Screening for Breast Cancer

Downloaded from www.bbc.co.uk/radio4 

THE ATTACHED TRANSCRIPT WAS TYPED FROM A RECORDING AND NOT COPIED FROM AN ORIGINAL SCRIPT.  BECAUSE OF THE RISK OF MISHEARING AND THE DIFFICULTY IN SOME CASES OF IDENTIFYING INDIVIDUAL SPEAKERS, THE BBC CANNOT VOUCH FOR ITS COMPLETE ACCURACY.

INSIDE HEALTH

Programme 3.

TX: 15.07.14  2100-2130

PRESENTER:  MARK PORTER

PRODUCER:  ERIKA WRIGHT

Porter

Today’s programme is all about screening for breast cancer.  Switzerland looks set to become the first country in Europe to halt routine screening. After reviewing the evidence the Swiss Medical Board felt the benefits were not great enough to justify continuing it.

Yet here in the UK a similarly comprehensive review – led by Sir Michael Marmot and looking at the same evidence – came to the opposite conclusion. Namely that the UK’s NHS Breast Cancer Screening Programme, offering women between the ages of 50 and 70 a mammogram every three years, should continue.

So, how can two groups of experts arrive at such different conclusions?  If you are confused, rest assured you are not alone.

Susan Bewley works in women’s health, she’s Professor of Complex Obstetrics at King’s College London, and being in her mid-fifties is eligible for screening but has decided not to take up her invitation.

Bewley

Like many of your listeners I’ve got friends and relatives with breast cancer and I’m frightened of breast cancer but I’ve looked at the figures and I don’t think they’re as good as people make out.  Everyone must make her own decision based on her preferences and her values but we all need the same good quality information and I’ve looked at that and I don’t think all women are getting it.

Porter

What have you seen that our listeners might not have?

Bewley

I’ve been following the debates and I’ve been reading the original articles, I read the Marmot Review and I’ve made the calculations.  It’s very painful to realise that all the women who’ve been told they’ve got cancer through screening they believe their lives have been saved but the truth of the matter is that probably only one in 15 of those women that lives have been saved, most will still live anyway and a few sadly will still die.  But a few have been told they’ve got breast cancer and have had treatments – that might be mastectomy, radiotherapy or chemotherapy.

Porter

And the problem in your case was you presumably don’t know, like all women, which of those 15 you were going to be?

Bewley

You can’t know which of those 15 you’re going to be.  For me with my preferences and my values looking at the figures there is a very small chance of my life being saved and I don’t even know if my life would be saved, I might not die of breast cancer but I might die of something else.

Porter

Of course one of the problems is that we doctors don’t know which of those 15 women either, so we find a cancer we then treat it.

Bewley

It’s true and that has to be done but I think that if you compare the prostate leaflets that men get, they are in black and white and they spell out that there may be risks, particularly there may be a risk of harm and impotence.  If you look at the leaflets for women they’re slightly pinkier and fluffier and they don’t give the full facts, they don’t warn women that there are harms if they follow this path.  And we’re trying to catch something that may or may not be serious.  There’s a very good analogy an American has made of trying to herd tortoises, rabbits and gazelles.  If you put a fence around them there really wasn’t much point catching tortoises, cancers that were very slow growing and may never have turned into anything harmful.  In between screening occasionally very aggressive cancers come through and we can’t catch the gazelles.  So we only catch a rabbit or two and yet some of the people who are doing screening are making out all cancers are the same.  We doctors know that when you look down a microscope something can look like cancer but it may not behave like cancer.

Porter

So do you believe that historically we’ve overblown the benefits and perhaps underestimated the risks?

Bewley

I do, I’ve looked back even at the original Forest Report which led to the introduction of screening and there were dissenters at the time who said it didn’t fulfil the screening criteria and that it may even have been political motivated.

Porter

You’ve gone very public with your stance, what sort of response have you had?

Bewley

Mostly I’ve had women doctors saying me too.

Porter

Professor Susan Bewley who is in the minority – around three quarters of women in the UK do take up their invitation to be screened - nearly two million every year in England alone.

The programme was introduced in the late eighties when Edwina Currie was Junior Health Minister.

Currie

Breast cancer in the 1980s and previously was still something women didn’t talk about except in hushed whispers to each other.  And when I looked at it it was the main cancer that was killing women and particularly women of working age, in fact that’s how we sold it to Margaret Thatcher – we said to her, you know what, we’ll get them back to work and they’ll be paying their income tax!  So it’s not only in practical terms of very good thing to do, in political terms it works as well and particularly if there’s a group that feel they’ve been perhaps a bit neglected and pushed to the back.  Women were still very much second class citizens in the ‘70s and the ‘80s and here was something that was being done just for them.

