Radiological and epidemiological aspects of mammographic screening

Abstract

Breast cancer is the most frequent cancer and the most frequent cause of cancer death in women worldwide. Service screening programs have been implemented in many countries in the western world, with the intention to reduce breast cancer mortality by detecting breast cancer in an early and curable stage. The Norwegian Breast Cancer Screening Program (NBCSP) has been nationwide since 2005 and offers biennial screening to women aged 50-69 years.

This thesis includes three papers focusing on quality and attendance in the NBCSP, based on results from the counties of Møre og Romsdal and Vestfold.

When reviewing prior mammograms from cases with screen-detected cancer, or cancer detected in the interval between two screening rounds, some of the cancers will be visible as pathological finding on the previous mammograms and might be classified as “missed” cases. The cancers cannot be regarded as missed in legal terms, but information about the extent and the mammographic features of such cases provides knowledge that might improve the quality of screening interpretation in the future. This thesis includes two review-papers of 130 interval and 209 screen-detected breast cancer cases from screen-film mammography (SFM) and full-field digital mammography (FFDM). The percentage of missed interval cancer was 30% at SFM and 33% at FFDM, whereas for screen-detected cancer the percentages were 21% and 20% respectively. The differences were not statistically significant. Mammographic features in cases missed at FFDM tended to differ from cases missed at SFM by a higher percentage of asymmetry and a lower percentage of calcifications. The implementation of FFDM as a new technique in mammographic screening has thus not reduced the challenge of missed cancer, but probably altered the type of mammographic features that is difficult to perceive and/or interpret.

The third paper deals with asymptomatic breast cancer outside an organized screening program. The attendance rate in Møre og Romsdal County is lower than the average for the NBCSP with 72% compared with 76% nationwide. It is expected that breast cancer cases diagnosed outside the organized screening program has less favorable prognosis. The low attendance rate has thus raised concern as to whether the program will be able to achieve its intended effect on reducing breast cancer mortality. Furthermore, the rate of interval cancers has been higher than recommended in the guidelines. We have analyzed how all 50-70 year old women in Møre og Romsdal got their breast cancer diagnosis during the period 2002-2008. We found that 20% of the interval cancers and 32% of the breast cancers in non participants were asymptomatic with comparable prognostic factors to screen-detected cancers in the program. The most common detection method was opportunistic screening at private institutes. Individual data regarding the detection method for all breast cancer cases are thus needed in order to fully evaluate the effectiveness of mammography as a screening test in Norway. Lack of such information might lead to an underestimation of the mortality reduction attributable to attending the screening mammography program.