Why I am still a screening sceptic
Essay: A leading cancer experts explains why, despite the ‘Jade Goody effect’, he has severe doubts about the benefits of cancer screening.
‘The largest threat posed by American medicine is that more and more of us are being drawn into the system not because of an epidemic of disease, but because of an epidemic of diagnoses. The real problem with the epidemic of diagnoses is that it leads to an epidemic of treatments. Not all treatments have important benefits, but almost all can have harms.’ (1)
A couple of weeks ago, I was showing off my newly acquired iPhone to my very bright 12-year-old granddaughter when she elected to Google my name. To my amazement, the top hit appeared to link my name with that of Jade Goody, the late reality TV star. As far as I could recall I had never met with her, I have never knowingly watched Big Brother, and I don’t even research or treat cervical cancer, which Goody died from in March 2009.
It appears that, in the piece brought up by Google, I was being stigmatised as the leading anti-screening dinosaur. Everything became clear when on 23 March 2009, The Times (London) published a full-page obituary on Goody’s short and tragic life, and reported that her dying wish was that women under the age of 25 should have access to screening for cervical cancer. The so-called ‘Goody effect’, with more women going for cervical screening tests, has already provoked a campaign and a response from the UK Department of Health.
I have nothing against the late Jade; in fact she reminded me of the type of patient that used to make my National Health Service (NHS) clinics in central London such fun. Nevertheless I feel the urge to defend myself and to explain why being a screening sceptic might, after all, place me on the side of the angels rather than to the darker side of Darth Vader.
For a start I must hasten to mention that some of my best friends are women. My beloved mother died of breast cancer and my equally beloved sister is a long-term survivor of the disease. Galvanised by this experience, I have devoted my whole professional life to fighting breast cancer, visualising it as a slavering beast and in moments of self-delusion seeing myself as St Michael fighting against the disease. (My wife’s reminder – that the brand name ‘St Michael’ is sewn in the back of my underpants – tends to bring me down to Earth.)
In addition, I actually know a thing or two about screening. In 1987, when I was employed as professor of surgery at Kings College Hospital in London, I was commissioned by the Department of Health to establish the first breast-screening unit following the publication of the Forrest Report of 1986. This unit was built near Camberwell Green in south-east London and served as the training centre for the South East of England. Furthermore, for the past eight years I have acted as chairman of the National Trial to evaluate PSA screening for the early detection of prostate cancer.
Suddenly, a confluence of events has thrown the subject of screening for cancer into the public eye again. To start with, there was the publication of a report of the Nordic Cochrane Centre on Mammographic Screening in the British Medical Journal in February 2009 (2), along with an open letter to The Times signed by me and 27 other experts, both of which provoked a debate on the wisdom of screening for breast cancer. More recently, last weekend The Sunday Times reported that more and more women are angry about having had needless breast cancer operations, after research showed that ‘10 patients will be treated unnecessarily for every life saved’. The paper reported my view that, instead of breast-screening the whole female population over the age of 50 every three years, women should be tested ‘according to their level of risk’ – an argument I will expand on in this essay (3).
Then there were two papers published in the New England Journal of Medicine in March, reporting on the early results of screening for prostate cancer by using the PSA blood test (4). This was accompanied by banner headlines in the Daily Telegraph demanding PSA screening on the NHS that could save 2,000 lives a year.
Later on, I will explain just how wrong this statement was; but before then it behoves me to explain some of the delusions and illusions of screening in general.
Some truths about screening
The majority of lay people could be forgiven for believing that one of the mainstays in the fight against cancer is ‘early detection’. This belief has generated a European-wide consensus that screening for cancer before it becomes symptomatic will save lives. It has also become the main plank in the UK government’s campaign to improve cancer survival rates in Britain to match the highest levels achieved in other European Union countries.
In the vanguard of this campaign, the NHS screening programme for breast cancer (NHSBSP) by mammography has been lauded as a triumph and has laid claim to the responsibility for the dramatic decline in breast cancer mortality since its initiation more than 20 years ago. Those of us who have remained sceptical from the start have been branded as either misogynists or fools. Furthermore, if it’s good enough for women, what about men’s health as well? How can early detection of cancer be a bad thing?
Although it seems counterintuitive, a growing body of informed opinion is moving in the direction of exposing early detection of cancer as just that: possibly a ‘bad thing’.
Biases of screening
Let us start by considering two separate but related issues: firstly biases of screening that give a false impression of benefit, and secondly the over-detection of cancer ‘look-alikes’ which, if left undetected, might never threaten a patient’s life.
