As a physician who takes care of women with breast cancer, and as a breast cancer advocate, I strongly disagree with the Drs. Jha and Ware’s commentary entitled “Mammography Trials.” The commentary contains multiple inaccurate statements about screening mammography. Drs. Jha and Ware’s most egregious is their opening statement, “There is so much uncertainty precisely because the benefits are so small at the population level.” Their limited review of mammography trials focuses solely on the Canadian National Breast Screening Study (CNBSS). They have chosen to ignore the results of all the other randomized clinical trials (RCTs).
In fact, eight large RCTs have definitively demonstrated that screening mammography in women 40–74 years of age decreases mortality from breast cancer. Smith et al. summarized the RCTs and showed a significant mortality reduction of 15–30% from mammography screening . In fact, the true mortality benefit of screening mammography is even greater than what has been shown for RCTs. RCTs underestimate the mortality reduction because women invited to screen may not undergo mammography (noncompliance) and women in the control group may undergo mammography on their own, outside of the trial (contamination). Drs. Jha and Ware correctly state that the strength of RCT is the length of follow-up. This statement is true, especially in well-designed trials. Tabár et al.’s Swedish two-county trial of mammographic screening was the first breast screening trial to show a reduction in breast cancer mortality from screening with mammography alone, finding a 30% reduction in breast cancer mortality among 40- to 74-year-old women invited to screening . With a long-term follow-up of 29 years, the benefits are even greater; the number of women needed to undergo screening to prevent one breast cancer death decreased to 414 women .
The commentary’s sole focus on CNBSS is misleading. This RCT had biased randomization with patients with palpable abnormalities included in the screening arm, poor image quality, low sample size, and poor training of radiologists. These flawed methodologies have been refuted and cannot be corrected with 25 years of follow-up. Drs. Jha and Ware’s commentary is simply a rehash of Gøtzsche and Olsen’s paper that ignore the flaws of CNBSS and exaggerate the limitations of other RCTs . CNBSS is an outlier that does not stand up to rigorous methodologies and peer review. Recently, Coldman and colleagues conducted a pan-Canadian study that found a 40% reduction in breast cancer mortality in women who underwent screening mammography and had 19 years of follow-up . So contrary to Drs. Jha and Ware’s conclusion, the results of RCTs and observational studies show that the benefits of screening mammography are large, indeed, at the population level.
References
1. Smith R.A., Duffy S.W., Gabe R., et. al.: The randomized trials of breast cancer screening: what have we learned?. Radiol Clin North Am 2004; 42: pp. 793-806.
2. Tabár L., Fagerberg C.J., Gad A., et. al.: Reduction in mortality from breast cancer after mass screening with mammography: randomised trial from the Breast Cancer Screening Working Group of the Swedish National Board of Health and Welfare. Lancet 1985; 1: pp. 829-832.
3. Tabár L., Vitak B., Chen T.H., et. al.: Swedish two-county trial: impact of mammographic screening on breast cancer mortality during 3 decades. Radiology 2011; 260: pp. 658-663.
4. Gøtzsche P.C., Olsen O.: Is screening for breast cancer with mammography justifiable?. Lancet 2000; 355: pp. 129-134.
5. Coldman A., Phillips N., Wilson C., et. al.: Pan-Canadian study of mammography screening and mortality from breast cancer. J Natl Cancer Inst 2014; 106: pp. dju261.