The study which we undertook was called the Aspirin Esomeprazole Chemoprevention Trial and the purpose of this was to actually look at the best ways of preventing a very incurable cancer called oesophageal adenocarcinoma.

So what we did is we looked for all the agents that might have an effect in decreasing cancer and there’s at least twenty of them, probably forty now, and we also looked at surgical intervention with surgical ligation of the lower oesophagus.

The two that came out by far the best were acid-suppressing drugs, proton pump inhibitors, and, indeed, low dose aspirin.

So we looked at a two by two factorial design so that we could address these two agents independently because that’s the most efficient way to do that.

Just to clarify, that was high dose, high dose with aspirin, low dose and low dose with aspirin?

That’s correct.

So it would be high dose proton pump inhibitor which is 40mg twice a day, which is right at the high end of what we would normally recommend, and low dose acid suppression which is 20mg per day which is the standard routine dose of proton pump inhibitor.

Then we used 300mg of aspirin versus no aspirin at all so there were effectively four arms in the study, two interventions.

Four arms which would be high dose proton pump inhibitor with aspirin, high dose without aspirin, low dose proton pump inhibitor with aspirin and low dose on its own.

And what was the duration of these treatments?

The duration of the study, the recruitment phase was four years, the follow-up was eight years minimum and a maximum of ten years.

When it came to the results, how are these bearing out?

The results have shown three very interesting features.

First of all, the high dose proton pump inhibitor is statistically very significantly different from low dose proton pump inhibitor at all the endpoints that we looked at which were overall mortality, cancer and high grade dysplasia.

But particularly in preventing the mortality proton pump inhibitors seemed to save some lives compared to low dose proton pump inhibitors.

The reason that’s important is we routinely recommend low dose proton pump inhibitors as the norm.

Secondly, with aspirin the primary endpoint for that was not significant on its own.

However, when we looked at concomitant use of non-steroidal drugs or anti-inflammatories, Brufen would be the commonest, and we factored for when the patients started taking these drugs because that was actually a protocol violation and we knew in real time when these drugs were taken.

So if we looked at the follow-up in these patients and stopped the follow-up and looked at the endpoints prior to them taking the non-steroidal drug we then had a significant effect for aspirin.

What that suggests to us, as we already know, is that non-steroidals, certainly some of them like Brufen, may very well on their own have some weakly cancer protective effect.

But we wouldn’t recommend people take them because they have so many other serious side effects.

Then the last aspect, which was very important, is that all we’ve talked about so far is the benefits for chemoprevention it’s very, very important to actually minimise the risks.

If you’ve got reasonable benefits but with still reasonable risks you can’t use chemoprevention.

What the third aspect of our study was is the risks were exceptionally low indeed.

In fact, 1% of our patients had serious adverse events which is much less than we expected.

So what we’ve got out of this is we’ve got some encouraging data but we can’t give advice to patients individually at the moment.

My recommendations would be as follows, it would be three things.

Number one, we’ve submitted our paper which is for review in a major journal at the moment.

We’ve already submitted that to one of my employers, the National Institute for Health and Care Excellence for them to independently look over our paper and the raw data to actually see whether they think our results are right.

Because it can’t come down to investigators or, indeed, investigating teams; it must come down to advisory bodies which are independent to come to these decisions.

All countries should do the same.

The second aspect about it is that we’d recommend that we follow up the patients for longer.

It’s very important to mention that we’ve looked for patients highly selected, pre-screened for adverse events to aspirin.

We’ve been endoscoping them every two years to look for problems so we’ve picked up problems very early on and we have picked them up relatively young, so round about the early fifties.

Once you start moving into your sixties, the side effects from low dose aspirin increase linearly and in some cases exponentially.

So it could be if you start taking your aspirin from your mid-sixties onwards we won’t see all the low risks that we’re seeing at the moment. So it’s very important to be cautious.

Last, I would say if you are the routine patient out there, you don’t know whether you’ve got Barrett’s, you have got Barrett’s and you don’t know what you should do, I would recommend you speak to your physician about it because in fact we’re very strongly encouraging patients not to self-medicate.

This is a calculated risk.

In fact, the big problem with chemoprevention, everybody is looking for low-hanging fruit, the magic bullet that can make people live longer.

I’m sad to say that isn’t the case so what you have to do is understand the risks that you are taking when you take a drug, particularly over the medium to long-term.

I think what we can conclude is that over the medium term that risk looks beneficial but we don’t know that if we’re to push it on for another five years whether that risk would actually change the risk benefit ratio completely.