The survival from cancer is measured from the time of detection until recurrence and death. However, there are biases that impact on these survival rates. If a frame shift in the chronology of the disease due to screening occurs, then survival is automatically extended even if the ultimate outcome is the same; this is called ‘lead-time bias’. Of course, if the ‘cancer’ detected would never have threatened a woman’s life in the first instance, then that lead time might be as long as 30 years. Next, bearing in mind that the interval between screenings is anything from one to three years, it is inevitable that the fast-growing tumours with a bad prognosis will appear during the intervals while the slow-growing tumours with a good prognosis will sit around until found by mammography; this is called ‘length bias’.
There is also another subtle bias that can be described as the ‘self selection’ bias, in that women who accept invitations for screening might be demographically different to those who ignore the invitation. For a variety of reasons, such women have better outcomes in the treatment of cancer forgetting whether they were screen-detected or not.
The only way to account for these biases is to consider all the clinical trials of screening versus no screening and look for the pooled results described in terms of mortality – that is, the number of women dying in the screened group compared with those dying in the control group. And the fact is that there is only a modest advantage to screening when one looks at it in terms of mortality, as described in the recent publication in the British Medical Journal of ‘Breast screening: the facts – or maybe not’ by Peter C Gøtzsche and his colleagues from the influential and independent Nordic Cochrane Centre (2).
In this milestone paper, the authors describe a synthesis of all the papers that describe both the benefits and harms of screening using absolute rather than relative numbers. They conclude the following: If 2,000 women are screened regularly for 10 years, one will benefit from the screening, as she will avoid dying from breast cancer. (The US Preventive Services Task Force derived a similar number independently in 2004 (6).) However, even this 1:2,000 ratio might be an overestimate. Remember these data were derived from the trials that were mostly started in the 1970s and reported in the late 1980s. Since then, as a result of improvements in treatment, such as the adoption of tamoxifen and adjuvant chemotherapy, we have witnessed a drop in mortality of 30 to 40 per cent both in the age group that are invited for screening (50 and over) as well as for the younger woman. So perhaps the correct number might be 1:2,000 x 0.6 – or 1:3,000.
Now, if screening were as non-toxic as, say, wearing a seat belt, then there would be no case to answer, and the fact that one woman benefited from screening for every 2,000 women screened would be a good thing. However, there is a very definite downside: the problem of the over-diagnosis of ‘pseudo-cancers’. It is deduced by the Cochrane report that for every life saved, 10 healthy women will, as a consequence, become cancer patients and will be treated unnecessarily. These women will have either a part of their breast or the whole breast removed, and they will often receive radiotherapy and sometimes chemotherapy. This is the tricky part of the story that is more or less denied by the screening fraternity, and therefore deserves some close attention.
The nature of over-diagnosed cancers
Screening for breast cancer is now adopted as an unequivocal good by most of the member states of the European Union. Invitations for screening promote this activity by being economical with the truth (7). And one of the most uncomfortable truths concerns the over-diagnosis of both in-situ and invasive breast cancers in screening populations (8).
Over-diagnosis of breast cancer doesn’t mean false positive rates but the detection and treatment of cancers which, left undetected, would never threaten a woman’s life and with which she would live, in blissful unawareness, until she died naturally of old age. We had always assumed that there was an over-diagnosis of duct carcinoma in-situ (DCIS), some of which had the potential of progressing to an invasive and life-threatening phenotype. However, there is now clear evidence that anything between 10 and 50 per cent of invasive cancers detected and treated radically as a result of screening would never threaten life (9). As a result, the overall mastectomy rate rises after any country implements screening in contrast to the message in the NHSBSP leaflet – ‘Breast Cancer: The Facts’ (10) – which implies that screening saves breasts. It doesn’t.
How can this possibly be? Don’t we know that if cancer is neglected it will progress to a life-threatening condition?
By way of illumination, let me propose that the pathological diagnosis of cancer at screening is based on a syllogism (a syllogism is a logical argument in three propositions – two premises and a conclusion) with the conclusion being specious. A simple example might be that people die from meningitis; people that have meningitis harbour meningococci in their nose; therefore, nasal meningococci are lethal. In fact, about 10 per cent of the population harbour these bacteria but only a small proportion develop meningitis. Cancer was defined by its microscopic appearance about 200 years ago. The nineteenth century saw the birth of scientific oncology with the discovery and use of the modern microscope. Rudolf Virchow, often called the founder of cellular pathology, provided the scientific basis for the modern pathologic study of cancer. As earlier generations had correlated the autopsy findings observed with the unaided eye with the clinical course of cancer 100 years earlier, so Virchow correlated the microscopic pathology of the disease.
