Mammography in the under 50s - 'Age' trial results

Team Leader: Dr Sue Moss

Location: Sir Richard Doll Building, Sutton

Section: Section of Epidemiology

'Age' Trial Results at 10 years' follow-up

The first results on observed breast cancer mortality from this trial have now been published :

Effect of mammographic screening from age 40 years on breast cancer mortality at 10 years’ follow-up: a randomised controlled trial

Sue Moss, Howard Cuckle, Louise Johns, Michael Waller, Lynda Bobrow, for the Trial Management Group, LANCET 2006: 368: 2053-60

In summary :

• The ‘Age’ trial is a randomised controlled trial, designed to look at the effect of inviting women for breast screening by mammography, from the age of 40

• Invited women were offered annual screening, as research suggests that screening may need to be more frequent in women under age 50 to produce a benefit

• The rate of death from breast cancer in those women invited for screening (the intervention arm) has been compared with that in those women not invited (the control arm)

• All women in both arms are eligible for the NHS Breast Screening Programme which invites women for the first time between the ages of 50 and 52

• A total of 160,921 women were randomised between 1991 and 1997, of whom 53,884 were invited for screening

• We have studied breast cancer deaths occurring up to the end of 2004, at which time there was an average of 10.7 years of follow up from the date women were randomised

• At this time the observed reduction in mortality from breast cancer in the intervention arm compared with the control arm was 17%, equivalent to an absolute reduction of 0.4 deaths per 1000 women invited, but this did not reach statistical significance

• The trial was designed to have an 80% chance of a statistically significant result if being invited for screening reduced breast cancer deaths by 20% at 10 years of follow up. However the number of women randomised was less than the planned 190,000, diminishing the chance of a statistically significant effect

• In women attending for at least their first screen, the reduction (adjusted for different background risk in attenders and non-attenders) was 24%, or an absolute reduction of 0.56 per 1000 women screened. Again this was not statistically significant. The possibility that these results are due to chance cannot be excluded

• We have combined the results with those from other trials which have included women under the age of 50. The result is consistent with previous meta-analyses showing a 16% reduction in breast cancer deaths in women invited to screening. However in other trials (although not the present one) some women screened will have entered close to age 50, and the effect of screening beyond age 50 cannot be excluded

• Even if screening is effective in this age group, the absolute benefit will be less than in older women, and more frequent screening is likely to be necessary

• There are disadvantages of screening, including the risk of false positive screen and resultant anxiety and unnecessary follow up, and the possible harmful effects of radiation

• Of those women attending for regular screening, 23% had at least one ‘false positive’ screen, resulting in a recall for further assessment at which no cancer was detailed. This percentage is higher than that estimated for women aged >50

• Further follow up of the trial is continuing. Meanwhile it is important for potential advantages and disadvantages of screening to be carefully explained to women