EU-TOPIA

EU-TOPIA is a micro-simulation research project to estimate the effects of screening programmes for breast, cervical and colorectal cancer in Europe.

One of the goals of the project has been to create a model that, on the basis of figures from the different countries, can calculate the outcomes of a screening programme in relation to the proportion of women who are called in for follow-up examinations, detection of breast cancer, reduced breast cancer mortality and overdiagnosis.

The Cancer Registry of Norway has sent data from the Mammography Programme for this purpose, and the results are shown below in Tables 1 and 2.

Among other things, the results give an indication of what we can gain and lose if the Mammography Programme changes in relation to how it is implemented today, where the age group 50 to 69 years is invited every other year. 

What has been done

The model is based on women aged 40-100 years who are followed from 2020 to 2050. This means that the women are followed from 2020 until they have a simulated date of death, either from breast cancer or from other causes, and until 2050.

In addition to calculating the effect of the Mammography Programme as it is today, the model is used for simulations that show what happens to different outcome parameters if the screening interval is increased or decreased, or if the age limits are changed by including younger and older people in the target group.

Information and results from the Mammography Programme have been incorporated into the model, so that the estimates apply specifically to Norway.

Findings

The calculations are based on the average attendance for the Mammography Programme, which is 75%. This will lead to 7 533 000 screening examinations in the period 2020 to 2050 if women 50-69 years are in the target group and a two-year screening interval is used. It is also estimated that 44 000 breast cancer tumours will be diagnosed in the Norwegian Mammography Programme (9000 precursors to breast cancer (Ductal Carcinoma in Situ - DCIS) and 35 000 invasive cases of breast cancer) (see Table 1 below).

In addition, it is estimated that 90,000 women aged 40-100 years will be diagnosed with clinically invasive breast cancer and 9,000 will be diagnosed with DCIS.

217,000 women will experience a false positive screening examination. This means that they are summoned to the hospital for supplementary examinations after a screening test, but the follow-up examination concludes that the result is negative, i.e. that it is not breast cancer.

Table 1 also shows that by extending the screening programme at both ends of the current age group – i.e. to 45-74 years – it is estimated that 11 170 000 screening examinations would be conducted (yellow row) and that we would have had 61 000 (49 000 +12 000) cases of breast cancer detected in the screening programme and 86 000 (78 000 +8 000) that were clinically detected. 448,000 women would have received a false positive screening result.

The incidence of breast cancer increases with age. This may explain a higher number of cancer cases among older than younger women. It is natural that the detection of cancer cases increases the shorter the interval, since there are more opportunities to find the tumour than, for example, a three-year interval.

The number of false positive screening results is highest for the 50-69 age group if the screening interval is one year (n = 454 000). Again, it is the frequency of screening examinations that is probably decisive, but we also know that younger women have a higher risk of false positive screening results as they are more frequently recalled due to suspicious findings in denser and more «unclear» breasts, compared to older women.

In addition, earlier mammograms are used for comparison in screening interpretation. More younger than older women come for screening for the first time and thus do not have old images for comparison.

The group estimated to have the fewest false positive screening tests was 50-69 years, screened every three years (145 000 false positive screening results).

While the incidence of false positive screening results was estimated among the women who attend screening, the calculation of breast cancer mortality is based on the effect of screening versus not screening.

This means that the entire age group is included, regardless of whether they have been screened or not.

If one again takes into account the population used in the simulations, women aged 40-100 years and screening programmes with a 2-year interval for women 50-69 years, the blue row in Table 2 shows that estimated breast cancer mortality is 21.2% lower if screening is offered versus a scenario without screening.

Based on the models, the EU-TOPIA group has estimated overdiagnosis in the current programme to be 6.0% – i.e. 3 000 of the 44 000 women who were diagnosed with breast cancer were overdiagnosed, i.e. that the tumour is slow-growing and probably would not have caused symptoms in the woman's lifetime.

The calculations also showed that for every prevented death from breast cancer in this group, 22 women would have experienced a false positive screening result.

Table 2 also shows that the estimated reduction in breast cancer mortality is 25.0% as a result of screening if the age group is extended at both ends, to 45-74 years, and the same two-year screening interval is retained (yellow row in Table 2). Overdiagnosis was estimated to be 6.9% – i.e. 4 000 of the 61 000 women who were diagnosed with breast cancer may have been overdiagnosed. For every death prevented, it was estimated that 39 women would experience a false positive screening examination.

More on what's been done

The project has been based on results from European screening programmes, cancer incidence, stage at diagnosis, treatment regimens and death, and on this basis simulated effects on disease-specific death, overdiagnosis and false positive screening results as a result of an offer, versus no offer, of mammography screening.

Some get clinically detected breast cancer (outside the Mammography Programme), while others participate in the programme and have breast cancer detected by screening. In the figures of the Mammography Programme, there will be no cases of cancer detected by screening for women under the age of 50 or over 69, and in the scenario that reflects the current programme, a 2-year screening interval has been included.

Different scenarios, such as annual versus three-year screening intervals, will mean that the number of screening examinations changes for the population.

For example, annual screening will mean 20 examinations for the 50-69-year-olds, and the three-year screening interval will mean 6-7 screening examinations. Including 45-69-year-olds in an annual screening programme will entail offering 25 examinations for the individual woman, while a two-year interval will mean 12-13 examinations. The number of surveys will naturally affect the effect estimates.