News Release

Mammograms benefit women up to the age of 75 and 3-yearly screening intervals are best

Peer-Reviewed Publication

ECCO-the European CanCer Organisation

Berlin, Germany: Breast cancer screening is effective, appropriate and reduces deaths from the disease in women aged up to 75 years old according to new research in over 860,000 women aged 70-75 presented today (Friday) at the 6th European Breast Cancer Conference (EBCC-6) in Berlin. In another study of nearly 100,000 women aged 50-62, also presented at EBCC-6 today, researchers found that the best interval for screening was every three years – a finding that counters arguments that women should have mammograms more frequently.

Many countries that run breast cancer screening programmes offer it to women aged between 50 and 70. However, in 1998 in The Netherlands, the programme was extended to women aged up to 75. The results presented today suggest that this is an appropriate upper age limit and saves lives without causing substantial harm by subjecting older women to over-diagnosis and over-treatment.

This is the first study to provide firm support for the idea that offering mammograms to women up to the age of 75 is effective, as, until 1998, very few women were screened at this age.

Mr Jacques Fracheboud, a senior researcher at the Erasmus Medical Center (Rotterdam, The Netherlands), and his colleagues found that from 2003, five years after screening was extended to women aged 70-75, there was a steady decline in deaths from breast cancer among women aged 75-79. By 2006 breast cancer mortality was 29.5% lower than the average for the period 1986-1997 for this age group – a time when breast cancer mortality in women aged 75-79 had remained stable. In 1986-1997 the average was 166 deaths per 100,000 women aged 75-79 and in 2006 it was 117 per 100,000.

It takes some years before a significant effect from screening can be seen, therefore breast cancer mortality was analysed for the ages 75-79.

“That means that women, aged 70-75 at the time that screening was extended to this age group, have become five years older and the reduction in breast cancer mortality shows that the screening has started to have a statistically significant effect,” Mr Fracheboud told a news briefing.

Out of the 7.37 million screening examinations performed between 1998-2006, 862,655 were for women aged 70-75. Among women aged 50-69, 81.2% participated in the screening programme, and 71.9% of women aged 70-75 took part. However, the participation rate for these older women increased significantly during this period, rising from 62.5% in 1998 to 77.6% in 2006.

In the 50-69 age group, 12.8 per 1000 women screened were referred for diagnostic assessment as a result of screening and breast cancer was detected in 4.5 per 1000, giving a positive predictive value (percentage of abnormal mammogram findings that were confirmed as breast cancer) of 36%. In the 70-75 age group, 16.4 per 1000 women screened were referred for diagnostic assessment and breast cancer was detected in 7.8 per 1000, giving a PPV of 47%.

“The difference shows that it is easier to find breast cancer in older women due to their breast tissue being less dense,” said Mr Fracheboud. “But it is not necessarily an argument for continuing screening beyond 75 because many tumours found at this stage are slow growing and may never reach the stage of causing a problem.”

He continued: “The results of our study suggest that screening women aged 70-75 has a strong impact on breast cancer mortality and that it is effective and appropriate up to 75 years. The cost per mammogram is approximately €50 – the same as for younger women. Before the programme was extended up to 75 there was a fear that older women would be more expensive because fewer would participate and it might take more time to screen them due to reasons such as lower mobility. However this seems not to be case, and participation rose among the older women.”

In the second study – the UK Breast Screening Frequency Trial – researchers randomised nearly 100,000 women to have either an annual or a three-yearly mammogram after the date of the first screening to which they were invited after their 50th birthday. In an earlier paper (European Journal of Cancer 2002), the researchers had predicted that annual screening would not result in a further reduction in breast cancer mortality over the three-year interval standard in the UK. The actual mortality rates presented today confirm this to be the case.

After an average follow-up time of over 13 years (the trial started in 1989), there were 373 breast cancer deaths (out of a total of 49,173) among the study group of women invited to attend annual screening, and 374 (out of a total of 50,162) in the control group of women invited to three-yearly screening. Among those women who actually attended the screening (as opposed to the whole group of women, including non-attenders), there were 209 breast cancer deaths in the study group and 231 in the control group. The researchers found no difference in mortality rates when they looked at only those cancers diagnosed during the three-year screening periods.

The absolute risk of dying from breast cancer was statistically insignificant between the two groups, with the control group having an absolute risk of only two per cent more than the study group.

One of the authors of the study, Professor Roger Blamey, a breast surgeon at Nottingham City Hospital (Nottingham, UK), told the news briefing: “There was a lot of criticism of the UK for having a three-yearly interval when breast screening was set up – screeners and advocacy groups said, without evidence, that it was too long an interval. Population screening in the UK costs around €100m a year. These results indicate that our earlier predicted mortality figures were accurate and that there is no evidence in favour of shortening the current three-year screening interval.”

###

Catalogue nos: 5LB & 4LB, Friday 11.00-13.00 hrs CEST (Hall 1)


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.