Recurrent colorectal cancer and questions raised by surgical intervention

Share :
Published: 11 Nov 2014
Views: 2493
Prof Tom Treasure - University College London, London, UK

Prof Treasure talks to ecancertv at NCRI 2014 about recurrent colorectal cancer which has spread to the lungs or liver and the questions raised over the efficacy of current treatment practices involving surgery, which may or may not be beneficial.

Click here for more information about the trial.

When colorectal cancer recurs, as it does in about half of patients who we hope to have cured, it can spread to the liver and to the lungs. For years now people have opportunistically taken out metastases from the lungs in the hope that it might help and that has become more common. Right from the earliest days there has been serious doubt about whether that really is an effective treatment and cures anybody. Clearly if you were lucky enough to catch just the one metastasis you might complete a cure but that’s not in the nature of blood borne disease. So it’s carried out in hope rather than expectation.

Because they’re very selective about who they operate on, when I say they I mean we as thoracic surgeons, it’s patients with one, two or three and they only show up after some time. But they are a very selective group of patients who might be in the most indolent end of the distribution. So the fact that some of them live for a number of years afterwards might be the selection of the more indolent cases, rather than the effectiveness of the surgery.

So what we decided to do some years ago was to try and subject that to a randomised trial and the acronym is Pulmonary Metastasectomy in Colorectal Cancer, PulMICC, and the design is this in essence. Because the majority of patients who have lung metastases would not be and are not considered for surgery, you will say no to some patients, patients who have just got one or two and they seem completely well otherwise and there’s a chance of curing them, surgeons are invited to take them out and they willingly do so. But if you say no to some and yes to others across a rather complex spectrum of disease, it stands to reason, it seems to me, that there are some where there is doubt between those two and indeed there are because when you sit in the multidisciplinary teams, the discussion instead of being a quick yes or no can become very protracted and then patients are brought back after another few weeks and so on. So these are the patients in whom we propose there is sufficient, perfectly evident uncertainty to put it to the patients so that they should be randomised.

We started the study three or four years ago as a feasibility study funded by Cancer Research UK. This model is that we recruit all patients who might be a candidate, and we’ve recruited 300 , and of them randomise those where the clinical team and the patient accept that there is sufficient uncertainty about what is the better course of action for them to allocate them at random and there are about 70. Now we’ve shown that it’s feasible to randomise but it’s a difficult thing to do, to randomise patients to have or not have rather dramatic treatment. It’s not like two different doses of a drug or two drugs very much like each other. So that’s recognised as being difficult but just because it’s difficult doesn’t mean that we don’t need to find out the answer.

When do you hope to find out?

The whole thing rather hinges at the moment on the committee CTAAC which will decide in a few weeks’ time whether we get funding to carry on the study for another three years. We would then need a couple of years after that to let time go by to see survival and non-survival emerging as the main outcome of interest. We have twenty centres nominally open in Britain of whom about a dozen have produced cases in enough numbers to matter. But also we now have, because we’re writing about it and talking about it and doing exercises such as this, interest from Germany, Italy, Spain, perhaps America but that’s a difficult one, Serbia and China. So if some or all of those come on board we could get good numbers and maybe a decent sized trial within the time frame.

What do you think will be the outcome of this study?

There’s a huge amount of data already in observational studies. I quite deliberately make no attempt to look under the bonnet; not everybody is able to do that but I can, it just goes along. What I can say is because of the nature of the patients who are being operated on, there will be a fair number alive at five years amongst those who are selected to have surgery. What has not been done before is to see in exactly the same group of patients randomised to not have it, whether they will also survive similarly. We have done various mathematical modelling studies where we take cancer registry patients who have not had this surgery and I’d have to say they survive quite well too. So if I were to predict I would say very little difference between the two but clearly those who are practising this surgery at present believe that there is a difference so we need to find out.