Porter

But do women still think the same today? And what factors should they consider before taking up the offer of screening?

We have assembled a group of experts with differing perspectives to debate the issue:

Sir Michael Marmot is Professor of Epidemiology and Public Health at University College in London and the man who led the Independent Breast screening Review in 2012 that found in favour of continuing the programme here in the UK.

Dr Caitlin Palframan is Head of policy at the charity Breakthrough Breast Cancer.

Inside Health’s Dr Margaret McCartney has successfully campaigned for women to be given clearer information about the possible downsides of screening in the leaflet enclosed with invitations.

And Professor Nikola Biller-Andorno is an ethicist from the University of Zurich, and a member of the advisory group that recommended screening be discontinued in Switzerland. Nikola, how did you arrive at your decision?

Biller-Andorno

The medical board looked at the existing literature, the same literature that was looked at by different countries, in the UK, in the Netherlands, etc., and it found that mammography screening had no effect on overall mortality – that is the overall number of people dying from any given cause – so no change caused by mammography screening.  And it found that mammography screening might prevent – it’s not quite sure but it might prevent about one death attributable to breast cancer for one thousand women that you screen.  But this effect was not sure and it was doubted in the most recent literature and what the medical board appreciated was that this was gain, so to speak, comes at a significant cost and that is about for each woman that you may save you have at least a hundred women that are bothered by false positives or that they even fall prey of over-diagnosis, which means that they will actually undergo treatment of a cancer that would have never bothered them, they may undergo radiotherapy, they may undergo chemotherapy, an operation for something which would never have bothered them.  And in light of this evidence that was available the Swiss Medical Board concluded that there was no basis really to introduce new programmes in Switzerland at this point and that the existing programme should be time limited and possibly phased out unless we get new evidence that suggests a more robust benefit of the programme. 

Now I am on the expert panel of the Swiss Medical Board as an ethicist and I do share the conclusions of the board because I think so far we’ve tended to overestimate the benefit and tended to neglect the real potential for harm.  I think for any woman who thinks she’s got breast cancer she’ll have to undergo partly aggressive therapy, she might think am I going to see my kids grow up – all these kind of things that are triggered.  I think if we can’t distinguish well between the woman who might profit from undergoing through all of this, whose lives will be saved and other women who really bothered in vain, unless we can really distinguish well between those two groups I think that’s a real problem for systematic screening.

Porter

Michael Marmot, you looked at similar data to the Swiss Medical Board when you carried out a recent review here in the UK and yet that review came to a different decision.

Marmot

Well we looked at all the questions that Nikola has just addressed and we did indeed come to a different decision.  Nikola quoted an absolute difference and she said that you need to screen a thousand women to save one death.  We said it was five times as beneficial as that – our calculation suggested you need to screen 180 women to avert one death.

Porter

That’s a big difference.

Marmot

It’s a huge difference.  I mean let me just say at once that I came to this, not as a screening expert, in fact I came to it with no priors at all about whether breast screening works or doesn’t work and looked in as dispassionate a way as possible at the nature of the evidence.  Now it turns out some people say you need to screen 2,000 women to save one death, some say you need to screen 100 women to save one death.  The two biggest sources of variation, apart from what’s going on in the head of the writer, are the age of women, so that, for example, the writer who said you need to screen 2,000 women to save one death was looking at younger women as well as older women.  Now if the absolute rate of breast cancer is low as it is in women under 50 then you’ll need to screen many more women to save one life.  The person who said you needed to screen 2,000 women was looking at trials that included younger women.

Porter

But are you confident, looking at the age groups that we screen here in the UK, the current programme, that your figure of less than 200 women need to be screened to save one death is accurate?

Marmot

I think it is as accurate as we can get given all the usual caveats and one of those important caveats is that these trials were done a long time ago and some of the experts who gave this evidence said we cannot base current policy on trials that were done one decade, two decades, even three decades ago.  And we considered that.  The problem is we don’t have more recent data in the form of trials, we do have more recent data in the form of observational studies so these are not randomly allocating women to being screened or not screened but observing cohorts of women who’ve been screened and cohorts that haven’t and the apparent benefit of screening is bigger.  Now we thought that there were methodological problems with these observational studies so we didn’t use the figures they gave us.

Porter

Caitlin Palframan, this is complicated and it’s going to be confusing for women to work out whether they should sign up to the National Breast Screening Programme, what’s your stance at Breakthrough?

