Jacky Luiten

128 | Chapter 9 General discussion and future perspectives The detection of breast cancer has come a long way since the days of only diagnosing palpable masses. The introduction and widespread use of mammography screening has led to the detection of asymptomatic, nonpalpable disease. The improvement of screening techniques resulted in the detection of even smaller lesions, which has accelerated the shift from breast amputation to breast conserving surgery. And most important, the detection of breast abnormalities at an earlier stage has contributed to the decrease in the mortality rates that are observed in countries in which mammographic screening has been implemented. 1 However, the improved digital screening techniques also resulted in an increased detection of DCIS and other pre ‐ malignant abnormalities. 2,3 These subclinical lesions can be divided into progressive and non ‐ progressive lesions. 4 Progressive subclinical lesions have the potential to present as clinical cancers in a person’s lifetime, while non ‐ progressive subclinical lesions remain subclinical or might even regress. 4 The detection of these non ‐ progressive subclinical lesions leads to overdiagnosis, which is defined as the detection of cancer by screening that would never cause symptoms or harms in the absence of screening. 4,5 Overdiagnosis is an unintended but unavoidable harm of screening mammography. The women who may benefit from early detection and treatment are those with progressive subclinical lesions. Screening mammography mostly detects high grade DCIS [ this thesis, chapter 2 ]. 3 It is postulated that, in case DCIS is progressive, low grade DCIS will progress to low grade invasive carcinoma over a long time period, whereas high grade DCIS will more likely develop into high grade invasive carcinoma over a shorter time period. 6,7 This assumption implies that the detection and subsequent treatment of high grade DCIS will result in a survival benefit. However, it is difficult to judge if the early detection of high grade DCIS by mammography screening confers a true survival benefit or if the lesion is only detected earlier without prolonging the survival time (lead time bias). The lead time of a breast tumor is the length of time from detection by screening until the appearance of clinical symptoms. Not considering lead time will lead to over ‐ optimistic results of screening mammography. As it is currently not possible to identify which subclinical lesions detected at mammography screening are non ‐ progressive and which are progressive, almost all patients with DCIS are offered treatment. This routine treatment of subclinical lesions will obviously result in overtreatment for certain women. There is no consensus in the literature about the estimated number of breast cancer deaths