However, the material he was studying came from the autopsy of patients dying from cancer. In the mid-nineteenth century, pathological correlations were performed on living subjects presenting with locally advanced or metastatic disease that almost always were predetermined to die in the absence of effective therapy. Since then, without pause for thought, the microscopic identification of cancer according to these classic criteria has been associated with the assumed prognosis of a fatal disease if left untreated. Therefore, the syllogism at the heart of the diagnosis of cancer runs like this: people frequently die from malignant disease; under the microscope this malignant disease has many histological features we will call ‘cancer’; ergo anything that looks like ‘cancer’ under the microscope will frequently be lethal.
I want to argue that some of these earliest stages of ‘cancer’, if left unperturbed, would not progress to a disease with lethal potential. These ‘pseudo-cancers’ might have microscopic similarity to true cancers, but these appearances are only a necessary rather than sufficient condition for a fatal disease. I also want to suggest that many of the ‘risk factors’ for the development of cancer are in fact the promotional agents of a latent condition that H Gilbert Welch has described as ‘pseudo-cancers’. If we stand back and take a broader look at nature, this shouldn’t be surprising. Conventional mathematical models of cancer growth are linear or logarithmic – in other words completely predictable at the outset. These mathematical formulae may be appropriate for designing theme-park rides but cannot begin to explain the exquisite organisation of cell proliferation and the complex inter-relationships of cells of different progeny.
Most natural biological mechanisms are non-linear or better described according to chaos theory. Prolonged latency followed by catastrophe should not be all that surprising. We accept that to be the case for prostate cancer, as we know that most elderly men will die with prostate cancer in situ and not of prostate cancer that has invaded. In fact, the UK national PSA screening trial for prostate cancer is predicated on that fact, with two a priori outcome measures defined: deaths from prostate cancer versus the number of cancers treated unnecessarily. So, why oh why does the breast cancer lobby remain in denial?
Of course, now that the cat is out the bag, and there is evidence of harms in breast cancer screening, they will have no choice but to include this fact in their invitations in order to avoid litigation from an irate woman in the future. And increasingly today, ‘the sins of the mother are visited on the daughter’: we now have cases of women with screen-detected DCIS whose daughters have had problems raising mortgages when the insurers have discovered this family history of breast cancer! (11,12) Such negative impacts of screening are likely to impact on the general emphasis and style of the breast cancer lobby.
However, even the concession by the UK Department of Health (DoH) on some of these facts, as reported in The Times recently, only acknowledges one cancer over-diagnosed for every life saved. How on earth does the DoH derive that number? The Cochrane report is transparent on this matter and the numbers are there for all to see. They simply record the numbers of breast cancers that appeared in the screened group (observed), and subtracted the numbers of cancers that were expected to appear as seen in the unscreened control populations in the trials that have now been followed up for the majority of these women’s lifetimes. And Observed minus Expected = Over-Diagnosis. How the DoH arrived at the figure of one to one is inexplicable given the Cochrane centre’s thorough research which shows a much higher rate of over-diagnosis. The DoH is trotting out figures that are 10 years out of date.
By way of a minor diversion, I want briefly to mention prostate cancer screening. In my preamble, I mentioned the banner headlines in one of our respected broadsheets – about prostate screening in Britain potentially saving 2,000 lives a year – following the publication of expert studies in the New England Journal of Medicine in March 2009. Well, unlike the writers of those banner headlines, I actually read the expert publications and not just the press release of a prostate cancer charity.
There were in fact two studies reported together with a lengthy editorial. The first study was American and produced a negative result. The larger European study was of borderline statistical significance and, for the statistically literate amongst you, showed a 20 per cent relative risk reduction of cancer-specific mortality over a 10-year period with a p value of less than 0.04. Translated into numbers that all lay men might understand: you would need to screen 1,400 men for 10 years to save one prostate cancer death at the expense of over-diagnosing 48 cases of cancer that would be treated with radical surgery that frequently leads to impotence and incontinence and on rare occasions death from the complications of surgery.
The editorial concluded that it was premature to make any kind of recommendation and urged that we wait for the outcome of the British trial. That happens to be the study that I chair. I can claim no credit for the elegant and ambitious design of that trial, which is all down to the three principal investigators: Freddie Hamdy, Jenny Donovan and David Neal. I was invited in as an independent chairman to see fair play. Five-hundred thousand men in the UK have been randomised for the trial. Those with a raised PSA are further investigated and, if in the end their prostate biopsies show localised cancer, they are then offered re-randomisation to three treatment groups: radical surgery, radical radiotherapy or ‘active monitoring’. The two primary outcome measures of this trial will be cancer-specific mortality and the rate of over-diagnosis. Health economics is factored in, too, and the secondary randomisation will allow measures of quality of life as well as length of life in the different treatment groups, while the ‘active monitoring’ arm will allow us to study the natural history of screen-detected prostate cancer.
