Cancer Rose, 16 August 2020
In August 2020, the final results of the UK Age Trial [I] were published.
For women aged 40-49 years
Between 1990 and 1997, this British trial included approximately 161,000 women aged 39 to 41.Women were randomly selected and an annual mammography up to the age of 48 was proposed for about one in three (53,883 women), while the remainder had no screening. All the women then joined the standard British screening program, which includes a mammogram between the ages of 50 and 69 every three years. The main aim of the trial was to determine whether screening could reduce mortality from breast cancer before the first mammogram of standard screening program, which starts after the age of 50.
Prior to these results, no evidence of a benefit
After 10 years, the UK Age Trial results showed a statistically significant decrease in the number of breast cancer deaths before routine screening program. The authors announced a 25% decrease in relative value, but this actually corresponds to a gain of only 4 deaths from breast cancer per 10,000 women screened and followed for 10 years. Furthermore, the results did not demonstrate a decrease in total mortality (or all-cause mortality).*
There was no statistically significant decrease in deaths from cancer when the results of all nine studies that included women aged 40-49 (not just the UK Age Trial) were considered.
*Only the total mortality includes all elements of patient management, hence also the effects of treatment, overdiagnosis and overtreatment.
This makes more sense because any cancer detected will be treated, the treatments themselves sometimes causing deaths, which will be counted in the «all cause mortality» , thereby better reflecting the screening reality.
After these results, even less evidence of a benefit
After 23 years, the UK Age Trial results no longer indicate a significant decrease in the number of breast cancer deaths in women screened between the ages of 40 and 49.
The authors of the trial write: «Overall, there was no significant reduction in breast cancer mortality in the intervention group compared with the control group» (p4 penultimate paragraph (ref1 PDF)).
Prior to these final results, analyses that considered the results of all trials already concluded that there was no measurable benefit. Or, at the time, the UK Age was one of those trials that supported screening program.
With these results, we can be even more affirmative in saying that attempting to screen for breast cancer before the age of 50 does not have any tangible benefit.
When should we expect a re-assessment of these results?
Another trial is underway to evaluate the possible benefit of expanding screening to women before age 50 and after age 69: the Age X Trial. Its results are not expected before 2026.
Controversy in Great Britain
The publication of these results caused a great deal of controversy in Great Britain. Not because they were challenged or questioned, because the trial is of a good methodological level.
But because the authors [iii], no doubt disappointed with the results, tried to hide their negative character by insisting on the results obtained at the end of a 10-year follow-up and not at the end of the 23-year follow-up.
Some popular media journalists were thus prompted to write that the results were in favor of the effectiveness of the screening, hence the controversy [iv].
Read more about:
[i] Duffy SW et coll. "Effects of mammogrpahic screening from age 40 years on breast cancer mortality (UK Age trial) : final results of a randomised, controlled trial" The Lancet Oncology online. 12 août 2020. Website www.thelancet.com/oncology. Doi : 10.1016/S1470-2045(20)30398-3
Document PDF of the study
[ii] Nelson HD et coll. "Screening for Breast Cancer: A Systematic Review to Update the 2009 U.S. Preventive Services Task Force" Recommendation. Evidence Synthesis No. 124. AHRQ Publication No. 14-05201-EF-1. Rockville, MD: Agency for Healthcare Research and Quality; 2016.
See particularly table 28 page 128.
[iii] Who are the authors of this study?
They are Duffy’s team and his staff from Queen Mary University in London.
Dr.Duffy is already well known in the world of screening, as he is one of the oldest pioneers in the promotion of screening and has published several studies seeking to quantify overdiagnosis, most often at its lowest range, according to him from 1 to 10%. (Overdiagnosis in mammographic screening for breast cancer in Europe: a literature review. Puliti D, Duffy SW, Miccinesi G, by Koning H, Lynge E, Zappa M and the EUROSCREEN Working Group. J Med Screen 2012;19 Suppl1:42-56.)
This was a review of studies, and this work had been highly controversial as a source of multiple rather crude biases. In their analysis the authors, Duffy and Puliti, had deliberately excluded many reference studies, Zahl’s in 2008 and Junod’s in 2011( Junod B, Zahl P-H, Kaplan Rm, Olsen J, Greenland S. An investigation of the apparent breast cancer epidemic in France: screening and incidence trends in birth cohorts. BmC Cancer. 2011 Sep 21;11(1):401.)
The Prescrire Review in 2006, after a careful analysis, as well as the exhaustive analysis made by Professor Autier, and many others even more recent, currently conclude that the rate of overdiagnosis may be between 30 and 50%.
Revue Prescrire :
*Dépistage des cancers du sein par mammographie Deuxième partie Comparaisons non randomisées : résultats voisins de ceux des essais randomisés. Rev Prescrire. 2014 Nov;34(373):842–6.
*Dépistage des cancers du sein par mammographie Première partie Essais randomisés : diminution de la mortalité par cancer du sein d’ampleur incertaine, au mieux modeste. Rev Prescrire. 2014 Nov;34(373):837–41.
*Dépistage des cancers du sein par mammographies Troisième partie Diagnostics par excès : e et indésirable insidieux du dépistage. Rev Prescrire. 35(376):111–8.
How screening promoters try to justify the alleged success:
In the screening group, after 10 years, for every 10,000 women in the screening group, there were 14 deaths from breast cancer, while for every 10,000 women in the control group, there were 20 deaths.This relatively small benefit (including adverse effects of screening) was statistically significant. It is this difference that the authors rely on to argue that there could be a benefit, which they proclaim as “25%” (20-14 / 20 = 25%).
However, at the end of the follow-up, after 23 years, for 10,000 women in the screening group, there were 39 deaths from breast cancer, while for 10,000 women in the control group, there were 44. This time the difference is not statistically significant. And in relative terms it is 44 – 39 / 44 = 11%.
Above all, the important thing to remember: no significant difference in all-cause mortality was found between the two groups, neither after 10 years, nor at the end of the follow-up, after 23 years
At the end of the trial, there were 650 deaths per 10,000 women in the screening group and 648 deaths per 10,000 women in the control group.