Mammography is the technology of choice for many of the world's national screening programmes for breast cancer. But several new studies suggest it isn't fit for purpose and that its cons far outweigh the pros
Itseemed such a good idea at the time: routinely screen every woman over the age of 50 and you'll reduce deaths from breast cancer by a third. And so, buoyed by such a benefit, the UK introduced its national screening programme using X-ray mammography in 1988.
On the face of it, it seems to have delivered. Around the time routine mammography was introduced in the UK, 42 women per 100,000 were dying of breast cancer every year in the UK; by 2011, this had dropped to 25 women.
Today in the UK, around 1.6 million women aged between 50 and 70 years are invited to undergo routine mammography screening every year. The National Health Service points out in the invitation that mammography is saving around 1,300 lives a year by detecting breast cancer early and is the method by which one-third of breast cancers are diagnosed (although it doesn't explain how the majority is discovered).
But this year, three major studies have said that women aren't being told the full story about mammography. The screening programme hasn't had much to do with the decline in breast cancer deaths, and the number of cases detected is outgunned by a factor of 10 to one for cancers 'seen' by the technology but which aren't actually there (false positives). In other words, for every woman correctly diagnosed, 10 others will go through the rigours of chemotherapy, radiotherapy or even mastectomy (breast removal) to treat a cancer that was never there.
Why the false positives?
Most false-positive readings happen when the mammogram picks up a case of DCIS (ductal carcinoma in situ) which, despite its name, very rarely turns into cancer. Because of mammography's limitations, the oncologist must assume that each DCIS detected is malignant and so begin a process of testing and even surgery.
Around 60,000 new cases of DCIS are detected by mammography screening every year in the US alone, and yet, fewer than 5 per cent, or 3,000, develop into breast cancer. This suggests that around 57,000 American women go through the trauma of unnecessary worry, biopsies and surgery every year for no reason.1
The Swiss Medical Board, an independent health assessor, started the ball rolling in February. It took a fresh look at the early studies that had ushered in the age of mammography and found them wanting. The studies, which dated from 1963 to 1991 and followed 1,000 women who had been screened and a further 1,000 who hadn't, advocated the uptake of mammography as a national screening technology, although the data didn't actually support such emphatic conclusions. In fact, the data suggested the cons outweigh the pros.
Although mammography might save one or two lives, it will detect a further 100 false-positive cases. The Swiss researchers said that health authorities should start setting a time limit on their national screening programmes with a view to winding them down, and that nobody should be thinking about setting up mass-screening programmes if they haven't already done so.1
Their conclusions have caused an uproar. The Swiss Cancer League said it was astonished, while other groups accused the researchers of being unethical, irresponsible and adding to the breast cancer death toll. But not everyone has taken that line. Eleven of Switzerland's 26 cantons (districts) currently offer routine mammography, but the German-speaking canton of Uri may not now be joining that number. After reading the board's conclusions, it is reconsidering its decision to introduce a screening programme.
In response to their critics last April, two of the board's researchers-Nikola Biller-Andorno and Peter J"uni-point out three very worrying trends uncovered by their analysis.2 The first was that the benefits of mammography are based on outdated trials, such as those they analyzed. Since the last of the trials carried out in 1991, better and more targeted cancer treatment has been introduced, and this may have had more to do with the drop in death rates from breast cancer than the widespread use of mammography.
Even after carefully re-reading the data, it is still not at all obvious that mammography benefits outweigh the harm. The original expectation that mammography would reduce breast cancer deaths by a third was based on conclusions of the UK's 1986 Forrest Report, which ushered in universal screening, but this relied on just two studies, both of which were inaccurate.
More recently, estimates for reductions in breast cancer deaths have varied from 10 per cent to 20 per cent, but even those benefits seem to come with strings attached.
The Canadian Breast Screening Study monitored the lives of nearly 90,000 women aged between 40 and 59, around half of whom had regular mammogram screening, while the rest had physical examinations and the usual community care.3 During the study, 3,250 women in the mammography group and 3,133 in the physical-examination group were diagnosed with breast cancer, and 500 and 505 of them, respectively, died. But although mammography was no better than physical examination at detecting cancer and saving lives, 22 per cent of cases detected by mammography were false positives, which triggered unnecessary invasive treatment.