This latter group will also allow us to study the molecular biology of the disease to see if we can learn to separate out the ‘poodles’ (cancers that will remain latent for the duration of life) from the ‘Rottweilers’ (those that are predetermined to invade and spread). If only we had our time over again: this is how the breast cancer trials should have been designed, too.
Is there a reasonable ‘exit strategy’?
Let me summarise in bullet points where we have arrived with breast cancer screening:
- The current NHS screening programme is based on the results of randomised controlled trials that were published before 1987 and started in the late 1960s and early 1970s;
- Some of these trials in retrospect were of poor quality;
- With mature follow up, and careful attention to biases, a relative risk reduction (RRR) in breast cancer specific mortality has been estimated as 15 per cent;
- In absolute terms, therefore, the numbers needed to screen over 10 years to prevent one breast cancer death is about 1:2,000. Anything better than this depends on ignoring the obvious biases in the trials, mathematical manipulation of the data that I either simply don’t understand or is based on those self-selected women who accept the invitation to screen (‘selection bias’);
- Along the way, the estimates of harm have increased. At the outset, the hazards of over-diagnosis were ignored, then as the rate of screen-detected duct carcinoma in situ (DCIS) shot up, it was still judged to be worth the cost. Now we recognise that the over-diagnosis of invasive cancers (IDC) that are not predestined to threaten a woman’s life is a problem. The extent of over-diagnosis is debateable, but I personally agree that if you include DCIS and IDC it mounts to about 10 cases treated unnecessarily for every life saved;
- Putting politics aside for the moment, I wonder how many of us would have voted for the NHSBSP in 1987, knowing what we know today?
- Furthermore, in spite of the wonderful advances we’ve made in imaging technology and treatment in the past 20 years, there has been only one new trial reported for screening and that was the trial for the under-50s that supported the 15 per cent estimate (13);
- In other words, we are using state-of-the-art imaging and modern therapy to service a programme based on data that is 20 years old. It is also worth re-iterating at this juncture that improvements in the treatment of symptomatic patients since the mid-1980s leaves a much narrower window of opportunity for screening, so that even our estimates of 1:2,000 based on these old trials might have to be multiplied by a factor of about 0.6;
- So where do we go from here? To close the breast-screening programme is politically unacceptable. I therefore want to make two practical propositions for research and development. One concerns ‘person preferences’ and the other concerns the more efficient use of scarce resources that I will refer to as risk assessment/risk management (RARM):
Since 1997, when I resigned from the NHSBS committee, I have publicly expressed my concerns on the issue of informed choice for women invited for screening. This has often led to ridicule and ad hominem attacks. I take no particular pleasure in the fact that the NHS has at last accepted the point and agreed to rewrite the letters of invitation. As The Times reported on 21 February 2009: ‘NHS rips up breast cancer leaflet and starts all over again.’
My concern is that they will repeat the mistakes of the past if we leave this task to the usual suspects. Furthermore, it is not for me to prejudge what level of benefit and what level of harm might influence the average woman to accept the invitation. For this reason I think there are two related areas of research. First, the development of an information pack that includes decision aids. This could be used in a person preference study where well women might be offered sliding scales of benefits and harms to find the point at which screening is judged acceptable. Second, the data resulting from such studies might then inform the next and perhaps more important area of research on more efficient ways of using scarce resources in the NHS.
The beauty of risk assessment/risk management is that it provides a platform for the management of all women in an attempt to reduce the mortality from breast cancer where mammographic screening is one component of an integrated programme. The first step is to set up a nationwide facility for risk assessment using a modern computer programme. Women would then be offered, not compelled, to accept this service. Initially a practice nurse could administer a questionnaire, but it would be quite easy to transfer this to a web-based programme for computer-literate members of the community. From the read-out, an initial triage could be agreed. Those at the most extreme end of the risk spectrum, say with a relative risk (RR) of greater than 10.0, could be invited to a clinical genetics consultation. At the other extreme, those with a RR of, say, less than 2.0, might be reassured and given lifestyle advice on diet, alcohol and exercise. (Please note that these risk ratios are for illustration only; the actual figures used could be derived from the ‘person preference’ studies and the cut off for genetic counselling is already broadly accepted.)
