Menopause Live (30 January, 2012) from IMS

The original decision to introduce mammographic screening in the UK was based on The Forrest report [1] which analyzed the costs and benefits of screening in terms of quality-adjusted life-years (QALYs) and found that screening would reduce the death rate from breast cancer by almost one-third with few harms and at low cost. Since that time, certain harms associated with breast screening have been identified including overdiagnosis and unnecessary surgery [2].

A recently published paper [3] set out to replicate the original Forrest report using baseline mortality and risk of surgery from English data (not originally used by Forrest) but also included the perceived harms noted above, using estimates based on those published by The Nordic Cochrane Centre [4] and by Nelson and colleagues [5,6] for The US Preventive Services Task Force.

The outcomes of 100,000 women aged 50 years were modelled in two cohorts, one screened, the other not. Outcomes measured were deaths from breast cancer, deaths from all other causes and the number of women having false-positive diagnoses and surgery, which were combined into QALYs. The authors found that inclusion of the effects of harms reduced the updated estimate of net cumulative QALYs gained after 20 years from 3301 to 1536 or more than 50%. Using Nordic Cochrane Centre estimates generated negative QALYs for the first 7 years of screening and only 70 positive QALYs after 10 years and 834 after 20 years. Results were similar when estimates used by Nelson were applied. Hence breast cancer screening might have caused net harm for up to 10 years from the start of screening.

Comment
In countries where mammographic screening has been introduced, mortality from breast cancer has been reduced by approximately 30% over 20 years, as predicted by The Forrest report [1]. However, the proportion of that reduction attributable to the screening process is probably less than half, with the major reduction being due to improved treatment of breast cancer. The most compelling evidence of this comes from a study by Kalager and colleagues [7] who compared incidence-based death rates between two groups of women living between 1996 and 2005 in Norwegian counties with, or without, screening and two historical control groups which, from 1986 to 1995, mirrored the other two groups. They found that, of the 28% relative reduction in mortality in the screened groups, only about one-third (10%) could be attributed to screening. Whilst this reduction is still significant, any real harms from the screening process might substantially reduce net benefit. Of the proposed harms (psychological, radiation exposure, false-positive recalls, unnecessary biopsies), by far the most controversial has been the claim of overdiagnosis. The Nordic Cochrane Centre [8] estimates this to be 52% and suggest that, therefore, women who avoid mammographic screening reduce the risk of being diagnosed with breast cancer by one-third [9]. However, estimates of overdiagnosis range from 1% to 54% and, depending on the denominators used, the same studies may vary estimates of overdiagnosis by a factor of four [10]. Debate on this topic is passionate and, sometimes, personal. The current paper under review has already been damned by no less than The American College of Radiology which accuse the authors of falsifying data (Menopause Live, 26 Dec 2011). This approach is not helpful, particularly from bodies that might be accused of having a vested interest in screening.

There is little doubt that screening has contributed to the reduction in breast cancer deaths over the past 20 years. The real question is ‘by how much?’. There is also little doubt that overdiagnosis occurs: it will in any screening program where the biology of individual tumors cannot be known. Again, the question really is related to the magnitude of overdiagnosis. Importantly, there is also little doubt that the majority of women believe in mammographic screening and few, if any, would regard surgery, which seemingly ‘cured’ them, as ‘unnecessary’.

Therefore, rather than abandon screening, we should acknowledge the shortcomings of the current screening programs and work collaboratively to maximize benefits and minimize risks, perhaps by redefining what constitutes a ‘suspicious’ mammogram, perhaps by employing different screening modalities, perhaps by increasing screening intervals and perhaps by reviewing what age groups will derive most benefit.

Rod Baber
Clinical Associate Professor of Obstetrics and Gynaecology, Sydney Medical School, The University of Sydney, NSW, Australia

References
1. Forrest P. Breast cancer screening. Report to the Health Ministers of England, Wales, Scotland and N Ireland by a working group chaired by Professor Sir Patrick Forrest. HMSO, 1986.
http://www.cancerscreening.nhs.uk/breastscreen/publications/forrest-report.html
2. WHO IARC. Breast cancer screening. International Agency for Research on Cancer/WHO, 2002.
http://www.iarc.fr/en/publications/pdfs-online/prev/handbook7/Handbook7_Breast-0.pdf
3. Raftery J, Chorozoglou M. Possible net harms of breast cancer screening; updated modelling of Forrest report. BMJ 2011;343:d7627.
http://www.ncbi.nlm.nih.gov/pubmed/22155336
4. Gøtzsche PC, Nielsen M. Screening for breast cancer with mammography. Cochrane Database Syst Rev 2009;4:CD001877.
http://www.ncbi.nlm.nih.gov/pubmed/19821284
5. Nelson H, Tyne K, Naik A, et al. Screening for breast cancer: an update for the US Preventive Services Task Force. Ann Intern Med 2009; 151:727-37.
http://www.ncbi.nlm.nih.gov/pubmed/19920273
6. Nelson H, Tyne K, Naik A, et al. Screening for breast cancer: an update for the US Preventive Services Task Force. Agency for Healthcare Research and Quality, 2009.
http://www.uspreventiveservicestaskforce.org
7. Kalager M, Zelen M, Langmark F, Adami HO. Effect of screening mammography on breast-cancer mortality in Norway. N Engl J Med 2010;363:1203-10.
http://www.ncbi.nlm.nih.gov/pubmed/20860502
8. Jorgensen K, Gotzsche P. Overdiagnosis in publicly organized mammographic screening programmes: a systematic review of incidence trends. BMJ 2009;339:b2587.
http://www.ncbi.nlm.nih.gov/pubmed/19589821
9. Gøtzsche PC, Jørgensen KJ, Zahl PH, Mæhlen J. Why mammographic screening has failed to live up to expectations from randomized trials. Cancer Causes Controls 2012;23:15-21.
http://www.ncbi.nlm.nih.gov/pubmed/22072221
10. de Gelder R, Heijnsdijk EA, van Ravesteyn NT, et al. Interpreting overdiagnosis estimates in population-based mammographic screening. Epidemiol Rev 2011;33:111-21.
http://www.ncbi.nlm.nih.gov/pubmed/21709144

Scientific Editor: Amos Pines (apines@netvision.net.il)
Editorial Consultant: Jean Wright (jwright.ims@btopenworld.com)