Screen‐detected breast cancer compared to symptomatic presentation: An analysis of surgical treatment and end‐points of effective mammographic screening

Publisher Website
Google Scholar
Abstract
Background: Mammographic screening has been shown to reduce mortality from breast cancer and to offer more opportunity for breast conservation surgery (BCS). The minimum standards (or surrogate end‐points) that need to be achieved by a screening programme if it is to reduce mortality have been derived from the Two County Study. Three surrogate end‐points that can be used to gauge the quality of the screening service are that 50% of the identified infiltrating cancers should be < 15 mm; at least 30% of grade 3 cancers should be < 15 mm; and 70% of screen‐detected cancers should have a negative axillary dissection. The present study assesses these end‐points of effective screening in an urban population referred to The Strathfield Breast Centre (TSBC). The screening end‐points and surgical treatment of one group of women referred with a BreastScreen New South Wales (NSW)‐detected breast cancer (screen group) were compared to all the other, mostly symptomatic, breast cancer referrals (symptom group). The problems with the current pattern of acceptance of mammographic screening in TSBC’s referral area are discussed. Methods: A prospective non‐randomized study was done via analysis of the prospective database at The Strathfield Breast Centre (TSBC). Results: There were 224 women in the screen group and 657 women in the symptom group. The mean tumour size was 18.1 mm in the screen group and 22.1 mm in the symptom group. There were significantly more small invasive cancers (< 15 mm) in the screen group (58%) compared with the symptom group (33%; P < 0.001). In the screen group there were more low‐grade tumours but 30% of grade 3 tumours were < 15 mm compared with 16% in the symptom group (P = 0.009). In patients with invasive cancers who underwent axillary dissection, there was a significant difference in axillary node negativity, being 72% in the screen group and 59% in the symptom group (P = 0.003). In the screen group 64% of women had BCS compared with 51% in the symptom group (P = 0.002). Conclusions: These end‐points of effective mammographic screening were met in the BreastScreen NSW group of women who were referred to TSBC despite the biases involved which could lessen the effectiveness of the screening programme. This crudely translated into a significant reduction in breast cancer mortality but selection and lead time bias has to be taken into account in evaluation of these data. There was a significantly greater chance of BCS in the screen group.