Abstract P2-04-03: Can the interval cancer ratio diminish the need for background incidence estimates as a mammography screening programme performance indicator?

Abstract

Abstract Background: Interval cancers (IC) are cancers that are discovered within the interval between two screening examinations, i.e. they present themselves clinically before the next scheduled screening examination. A mammography screening programme that is run in a population with a high breast cancer incidence can have a high interval cancer rate, and still have as protective an effect as a screening programme with a low IC-rate run in a population with a low breast cancer incidence. The effectiveness of a mammography screening programme is therefore normally evaluated by comparing the IC-rate to the expected breast cancer incidence without screening, i.e. the proportional interval cancer rate. The expected breast cancer incidence in absence of screening, or background incidence is, however, difficult to approximate, as the introduction of a screening program makes it difficult or impossible to find a comparable population group that is not screened. We propose to use a new measure, the number of interval cancers out of the total number of cancers, called the interval cancer ratio, as an alternative measure for the burden of interval cancers in an organized mammography screening programme. We seek to validate this new measure by comparing it to the old measure, the proportional interval cancer rate. Method: We did a systematic review and included studies: 1) covering a service screening programme, 2) with inclusion of women aged 50-69 years, 3) that contained observed data, papers based on modeling only were excluded, 4) that contained data on number of interval cancers and number of women screened, or on interval cancers rate, and 5) that contained information on estimated breast cancer incidence rate of the background population. This resulted in 6 papers describing 12 mammography screening programmes. Results: The IC- ratio in studies including only initial screens, varied from 0.13 to 0.32 while the proportional interval cancer rate varied from 0.24 to 0.55 in the same studies. In studies including only subsequent screens the IC-ratio varied from 0.21 to 0.47 with a proportional interval cancer rate of 0.26 to 0.67. One study with mixed initial and subsequent screens with a majority of initial screens had an IC-ratio of 0.18, whereas the proportional interval cancer rate was 0.33. In the study with a majority of subsequent screens had an IC-ratio was 0.32 and the proportional interval cancer rate was 0.67. There is a strong linear relationship between the interval cancer ratio and the proportional interval cancer rate in initial screens. In subsequent screens the trend is linear but is less obvious. Conclusion: Our main aim with this study was to propose and validate the interval cancer ratio as an alternative measure for the burden of interval cancers. The new measure seems to capture the burden of interval cancers just as well or better than the old measure, without the need for increasingly difficult estimations of background incidence, which makes it a more accessible tool when evaluating mammography screening programme performance. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-04-03.