The aim of this study is to analyse whether immunohistochemistry (IHC) applying a broad set of markers could be used to categorise ductal carcinoma in situ (DCIS) of the breast in distinct subgroups corresponding to the recently defined molecular categories of invasive carcinoma. Immunohistochemistry of pure DCIS cases constructed in tissue arrays was performed with 16 markers (oestrogen receptor (ER), progesterone receptor (PR), androgen receptor (AR), Bcl-2, p53, Her2, insulin-like growth factor receptor, E-cadherin, epithelial membrane antigen (EMA), CA125, keratins 5/6, 14, 19, epidermal growth factor receptor, S100, and CD31). Results in 163 cases were analysed by unsupervised hierarchical clustering. Histological classification was performed by review of whole tissue sections and identified 36 well-, 55 intermediately, and 72 poorly differentiated DCISs. Unsupervised hierarchical cluster analysis categorised DCIS into two major groups that could be further subdivided into subgroups based on the expression of six markers (ER, PR, AR, Bcl-2, p53, and Her2). In the major predominantly ER/Bcl-2-positive (luminal) group, three subgroups (ARpositive (n ¼ 33), AR-negative (n ¼ 40), and mixed (n ¼ 34)) could be identified and included 34 well-differentiated DCISs. Within the major predominantly ER/Bcl-2-negative (nonluminal) group, a Her2-positive subgroup (n ¼ 34) was characterised by 31 poorly differentiated lesions. Eight triple-negative lesions, including one positive for keratin 5/6 and two positive for p53, were encountered. Intermediately differentiated DCIS shared a comparable IHC staining pattern with well-differentiated DCIS that was distinct from poorly differentiated DCIS (Po0.001). Ductal carcinoma in situ could be categorised by IHC into two major groups and five subgroups using six markers. Morphologically, intermediately differentiated DCIS seems to have more biological similarities with well-differentiated lesions as compared to poorly differentiated lesions.

SN biopsy should be considered in women with an initial diagnosis of DCIS on core-needle biopsy who are at risk for invasion; this includes women with a palpable lump, a mass on mammography, and intermediate or poor tumour grade.

Background: Sentinel node biopsy (SNB) is standard in assessing axillary lymph node status in patients with clinically node-negative breast cancer. The 5-year analysis of AMAROS trial showed that if locoregional treatment is advised after a tumor-positive axillary SNB, axillary radiotherapy (ART) is a reasonable alternative for an axillary lymph node dissection (ALND) with less side effects, though follow up was relatively short. Here we present the 10-year follow up data. Methods: From February 2001 to April 2010, patients with primary breast cancer stage cT1-2N0M0 were enrolled in the EORTC phase III non-inferiority AMAROS trial by 34 European sites. Patients were randomized between ALND and ART in case of a tumor-positive SNB. The primary endpoint, axillary recurrence rate (AxR) is now assessed at 10 years in the ITT population using Fine and Gray cumulative incidence method with deaths as competing risks, as well as secondary endpoints: overall survival (OS), distant metastasis free survival (DMFS), second primaries (including cancers other than breast cancers and contralateral DCIS) and locoregional recurrences (LRR). Little extra information beyond 5 years was available concerning Quality of Life and morbidity. Data collection is still ongoing and will be presented later. Results:Of the 4806 patients entered, 1425 patients had a tumor-positive SNB: 744 in the ALND-arm and 681 in the ART-arm, 60% with a macrometastasis. Both treatment-arms achieved a median 10-year follow-up and were comparable regarding age, tumor size, grade, tumor type and adjuvant systemic treatment. In the group who had ALND, the 5-year AxR was 0.41% (95%CI: 0.00;0.88) (4/744) and the 10-year AxR was 0.93% (95%CI:0.18;1.68) (7/744). In the group who had ART, the 5-year AxR was 1.04% (95%CI: 0.27;1.81) (7/681) and the 10-year AxR was 1.82% (95%CI: 0.74;2.94) (11/681) (HR 1.71, 95%CI: 0.67;4.39, p = 0.37). Sensitivity analysis, considering deaths and distant recurrences as competing risks, revealed consistent results. There were no significant differences between treatment arms regarding OS (ALND: 84.6% (95%CI: 81.5;87.1), ART: 81.4% (95%CI: 77.9;84.4), HR 1.17, 95%CI: 0.89;1.52, p= 0.26) and DMFS (ALND: 81.7% (95%CI: 78.5;84.4), ART: 78.2% (95%CI: 74.6;81.3), HR 1.18, 95%CI: 0.92;1.50, p=0.19). Cumulative incidence estimates of 10-year LRR are 3.59% (95%CI: 2.12;5.06) (ALND) versus 4.07% (95%CI: 2.49;5.65) (ART) (p= 0.69). More second primaries were observed after ART: 75/681 (21 contralateral breast) as compared to ALND: 57/744 (11 contralateral breast) (p = 0.035). All results are consistent in the per protocol analysis of patients with a tumor-positive SNB. Conclusion: Axillary recurrence after 10 years in patients with a tumor-positive SNB who were treated with ART is extremely rare and not significantly different from patients who were treated with ALND. OS, DMFS and locoregional control are also comparable. Second primaries including contralateral breast cancers are more frequently encountered after ART, but the difference is still low in absolute numbers. Thus, ART is a safe treatment for breast cancer patients with a tumor-positive SNB. Citation Format: Rutgers EJ, Donker M, Poncet C, Straver ME, Meijnen P, van de Velde CJ, Mansel RE, Blanken C, Orzalesi L, Klinkenbijl JH, van der Mijle HC, Veltkamp SC, van 't Riet M, Albregts M, Marinelli A, Rijna H, Tobon Morales R, Snoj M, Bundred N, Chauvet MP, Merkus JW, Petignat P, Schinagl DA, Coens C, Peric A, Bogaerts J, van Tienhoven G. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer patients: 10 year follow up results of the EORTC AMAROS trial (EORTC 10981/22023) [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr GS4-01.