Feature

Mammography -- to screen or not?

Opinion is divided over whether it is effective -- in both health and economic terms -- to screen women in their 40s for breast cancer.

John Le Heron

The following three opinions have appeared in various issues of the Journal of the American Medical Association this year:

From the United States National Institutes of Health Consensus Conference on breast cancer screening for women aged 40 to 49:

At the present time, the available data do not warrant a single recommendation for mammography for all women in their 40s.

From the American Cancer Society, which has noted that women in their 40s should be screened at yearly intervals, rather than every two years, as in an earlier recommendation:

Over the years, one thing has become clear. If you screen women under age 50 at intervals of 2 or more years you will not be able to accomplish nearly as much as you might if you screen them on an annual basis.

From the United States National Cancer Institute:

Women in their 40s who are at average risk should get a screening mammogram every 1 to 2 years.

Why are such divergent recommendations being expressed by ostensibly responsible bodies? The controversy can be added to with the American College of Radiology, the American Medical Association, and the American College of Obstetricians and Gynaecologists all recommending screening in this age group, while the American College of Physicians, the U.S. Preventive Services Task Force, and the American Academy of Family Practice do not recommend screening for women in their 40s.

With the imminent commencement of the national breast screening programme in New Zealand, it is perhaps relevant to briefly review what underlies these differences.

Statistics on mortality from breast cancer show a strong relationship with age. In the US the mortality from breast cancer for women in their 40s is 30 per 100,000, while for women aged 65 and over it is 126 per 100,000.

Other factors are relevant: breast cancer in younger women tends to be more virulent, growing faster, and killing faster; and detection of cancer in younger breasts is more difficult than in breasts of older women due to the presence of greater amounts of glandular tissue (with similar x-ray properties to tumour tissue).

Difficulties with Screening

Figures of performance typically quoted are that mammography will detect about 85% of cancers in older women, while this drops to about 60% for women in their 40s. These factors mean that any screening younger age groups of women is a very difficult job -- fewer cancers to detect, coupled with a less accurate detection process. And similarly, demonstrating the efficacy of such screening programmes is difficult -- large populations need to be studied, with shorter screening intervals, using optimal mammographic technique, and longer follow-up.

The fundamental measure of the success of a screening programme is whether the mortality rate can be lowered as a result of the screening. The evidence from international studies is that breast cancer deaths in the population of women aged 50-69 can be reduced by about 30% by good mammography screening programmes.

However the evidence is not so unequivocal for the 40-49 age group. This is partly because the majority of the studies were not designed to answer this particular question, and the one study that was so designed appears to have been badly confounded by other factors. It is also partly because of the difficulties outlined above.

The reduction in mortality from screening women over 50 appears after about five years. Any hint of benefit for women screened in their 40s is not apparent until more than ten years later.

Why the difference? Those with an anti-screening stance say that the women in the studies were often in their late 40s at entry into the trials, and that they were in effect 50s women by the time the benefits of being screened were being realised. The pro-screening group argue that the inappropriate screening intervals (typically two years, but up to three years) used in the trials led to delay in the appearance of benefit.

Policy Problem

Where does this place the policy makers? The advocates for screening women in their 40s claim that the studies do show a reduction in mortality, be it a smaller reduction than for women in the older age groups. They then claim additional support for screening by pointing out that modern mammographic techniques are superior to those used in the existing studies, and that the screening intervals in the studies were too long, making the programme less effective in reducing mortality.

On the other hand, the opponents to screening for the 40s women claim that the same studies do not show a reduction in mortality. They add weight to their case by considering the psychological harm caused by the relatively large number of false positives (estimated to be in the range of 5-11%) that occur with mammography in this age bracket.

A further comment made by the opponents is that detecting cancers at any earlier stage is not the same as saving lives. The virulence of breast cancer in younger women may negate the advantages of early detection. (This of course runs counter to the argument in the paragraph above of needing to screen frequently because of the same virulence.)

It would be nice to think that the studies could be interpreted in only one way, but the reality is that they contain many ambiguities, and they do not represent current practice. Because the science is unclear, other factors come into play. Radiologists are face-to-face with women who develop breast cancer, and feel strongly that such personal tragedy could be averted with screening.

Those sceptical of screening younger women tend to be versed in the science of evidence-based medicine. For these, the number of lives saved is not the only consideration. The number of women required to be screened to save one life has to be considered, and so does the impact of the false positives and unnecessary surgery. In other words, the goal posts are shifting. Reduced mortality is now not the only measure, and the human price of screening large numbers of women is part of the equation.

It is worth noting that possible radiation effects (radiation-induced breast cancers) are not seen as a factor in the debate. Current estimates of such cancers are a strong function of age, with very young women being most at risk. The emphasis is on producing optimum diagnostic images consistent with a minimum of patient dose. This is achieved by using state-of-the-art technology for the x-ray machines and imaging devices, and having appropriately trained professionals involved in the imaging process.

Where does this leave breast screening in New Zealand? The nationwide breast screening programme will commence with asymptomatic women aged 50-64 eligible for screening. Extension to women older than 64 will be considered again once the programme is in place and running well.The Ministry of Health has not recommended routine screening for women in their 40s, but will continue to monitor the latest information and, if sufficient new evidence becomes available, the screening guidelines will be reassessed.

John Le Heron is with the National Radiation Laboratory.