The prestigious Nordic Cochrane Centre in Copenhagen, led by Peter Gotzsche, came to similar conclusions. They took another look at eight studies of mammography involving around 600,000 women aged 39 to 74. Mammography didn't reduce the number of deaths from cancer over a 10-year period, even though the rates of lumpectomy, mastectomy and use of radiotherapy were far higher among women who had been screened. The women treated went through psychological distress, anxiety and uncertainty for years-and yet, for too many of them, it was utterly unnecessary because the mammogram had got it wrong. The Cochrane researchers estimate that, for every 2,000 women screened, one will avoid dying from breast cancer and 10 healthy women will get a false-positive reading.4
Then there are the two mismatched worlds that seem to be happily coexisting: the researchers who keep uncovering inconvenient truths about mammography; and the women who are the targets of screening. The two Swiss researchers were "disconcerted" by the women's perceptions about the benefits of mammography. One study by the Swiss Cantonal Health Office in Bellinzona, which surveyed 4,140 women, found that 68 per cent thought mammography reduces the rate of breast cancer deaths by at least half. As the research team concluded, this rose-tinted view of mammograms throws into doubt the possibility of true informed consent.5
Looking back on the controversy they have created, the Swiss researchers note that the main argument against their conclusion was that it contradicted the global consensus of leading experts in the field, which "made us appreciate our unprejudiced perspective resulting from our lack of exposure to past consensus-building efforts by specialists in breast-cancer screening". In other words, file under 'Emperor's New Clothes'.
Essentially, mammography simply isn't fit for purpose, they conclude. "From an ethical perspective, a public health programme that does not clearly produce more benefits than harms is hard to justify. Providing clear, unbiased information, promoting appropriate care and preventing over-diagnosis and over-treatment would be a better choice."
What's the alternative?
For 30 years until the mid-1980s, thermography, or digital infrared thermal imaging, was considered a viable alternative to mammography; radiologists saw it as complementary and often used the two screening technologies together. Essentially, mammography picks up mass while thermography registers activity as reflected by heat.
It is also safer than mammography, as it neither subjects the patient to a dose of radiation nor needs to have the breast crushed between plates.
Early studies supported the technology with enthusiasm. In one, 10 per cent more cancers were detected when thermography and mammography were used together,1 while another found that thermography was the better 'first alarm' in 60 per cent of cancer cases detected.2
Thermography fared even better in a study of 39,802 women, screened over a three-year period, in which researchers concluded that 30 per cent of the cancers would never have been detected had only mammography been used.3 Overall, an abnormal infrared image is the strongest risk indicator for the future development of breast cancer, and is 10 times as significant as a family history of the disease. In one study of around 58,000 women screened with thermography, more than a third of those who had an abnormal reading developed cancer within five years. The researchers concluded that "an abnormal thermograph is the single most important marker of high risk for the future development of breast cancer".4
Other viable alternatives include clinical examination and ultrasound. Ultrasonography is an accurate follow-up procedure when a mammogram detects an abnormality and the woman is at moderate risk for breast cancer.5
SPECIAL OFFER ON 'THE PROMISE' DVD
The Promiseis one of those films that every woman thinking about mammography should see first. It explores the politics behind the decision to launch a national screening programme for all women over the age of 50. Eminent figures in the worlds of medicine and radiology are interviewed, including Prof Michael Baum, who pioneered the introduction of the UK's mass screening programme, and WDDTY editor Lynne McTaggart.
A DVD of The Promise is available to WDDTY readers at the special price of lb12.24 including p&p (a 10 per cent discount). To order your copy, go to http://medscans.co.uk/product/the-promise-about-the-harms-of-the-routine-breast-screening-program/ and key in WDDTY in the special promo box to claim your discount.
N Engl J Med, 2014; doi: 10.1056/NEJMp1401875
BMJ, 2014; 348: g366
Cochrane Database Syst Rev, 2013; 6: CD001877
Int J Epidemiol, 2003; 32: 816-21
why the false positives? References
JAMA, 2009; 302: 1685-92
what's the alternative? References
Interamer J Rad, 1987; 12: 337
Spitalier H et al. 'Does infrared thermography truly have a role in present-day breast cancer management?', in Gautherie M, Albert E, eds. Biomedical Thermology: Proceedings of an International Symposium. New York: A.R. Liss, 1982: 269-783
Ann N Y Acad Sci, 1980; 335: 492-500
Cancer, 1980; 45: 51-6
Ann Intern Med, 2003; 139: 274-84