Palframan

I think you’re right this is an incredibly complicated debate and it’s a shame in a way that it’s played out so publically in the media because it is very confusing and we know that from speaking to the women that we work with.  Our view at Breakthrough Breast Cancer is that the review that Michael Marmot is talking about is the most relevant to the UK programme but there are risks and benefits to the programme, so the primary risk being the one that’s spoken about the most over diagnosis or we prefer to call it over treatment where you detect a cancer that is a real cancer but would not have caused any symptoms or been life threatening in that woman’s lifetime, so they receive treatment that they wouldn’t have needed if they hadn’t gone for screening.  So we believe it’s really important that women have all the facts in front of them and decide whether or not to attend.  But overall we definitely support the programme and we believe that it provides benefit.

Porter

Margaret McCartney, listening to Michael and Nikola, two comprehensive reviews there that have come to a different decision.

McCartney

To me the big thing that’s coming out of the discussion so far is how much uncertainty there is and actually in the Marmot Report, itself, when it talks about the numbers around breast cancer screening it does say the figures quoted give a spurious impression of accuracy.  So really we’re using a lot of numbers but there’s a lot of uncertainties attached to them, a lot of different reviewers have come out with really quite different numbers.  But I think it is generally agreed that breast cancer screening will pick up some breast cancers usefully but will also pick up some breast cancers unusefully.

Porter

So let’s try and simplify this, to all of you we agree that screening for breast cancers can save lives from breast cancer, can prevent deaths from breast cancer but what we’re uncertain about is at what cost that might come.  What’s the downside, Margaret, of entering a screening programme?

McCartney

Yeah well the problem is because the treatments for breast cancer can incur some harms, that means that you may die of breast cancer treatment rather than the breast cancer itself.  Now obviously if that treatment was saving your life that’s a treatment you would like to have but supposing it’s a treatment that you didn’t need and you didn’t benefit from because your breast cancer was over diagnosed, that means you get the risks of the treatment without the benefits of the treatment.

Porter

But this is an alien concept to most listeners – how can you over diagnose a breast cancer?  Surely the earlier you catch it the better?

McCartney

Absolutely and this is one of the real issues with screening is that it is counterintuitive.  So if you go and you do a CT scan on a lot of people that are walking down a street just now you will find things that you didn’t expect, things they didn’t get symptoms from.  And we know from post-mortem samples of breast tissue around about 9% of women will have breast changes called DCIS – ductal carcinoma in situ – but these women did not live knowing that there this cancer and they died from something completely different.  So the problem is that you can pick up a lot of disorders or diseases, things that appear to be disorders or diseases but they don’t behave like disorders, they exist quite benignly within that person and they don’t go on to cause problems.  But if you pick them up from a screening programme you cannot predict for that person what will happen and that woman is then offered treatment but treatment that she may not have benefited from.

Porter

Michael?

Marmot

Yes I think what Margaret said is incredibly important and I agree completely with the emphasis not just on the benefit but on the harm.  And we spent a lot of time in my review looking at this question of over diagnosis.  The word I think that Nikola used was being treated in vain, I don’t think that’s the right way to think about it.  If you get fire insurance for your house and your house doesn’t catch fire and you say well I don’t need that insurance, that was in vain, if I’d known my house wasn’t going to catch fire why would I have had fire insurance.  The problem is what’s being picked up and mammographic screening is cancer, there are false positives as well but leaving that question aside, is cancer and we’ve got evidence from all the experts – the pathologists, the radiologists, the screening people – and we said surely you must be able to tell which cancer picked up at screening is going to turn out to be benign and would never have come to attention and which one will go on to be invasive and cause death.  And the answer from the experts is no we cannot tell the difference at screening.  So you’re picking up a cancer and the woman doesn’t know and her doctor doesn’t know whether this is the one that would never have come to attention, as Margaret described, or whether it’s the one that’ll go on to kill you.  So what best practice is given that cancer you treat it.  Now in practice – and we looked at this evidence – we were not persuaded that the treatment for breast cancer was killing women, we were not persuaded by that evidence at all and it turns out that the ductal carcinoma in situ, the DCIS, which are largely picked up by screening, they tend not to be the ones that are treated with radiotherapy or with chemotherapy, the evidence shows that they’re the ones that tend to be treated with a partial mastectomy, a lumpectomy, or perhaps mastectomy but not with the extra treatment of radiotherapy or chemotherapy.  Now being treated, if there were some concept that this was unnecessary, is a terrible thing.  On the other hand nobody knows when we find that cancer whether this is the bad one or the one that’s not bad.  And given that what we said in our review women themselves have to be brought into the picture.

