Abstract Background: Epithelial atypia in the breast is most frequently detected through breast screening. Internationally, guidelines differ between advising diagnostic surgical excision of screen-detected atypia versus vacuum-assisted excision (VAE), often with annual follow up surveillance mammography. Prospective data with detailed follow-up and outcomes for subsequent cancer events is lacking. We evaluated a large prospective cohort of screen-detected atypia (Sloane Project) and an expert consensus meeting to develop management recommendations. Methods: We undertook an observational analysis of women with screen-detected epithelial atypia (Atypical Ductal Hyperplasia (ADH), Flat Epithelial Atypia (FEA), Lobular In Situ Neoplasia (LISN), or mixed atypia) reported to the National Health Service (NHS) Breast Screening Programme (BSP) between 01/04/03 and 20/06/18. We estimated cumulative incidence of invasive cancers per 1000 women within 1 and 2 subsequent screening rounds using cumulative incidence functions and repeated these by women’s age, type of atypia and year of diagnosis. Flexible parametric modelling produced hazard ratios (HR). The results were presented at a consensus meeting with clinical experts, patient representation, national decision-makers and commissioners from the NHS in England. A consensus on management recommendations was reached. Results: Among 3238 women (19088 person-years of follow up) with screen-detected atypia, diagnoses increased from 2010 (n=119) to 2015 (n=502). This coincided with the introduction of digital mammography into English screening centres from 2010 and changes in biopsy technique. A total of 141 invasive breast cancers were detected up to December 2018. Cumulative incidence of invasive breast cancer per 1000 women with atypia (ADH, FEA, LISN or mixed) was 14.2 (95% CI 10.3, 19.1) and 45.0 (36.3, 55.1) at 3- and 6-years post atypia based on 40 and 94 cancers, respectively. Women diagnosed with atypia during 2013-2018 had an invasive cancer detected less frequently within 3 years than women with an atypia diagnosis between 2003-2007 or 2008-2012 (6.0 invasive cancers (3.1, 10.9) per 1000 women in 2013-2018 Vs 24.3 (13.7, 40.1) and 24.6 (14.9, 38.3) in 2003-2007 and 2008-2012), suggesting that atypia diagnosis may increasingly contribute to ‘overdiagnosis’. Cancers detected were similar to the general screening population regarding grade, size and nodal involvement. Ipsilateral and contralateral cancers were detected in equal numbers, supporting the concept that atypia represents a risk factor rather than precursor of invasive cancer in the short term. Results by age and type of atypia did not suggest a need for risk-stratified management. Considering subsequent invasive cancer, VAE appeared to be as safe as diagnostic surgical excision of atypia (HR 0.75 (0.45, 1.25)). The expert consensus considered that increased screening of women with screen-detected atypia over and above routine 3-yearly screening was not beneficial, given the overall low numbers of cancers at three years. This applied to all types of atypia and all women who received digital mammography and diagnosis of atypia on core or vacuum-assisted biopsy followed by VAE within a quality assured NHS BSP. Conclusions: Breast atypia represents a risk factor for invasive cancer and does not necessitate surgical excision. Changes to mammography (digital vs plain film) and biopsy techniques (gauge of needle and use of vacuum-assistance) may lead to detection of atypia which is then ‘over-managed’. Current recommendations for annual mammography for 5 years after a diagnosis of epithelial atypia may be overly cautious. We suggest that women with screen detected atypia should be followed up with routine screening appointments. Citation Format: Karoline Freeman, David Jenkinson, Karen Clements, Matthew Wallis, Sarah E Pinder, Elena Provenzano, Hilary Stobart, Nigel Stallard, Olive Kearins, Nisha Sharma, Abeer Shaaban, Cliona Kirwan, Bridget Hilton, Alastair Thompson, Sian Taylor-Phillips. Management guidelines for epithelial atypia diagnosed in breast screening, based on an analysis of a large prospective cohort study (Sloane Project) [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-17-11.
Read full abstract