Karsten Juhl Jørgensen, MD, DrMedSci, Deputy Director, The Nordic Cochrane Centre, Rigshospitalet, Copenhagen, Denmark.
Breast cancer screening has been disputed for many years. The origin of the dispute is the conflicting results reached by the nine original randomised breast screening trials. One of these trials estimated a 42% reduction in breast cancer mortality while others estimated no reduction, or only a smaller benefit . The most important harm is also disputed. As for prostate cancer screening, there is now general agreement that breast screening leads to the detection of some tumors that would not have been detected or caused health problems in the absence of screening [2,3,4]. The dispute relates to how many such overdiagnosed tumors there are.
The very optimistic estimates of the benefits of mammography screening are at odds with observations. Fortunately, there have been great reductions in breast cancer mortality since the 1990’s, but comparisons between various countries and age groups indicate that this is mainly due to improved treatment, primarily the introduction of adjuvant therapy from the 1990’s [5,6].
In our recent study in Annals of Internal Medicine , we took advantage of a unique situation. Denmark has had a 17 year period from the early 1990’s where only 20% of women aged 50 to 69 years have been offered screening because only two Danish administrative regions chose to introduce it. Screening was rare outside these regions, providing us with a contemporary unscreened “control group” of same-age women. We could therefore separate the effects of screening from other factors and showed that while breast screening has led to the detection of many more small breast tumors, this did not lead to a drop in the occurrence of large tumors due to earlier detection as expected. Such a drop is necessary for screening to lead to improved survival and less use of comprehensive surgery – the major benefits of screening. It also indicates substantial overdiagnosis. We calculated that likely, 1 in 3 breast cancer diagnoses in a population of women offered screening is overdiagnosed.
To understand the importance of overdiagnosis, imagine the psychological stress that a woman and her family experience after receiving a breast cancer diagnosis. You likely know someone who has found themselves in this situation. The diagnosis almost invariably leads to treatment – removal of part of or the whole breast, radiotherapy and chemotherapy. This treatment has well known physical harms such as lymphoedema in the arm on the affected side; increased mortality from heart disease from left-sided radiation; and increased cancer risk from the chemotherapy. Then imagine that all of this was in vain – the diagnosis, worry and treatment was unnecessary.
Contrary to Denmark, neighbouring Sweden has had nation-wide breast screening since the late 1980’s, of a wider age range. Participation rates have been very high, >80%. But while Danish women in the age range offered (50 to 69 years) have experienced a drop in breast cancer mortality of 26%, the same age group in Sweden has experienced a drop of only 16%. Remember that breast screening only covered 20% of these women in Denmark. For Danish women below age 50 years, who have never been offered screening, the drop was 50%, much greater than in the screened age range. The corresponding figure for Sweden was only 36%, despite that some Swedish women in their 40’s are offered screening . In 2010 in BMJ, we compared breast cancer mortality rates for these age groups between Danish areas with and without breast screening and found that the magnitude of these declines was indistinguishable. While this evidence is not strong enough to exclude that breast screening can have a small effect, it does indicate that the vast majority of the credit for these declines should go to better treatment.
So why did some trials find a large effect of breast screening? The likely explanation is that these trials were not rigorously designed . One of the best designed trials according to independent groups of experts that use standard criteria for quality assessment was in fact a trial performed in Canada . This trial did not find an effect of breast screening. Another well-designed trial, from Malmö in Sweden, found a comparatively small effect . The latest rigorously designed trial, from the UK, found the same . When there is conflicting evidence and this can be explained by differences in trial design, the general (and sensible) advice is to trust the results of the well-designed trials . Even if these results are unwelcome.
The explanation why the benefit of breast screening is small may be ascribed to the biology of breast cancer. This is not one disease, but rather a spectrum of diseases. Some cases grow fast and spread early while others grow and spread slowly or not at all. Unfortunately, screening at intervals will predominantly identify those cancers that are present for the longest time – the slow growing ones with a good prognosis. The fast growing, aggressive ones pop up between screening rounds. They have already sent out their metastases at the time of detection and have become a systemic disease. Fortunately, our systemic (adjuvant) treatments have proven beneficial effects . We are not left powerless even if breast screening has not delivered the benefits we all hoped for.
