Mammography - The Hidden Downside - Part V
Over-diagnosis is an acknowledged problem with screening mammography, leading to treatment that for some people may be both unnecessary and intrinsically damaging in its own right. The danger of a false positive reading, with all the attendant anxiety and ensuing interventions, is also always a risk in current screening mammography programs. Similarly, the real possibility of a false negative - a clean bill of health that turns out to be illusory - is inherent in screening mammography. Moreover, there is no guarantee whatever that a breast cancer identified by screening mammography will be curable.
Furthermore, as we have seen, for some premenopausal women, particularly those with node-positive disease, there is the additional danger that early diagnosis by means of mammography may actually reduce survival rather than extend it. It is worth noting that mammography screening for premenopausal women is not recommended in any other country except the US.
For older (postmenopausal) women, the benefits of mammography may be marginally greater, at least over time, although here again, there is a danger of over-diagnosis, and of high false positive (and negative) results.
Meanwhile, the debate over screening mammography continues unabated. The US medical profession continues to stand unwaveringly behind its recommendation that women aged 40 and up should undergo annual mammography. Just last month, for example, the New England Journal of Medicine (NEJM) published a paper that made headlines all over the world. It claimed that mammography had been proven responsible for saving lives from breast cancer. It is therefore worth examining this report a bit more closely.
It should be borne in mind that this was not actually a new clinical trial. Instead, this study was based on what are called ‘computer modeling techniques’ (i.e., statistical inferences and predictions based not on direct observations of patients but on computer simulations). These techniques were used to re-analyze seven prior studies of the effectiveness of mammography. In addition, no modifications or allowances were made in order to achieve consistency between the seven studies. Five out of the seven studies showed that mammography had contributed less to the decline in death rates than had improvements in treatment.
The most vocal proponents of screening mammography tend to claim that screening reduces the death rate by anywhere from 45 percent to 64 percent. However, in this study screening mammography was only found to have contributed approximately 15 percent to the decline in death rates from breast cancer, while improvements in treatment were found to have contributed approximately 19 percent (Berry 2005).
The usefulness of this study, and the validity of its conclusions, are further undermined by the fact that the sample population spanned the entire age range, from 30 to 79 years. No attempt was made to separate women into different age groups. As Professor Cornelia Baines of the University of Toronto pointed out, this is a particularly important omission since the natural history of the disease varies widely in different age groups. For women in the age group 30-49, mammography’s benefits are the most questionable of all - a fact that was entirely ignored by this study (Baines, personal communication).
Yet despite this latest favorable NEJM article and despite the incessant repetition of the “mammography saves lives” mantra, there is, astonishingly, still no consistent, substantial scientific evidence that regular mammography results in a significant reduction in mortality from breast cancer. In an important paper published in 2000 in the prestigious journal Lancet, Swedish researchers, working on behalf of the international Cochrane Review organization, reviewed the quality of the major mammography trials to date and came to the following conclusions:
“Screening for breast cancer with mammography is unjustified. If the trials are judged to be unbiased, the data show that for every 1000 women screened biennially throughout 12 years, one breast cancer death is avoided whereas the total number of deaths is increased by 6″ (Gotzsche 2000).
In a paper examining the contradictory evidence concerning mammography screening, Steven Goodman, MD, a biostatistician at the Johns Hopkins Sidney Kimmel Cancer Center, Baltimore, MD, has written:
“If we take a step back, this controversy looks almost Swiftian when we consider that even under the most optimistic assumptions, mammography still cannot prevent the vast majority of breast cancer deaths. There will come a time when all the study patients have been followed up, all the analyses have been done, all the expert groups have met, and all the editorials have been written, and we still won’t be sure how much benefit and how much harm are caused by mammography. We must find good ways to help women deal with this uncertainty, for that time is imminent” (Goodman 2003).
The Role of Breast Self-Examination
So how can a woman prudently and effectively improve her chances of detecting breast cancer? What alternatives and adjuncts are there to annual mammography? What new developments are in the pipeline?
First, the value of a really thorough clinical breast examination (CBE) and breast self-examination (BSE) has been routinely downplayed and underestimated by the medical profession. A surprisingly high proportion of breast cancers are actually discovered by women themselves, without the aid of anything more high-tech than their own familiarity with the way their breasts feel, month by month. The American Cancer Society admitted some twenty years ago that almost 90 percent of all breast cancers were actually initially detected by women themselves (Ross 1987).
Although false positives do occur with clinical breast examinations, they are, perhaps surprisingly, less than half as common as they are with mammography. One study has shown that CBE is more sensitive (i.e., better able to detect abnormalities) than mammography in younger women with denser breast tissue. The same study found that CBE is better than mammography at detecting dangerous ‘interval’ cancers. Indeed, this study found that the combined use of CBE with mammography detected more abnormalities than either modality used alone (Elmore 1998).
The Canadian National Breast Cancer Screening study concluded that in women aged 50-59 years, “the addition of annual mammography screening to physical examination has no impact on breast cancer mortality.” That is, even though mammography was able to detect cancers at a smaller size, before they became large enough to be detected by CBE or BSE, this still did not improve survival rates, because “the majority of the small cancers detected by mammography represent pseudo-disease or overdiagnosis” (Miller, 2000).
This is not to suggest that there is no place for mammography in the detection of breast cancer. Far from it. Mammography certainly does have a place, and a useful place. As a diagnostic tool, it is the best we currently have. However, screening and diagnosis are two entirely different things.
Screening is intended to pick up possible abnormalities in otherwise healthy individuals, whereas diagnosis is the method whereby an abnormality, often initially detected by screening is more closely examined in order to identify its true nature (i.e., what its origins are, and whether the abnormality is benign or malignant, for example). The characteristics that make a good screening test are not by any means the same as those that are needed for diagnostic purposes.
The judicious and selective use of an imaging technique as a diagnostic tool in patients who have already been identified as having a suspicious lesion is not at all the same thing as the blanket application of an imaging technique in the mass screening of an entire population.
In the headlong rush towards mass mammography, the value of really thorough clinical and self breast examination has been almost entirely eclipsed. As we have seen, there are distinct dangers and drawbacks to mass screening with mammography, and the standard recommendation of annual mammography for all women over 40 is badly in need of revision. Particularly in these younger women, in whom mammography is both more dangerous and less able to detect abnormalities, regular CBE and BSE are of great importance (Baines 1997; Epstein 2001).
- Ralph W. Moss, Ph.D.