Those in between these two extremes could then be invited to a special clinic for the second step. At this clinic, women of 45 or older could have a mammogram to determine breast density that might also be kept as a baseline but also provide additional evidence about risk. (The greater the mammographic density, the higher the risk.) Those with radiological abnormality at this stage would be investigated in the accepted way. If the mammographic density is low, and the repeat estimate falls below a RR of 2.0, then they would be reassured and given lifestyle advice. Those who remain with a RR of between 2.0 and 10.0 would be offered screening. In addition, those who were pre-menopausal might be offered prevention with tamoxifen and those who were post-menopausal could be offered entry into the IBIS II trial (a study comparing tamoxifen with arimidex for the chemo-prophylaxis of breast cancer). A recent paper in the Journal of the National Cancer Institute supports the validity of this approach (14).
To carry on regardless with certain cancer screening programmes is no longer acceptable; and neither is political spin the answer. Women are getting smarter and smarter on these issues; the demand for change doesn’t just come from grumpy old men like me, but also from the legions of wise women represented by the signatories of the letter in The Times.
The changes I have in mind are not destructive but constructive. The NHSBSP has indirectly led to the provision of the best specialist services for the diagnosis and treatment of symptomatic breast cancer in the world, riding on the back of the screening units. The centralisation of care has led to the rapid recruitment into randomised controlled trials for the treatment of cancer that is the major contributor to the dramatic fall in breast cancer mortality in the UK over the past two decades. If we can now add to this the prevention of the disease and a risk-adjusted screening programme, then everyone is a winner.
Professor Michael Baum worked for 30 years as a surgeon specialising in breast cancer, and is now professor emeritus of surgery at University College London.
Dr Michael Fitzpatrick warned of the dangers of smear testing post-Jade. Dr Michael Baum reviewed Devra Davis’ The Secret History of the War on Cancer. In the spiked/Pfizer ‘What inspired You?’ survey, Dr Baum explained how he got interested in the natural history of breast cancer. Elsewhere, he told the self-appointed custodians of men’s health: ‘Get your hands off my balls.’ Brendan O’Neill wondered if we were witnessing a people’s rebellion against the cancer cops. Or read more at spiked issue Health.
(2) ‘Breast screening: the facts-or maybe not’, Gøtzsche PC et al, British Medical Journal 2009, 338; 446-448
(3) Anger at ‘needless’ breast cancer ops, The Times, 19 April 2009
(4) ‘Mortality results from a randomized prostate screening trial’, Andriole et al, NEJM 2009, 360: 1310-1319; ‘Screening and prostate cancer mortality in a randomized European trial’, Schroder et al NEJM 2009, 360:1320-1328
(6) US Preventive Services Task Force. Screening for Breast Cancer: Recommendations and Rationale. Ann Intern Med 2002;137: 344-6
(7) ‘Screening and choice: Informed choice for screening: implications for evaluation’, Les Irwig, Kirsten McCaffery, Glenn Salkeld, Patrick Bossuyt, British Medical Journal, 2006;332:1148-1150
(8) ‘Rate of over-diagnosis of breast cancer 15 years after end of Malmö mammographic screening trial: follow up study’, Zackrisson S, Andersson I, Manjer J and Garne JP, British Medical Journal, 2006; 332: 689-92
(9) Should I be tested for Cancer?, H Gilbert Welch University of California Press, 2004
(10) ‘Incidence of breast cancer in Norway and Sweden during introduction of nationwide screening: prospective cohort study’, PH Zahl, BH Strand, J Maehlen, British Medical Journal, 2004; 328(7445): 921-4.
(11) ‘Randomised Clinical Trials: the patient`s point of view’, Hazel Thornton, included in Ductal Carcinoma in Situ of the Breast, (Ed) Melvin Silverstein, Williams and Wilkins 1997.
(12) ‘Insurance repercussions of mammographic screening: What do women think?’, Claire Davey, Victoria White, Jenette E Ward, Medical Science Monitor, 2003; 8: LE44-45
(13) ‘Effect of mammographic screening from age 40 years on breast cancer mortality at 10 years’ follow up: a randomised controlled trial’, SM Moss, H Cuckle, A Evans, L Johns, M Waller, L Bobrow, The Lancet, 2006, 368; 2053-2060
(14) ‘Prevention of Breast Cancer in Postmenopausal Women: Approaches to Estimating and Reducing Risk’, Steven R Cummings, Jeffrey A Tice, Scott Bauer, Warren S Browne, Jack Cuzick et al; J National Cancer Institute, 2009;101
To enquire about republishing spiked’s content, a right to reply or to request a correction, please contact the managing editor, Viv Regan.