they appear at the moment. There are really two different areas: one is the short-term effects of aspirin on cancer growth and metastasis and the other, apparently separate, effect is the long-term effect of aspirin on reducing the development of new cancers. So these are separate areas. One of the clues to the possible short-term effect of aspirin on cancer outcomes comes from the trials of aspirin done in the prevention of vascular disease where strangely there’s quite a significant reduction in non-vascular death which, when we looked into it, turned out to be due to fewer cancer deaths in the aspirin group. So that was some evidence that there might be an effect of being on aspirin on cancer growth because these were relatively short-term effects over a few years and all of those cancers would actually have been present at randomisation in those trials, they just weren’t yet diagnosed and they presented a year, two, three years later.

We then took those trials where it was possible to look in more detail and find out more about those cancers, how did they differ between the aspirin and the placebo groups and we found that there was quite a big difference in the risks that the cancers had already spread, so-called metastasis, so blood-borne, spread around the body. That was about 40-50% less common in the aspirin groups than the placebo groups, supporting some earlier work from many years ago in animals that aspirin has a benefit on cancer by preventing the blood-borne spread of cancer cells around the body. There are now several trials ongoing to see whether that effect can be replicated when the aspirin is started after a cancer has been diagnosed and treated rather than before.

So that was the first issue. The second issue is whether aspirin has a longer term effect on the incidence of cancers. The problem with that is that it takes, for many cancers, at least ten years to go from the first abnormalities that aspirin might prevent the development of to actually presenting with a cancer that has grown over the years and is now causing trouble. So you really need very long-term follow-up of trials to get any signal possibly that that might be going on. So we managed to get twenty year follow-up from some big aspirin trials done in the late ‘70s and mid-80s to see whether there was any delayed effect of a few years of aspirin versus placebo on cancer. We found exactly that, that after ten years there was suddenly quite a large reduction in colorectal cancer incidence in the aspirin groups compared to the placebo groups and also for some other GI cancers, in particular oesophagus and stomach and one or two other cancers. That has subsequently been confirmed in long-term follow-up of some independent trials so it’s now looking like a definite effect.

It looks as though platelets, and there has been a long history of work on this, that platelets are important in in some way facilitating either the survival of cancer cells in blood or the transition of the cancer cells from the blood stream to a distant organ. So if that is the case you might well expect an anti-platelet drug that reduces the function, in a way, of platelets, their stickiness, might interfere with that process and there was some animal work from the late 1960s that suggested that was the case but strangely it hadn’t been followed up on in great depth.

Would you advocate the use of aspirin for a high risk, healthy individual?

I think if you had, for example, one of the genetic syndromes like Lynch syndrome which John Burn has been looking at in trials, then my own view is that the evidence in that situation is now fairly compelling that if you are at very high risk of colorectal cancer then the benefits of aspirin would outweigh the risks. I think we’re less good in the general population at predicting colorectal cancer. There are some predictors but I think we probably need to be better at identifying who really is high risk because again if we could do that, and there are some models, then it’s very likely that those individuals probably would benefit from aspirin.

Is the evidence for its effect on other types of cancer becoming clear?

It is becoming clearer for the other GI cancers, so oesophagus and stomach and small bowel. I think we need more data for the common non-GI cancers like breast and lung and prostate where the evidence is less clear at the moment.