Contrary to breast screening, other types of cancer screening do deliver on their promises. When we detect precursors to cervical cancer with screening, invasive cervical cancer incidence and mortality goes down . When we detect polyps in the intestine with screening for colorectal cancer using sigmoidoscopy, colorectal cancer incidence and mortality goes down . But when we detect and remove breast cancer “precursors” (called DCIS) due to breast screening, incidence of invasive breast cancer does not go down. It rises sharply . The reason we do not see the benefits of screening for breast cancer is thus not due to flawed methods. It is because breast cancer screening is more akin to prostate cancer screening than to screening for colorectal and cervical cancer.
When it comes to breast cancer screening, we need to inform women about its harms and focus our resources on treatment for the sick rather than screening healthy women.
Click here to read the Table of Contents for the series: Breast Cancer Screening, Mammography and “Alternative Facts”
The opinions expressed in this article are the author’s own and do not reflect the view of Cancer Knowledge Network or Multimed Inc.
- Gøtzsche PC, Jørgensen KJ. Screening for breast cancer with mammography. Cochrane Database Syst Rev 2013, Issue 6. Art. No.: CD001877. DOI: 10.1002/14651858.CD001877.pub5.
- Independent UK Panel on Breast Cancer Screening. The benefits and harms of breast cancer screening: an independent review. Lancet 2012;380:1778–86.
- Oeffinger KC, Fonthom TH, Etzioni R et al. Breast cancer screening for women at average risk. 2015 guideline update from the American Cancer Society. JAMA 2015;314:1599-1614.
- Siu AL, on behalf of the U.S. Preventive Services Task Force. Screening for Breast Cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med.2016;164:279-296.
- Autier P, Boniol M, Gavin A, Vatten LJ. Breast cancer mortality in neighbouring European countries with different levels of screening but similar access to treatment: trend analysis of WHO mortality database. BMJ 2011;343:d4411.
- Autier P, Boniol M, LaVecchia C, et al. Disparities in breast cancer mortality trends between 30 European countries: retrospective trend analysis of WHO mortality database. BMJ 2010;341:c3620.
- Jørgensen KJ, Gøtzsche PC, Kalager M, Zahl PH. Breast cancer screening in Denmark: a cohort study of tumor size and overdiagnosis. Ann Int Med 2017;DOI:10.7326/M16-0270.
- Jørgensen KJ, Zahl PH, Gøtzsche PC. Breast cancer mortality in organised mammography screening in Denmark. A comparative study. BMJ 2010; 340:c1241.
- Miller AB, Wall C,Baines CJ, Sun P, To T, Narod SA. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ 2014; 348:g366. doi: 10.1136/bmj.g366.
- Zackrisson S, Andersson I, Janzon L, Manjer J, Garne JP. Rate of
over-diagnosis of breast cancer 15 years after end of Malmö mammographic
screening trial: follow-up study. BMJ 2006;DOI:10.1136/bmj.38764.572569.7C.
- Moss S, Wale C, Smith R, et al. Effect of mammographic screening from age 40 years on breast cancer mortality in the UK Age trial at 17 years’ follow-up: a randomised controlled trial. Lancet Oncol 2015;16:1123-32.
- Schünemann H, Brozek J, Gyatt G, et al. GRADE Handbook: Handbook for grading the quality of evidence and the strength of recommendations using the GRADE approach. Available at: http://www.guidelinedevelopment.org/handbook/. Accessed 12. May 2016..
- Early Breast Cancer Trialists’ Collaborative Group. Aromatase inhibitors versus tamoxifen in early breast cancer: patient-level meta-analysis of therandomised trials. Lancet 2015;386:1341-52.
- Raffle AE, Alden B, Quinn M et al. Outcomes of screening to prevent cancer: analysis of cumulative incidence of cervical abnormality and modeling of cases and deaths prevented. BMJ 2003;326:901-5.
- Holme Ø, Schoen RE, Senore C, et al. Effectiveness of flexible sigmoidoscopy screening in men and women and different age groups: pooled analysis of randomised trials. BMJ 2017; 356:i6673