Porter

Nikola, what sort of response was there to the Swiss Medical Board’s decision when – I mean effectively they were saying that they felt that it was overestimating the benefits and underestimating the risk – did the public agree?

Biller-Andorno

This is a really complex issue and I think if we want women to make a genuine decision on these things I think we have to be up front with that.  I think it’s not enough in an information flyer to women to say well you might have a benefit and there is some risk of your being over diagnosed, over treated but I think there needs to be a bit of qualitative data, the best data we can put into such a flyer, so that women know it’s not just I’m very likely going to benefit and there’s a small risk of my not benefiting but that it may well be the other way round.  I think many people want to believe that screening works, it’s been so much effort to build these programmes, it’s a wonderful idea to be able to save women dying from breast cancer, so I think there’s a lot of desire to believe that mammography screening works.  And we know that in particular women who participate in screening programmes drastically overestimate the benefit.

Porter

And who can blame them given that even experts seem to struggle to nail down the scale of the benefits. While a five-fold or more variation in estimated benefits may seem a lot, it is actually not that unusual in this type of research.

The figures used are often best estimates extrapolated from complex data – each with a different margin of error. Take the best of one range, and the lowest of another, and it’s all too easy to end up with a very different conclusion.

And how you interpret statistics depends on your inherent bias. Are you heavily invested in screening like the NHS? Or are you concerned that the risks are being underestimated? Such bias exerts a subtle effect, but one even the most scrupulous researchers recognise.

Or, to put it another way, if you torture statistics enough they can often be made to tell you what you want.

Moira Adams is a Trustee of Challenge Breast Cancer Scotland.

Adams

We differ from most of the other charities in as much as we do not actively promote the screening programme.  What we actively promote, as a cancer charity, a breast cancer charity, is informed choice for women.  We believe that women should make an informed choice on screening and should not be coerced into going for screening by paternalistic leaflets from the NHS because the NHS is biased in favour of screening because it’s the NHS that runs the screening programme, so it has to be biased in favour of screening.  No matter how impartial they try to be at the end of the day it’s in their interest for the screening programme to work, they need 85-95% take up for it to be financially viable.  So it has to be biased, so we would not promote NHS screening.

Porter

Caitlin Palframan, that stance is an interesting one, sort of independent stance if you like, so presenting the information and let people make their own decision, do you feel that’s what you should be doing in situations like this?

Palframan

The NHS is an extremely financially constrained organisation, the NHS will not offer something that they don’t feel is cost effective and provides benefit to women.  I completely agree that there needs to be informed choice about screening.  Now I think in our minds what we’d like to focus on is where we go from here, so we know without new trials, which are not going to happen at this point, we will always have some uncertainty about the benefit of the programme and the risk of the programme and what we’d like to see is improvements made that can maximise the benefits while reducing the risks of the programme.  And I think there are two – those changes are two fold that could be made.  One is getting to the bottom of this issue which we’ve discussed around which breast cancers will progress and understand whether that’s going to go on and cause that woman problems or whether she can live with it healthily for the rest of her life, that will make a huge difference and remove this issue of overtreatment.  So that research is ongoing and something that the breast cancer charities are actively looking to fund, it’s so important that we’re able to make those predictions.  The other thing that would be extremely useful is if we can better tailor the programme, so we’re offering screening to those who are most likely to benefit, the ones who are most likely to develop a breast cancer that’s going to cause them harm.

McCartney

The idea that we can separate them out I think is something that is no way verified in current scientific knowledge and I don’t really see much progress that’s been made on that before now or even from now.  To me the big problem is that the NHS programme is judged on how many women attend, not on how many women made a good informed choice about whether they want to attend or not.  So in many ways there’s a kind of value judgement made in that screening is better for the NHS through breast screening and that creates I think the bias that Moira, speaking earlier, was talking about.  So I think we have to separate that out and help women to make a good informed choice without telling them what to do or sending them a leaflet that doesn’t contain all the information they would need with a time and a date to turn up.

Porter

But there has been a change in the information given to women recently in certainly England and Wales, I don’t know about Scotland.

McCartney

Yeah, no there has been but certainly still it says we want you to make an informed choice at the beginning but then I would argue that a lot of the information isn’t really quite complete and doesn’t explain that actually you’ve got a much higher chance of being over diagnosed with a breast cancer at screening as you are to have your death from breast cancer prevented.

Porter

Michael?

Marmot

I think the issue of uncertainty is being overplayed.  I deal with a variety of issues in public health and I would say there’s more evidence relating to screening for breast cancer than in almost any other public health issues with which I have to wrestle on a daily basis.  There’s an enormous amount of evidence.  It is always the case that trials done yesterday are being applied to women and men tomorrow, so that’s not unique to this problem.  Secondly…

Porter

But I suppose the problem here it’s not actually yesterday though Michael in terms of when we’re talking decades ago and there have been significant advances…

Marmot

But that is what we have, simply to throw your hands up and say they’re old trials, everything could have changed, well then what?  So the second issue is well treatment may have improved and that argument was certainly put to the panel that I chaired and we looked at it.  But the argument was put both ways because treatments improved that might make screening more effective or it might make screening unnecessary.  And overwhelmingly – overwhelmingly the reduction in mortality given a breast cancer incident that has occurred has been because of improvements in treatment, no question about it.  The benefit of screening pales by comparison with the benefit of improvements in treatment.  But that does not mean that we should say well there’s uncertainty and the screening’s only at the margin therefore we throw up our hands.  We said very clearly in my review that the individual woman’s decision is absolutely central and we said should be at the heart of a modern healthcare system.

McCartney

But Michael the reason why some women have their lives saved by breast cancer screening is because other women are over treated by it, so the more women that enter the programme you’re making a value judgement that it’s worthwhile for some women to have their death delayed from breast cancer but to do that other women have to be over treated.  So you have to be absolutely honest with those women and say that this is a balance of risks.

Marmot

And we said it’s absolutely of vital importance to communicate that information to women, as I believe it is being communicated.  So we’re saying it is complicated but we can decide things and we know that there is a benefit and we know that there is a harm.  I’m not for one moment suggesting that we should downplay the hazard of over diagnosis, it’s very real, it exists, we documented it, we reported it and it must be put alongside the benefit.

Porter

Caitlin.

Palframan

The information that’s provided on breast screening has had more attention paid to it and in the construction of it than virtually any other health information that’s out there.  So the information that’s come out from the NHS programme was developed with risk perception experts, with cancer charities, with patients, with critics, with supporters of the programme – all those people contributed to it.  So it may not be perfect but I do believe it’s about as good as it’s going to get.

Porter

Nikola, I suppose the one question that women want answered is am I going to be the woman whose life is saved or the woman who has to undergo unnecessary treatment and we can’t answer that?

Biller-Andorno

Right, we put up so easily with the fact that when we’ve got a really old trials, we’re talking trials that have been conducted 50, 40, 30 years ago and so many things have changed since then in terms of treatment etc.  So I’m wondering why can’t we do a randomised control trial these days that would be designed in a way that would respect women’s choice, there might be women who might be quite willing to be randomised into a group that either receives screening or does not receive screening, for instance.  So – and when I raised this idea I frequently get frowns from people saying well that’s obviously unethical, I’m not sure at all that it’s unethical and I think we should give more thought than we have to actually just designing another – a new study that would help us answer the questions that we have such a hard time finding.

McCartney

Nikola, I have to say that’s the only circumstance under which I would have breast cancer screening would be as part of a randomised control trial.

Porter

But Michael you suggested that we’re never going to see that trial.

Marmot

Well no there is such a trial in England at the moment which is extending screening two or three years younger than 50 and three or four years older than 70.  So it’s not a trial of women in the current screening age but it is looking at one extra screening before women get to 50 and an extra screening after women reach 70.  It will give us a more contemporary estimate for this group that’s slightly younger and slightly older than the current screening age of 50-70, it will give us an estimate contemporaneously of the benefit of screening, so that is actually going on as we speak.

McCartney

I think we’ve got a real hesitancy about allowing women a real free choice, including the choice not to have screening.  And the issue for me is not to think about whether we’re putting women off or whether we encourage them to have it, it’s to help people make really good informed decisions.  And that decision what it will be will be, you should be allowed to choose not to have it, just as you can be allowed to choose at present times to have the screening.

Porter

Caitlin.

Palframan

My personal view is that absolutely I would 100% go for screening but I wouldn’t want that to sway anyone else, it’s very important that they make their own choice based on the risks and benefits.

Porter

Nikola, as an ethicist, where do you stand – would you put yourself forward for screening if you were invited?

Biller-Andorno

As an ethicist I wouldn’t want to volunteer my position because I would want to encourage women to think for themselves and to really exhaust the potential of their making an informed autonomous choice, I think that’s most important.

Ends