First International Consensus Conference on lesions of uncertain malignant potential in the breast (B3 lesions)

The purpose of this study is to obtain a consensus for the therapy of B3 lesions. The first International Consensus Conference on lesions of uncertain malignant potential in the breast (B3 lesions) including atypical ductal hyperplasia (ADH), flat epithelial atypia (FEA), classical lobular neoplasia (LN), papillary lesions (PL), benign phyllodes tumors (PT), and radial scars (RS) took place in January 2016 in Zurich, Switzerland organized by the International Breast Ultrasound School and the Swiss Minimally Invasive Breast Biopsy group—a subgroup of the Swiss Society of Senology. Consensus recommendations for the management and follow-up surveillance of these B3 lesions were developed and areas of research priorities were identified. The consensus recommendation for FEA, LN, PL, and RS diagnosed on core needle biopsy or vacuum-assisted biopsy (VAB) is to therapeutically excise the lesion seen on imaging by VAB and no longer by open surgery, with follow-up surveillance imaging for 5 years. The consensus recommendation for ADH and PT is, with some exceptions, therapeutic first-line open surgical excision. Minimally invasive management of selected B3 lesions with therapeutic VAB is acceptable as an alternative to first-line surgical excision.


Introduction
Breast lesions classified as lesions of uncertain malignant potential (B3) are a heterogeneous group of abnormalities with a borderline histological spectrum, and a variable but low risk of associated malignancy [1]. They encompass a & Christoph J. Rageth c.rageth@brust-zentrum.ch spectrum of histological diagnoses including atypical ductal hyperplasia (ADH), flat epithelial atypia (FEA), classical lobular neoplasia (LN), papillary lesions (PL), benign phyllodes tumors (PT), and radial scars (RS), each with variable rates of upgrade or long-term increased risk of breast cancer [2]. Histological diagnosis of a B3 lesion is made by either core needle biopsy (CNB) mostly using a 14G spring-loaded CNB or by vacuum-assisted biopsy (VAB) using a 7G-11G device either under ultrasound, stereotactic, or MRI guidance following informed consent and local anesthetic. Occasionally it is an incidental finding on a specimen which has been excised surgically. Between 4 and 9 % of all CNBs are classified as B3 lesions with numbers increasing due to advances in diagnostic imaging such as highly sensitive MRI scanning and interventional techniques such as VAB [3]. However, the positive predictive value for malignancy has been falling (from 29 to 10 %) [4,5]. Management of B3 lesions provides a challenge to the multidisciplinary team as diagnostic surgical excision is no longer the only available treatment. Minimally invasive breast biopsy, or VAB, facilitates removal of larger volumes of tissue than a CNB equivalent to a small wide local excision and allows the same diagnostic accuracy as open surgery [6]. For many B3 lesions, instead of surgical excision, VAB may be sufficient for therapeutic excision which would benefit the patient and save on healthcare costs by obviating the need for surgery [7].
The evidence base for appropriate management of B3 lesions in the breast is limited. Practice varies greatly from country to country. This article provides a review of the literature for the six different B3 lesions documented including the analysis of 2 large Swiss B3 histology databases followed by consensus recommendations by an expert panel taken after a voting by the participants of the symposium organized by the International Breast Ultrasound School (IBUS) and the Swiss MIBB group-a subgroup of the Swiss Society of Senology in January 2016, in Zurich, Switzerland.

Methodology
The first International Consensus Conference on lesions of uncertain malignant potential (B3) was held with international experts as part of the IBUS seminar in January 2016 in Zurich. These meetings have been held bi-annually since 2001 with discussion of therapeutic management options for B3 lesions. The meeting in January 2016 had 90 participants with the multidisciplinary expert panel (all the aforementioned authors) comprising nine radiologists, two pathologists, one surgeon, and three gynecologists. Each of the B3 lesions was discussed in turn with reference to the published literature and the analysis of the MIBB [8] working group database (histology from 22,072 VABs).
A set of recommendations for the management of B3 breast lesions was prepared building on the current practice of the Swiss MIBB working group. Recommendations for management of B3 breast lesions following histological diagnosis were either: (i) surveillance (defined as 6 monthly or yearly mammography and/or ultrasound, depending on the imaging findings) [9], (ii) therapeutic VAB excision, or (iii) therapeutic open surgical excision. All participants at the Consensus Conference were invited to vote on all recommendations and 50 of the 90 participants decided to. 27 (57 %) were radiologists, 2 (4 %) pathologists, 2 (4 %) surgeons, and 16 (34 %) gynecologists. Nearly two-thirds of those voting had more than 10 years' experience in breast disease diagnosis and management.
There were 3344 ''pure'' B3 lesions in the MIBB database (15 % of all lesions). Following presentations of each B3 lesion in detail reviewing the published literature, three questions were asked in turn: Q1. If a CNB returned a B3 lesion on histology, should the lesion be therapeutically excised?
Q2. If so, should it be excised therapeutically using VAB?
Q3. If the VAB returned a B3 lesion on histology and if the lesion was completely removed on imaging, is surveillance acceptable or should a repeat VAB or surgical excision be performed?
A panel discussion followed the voting and consensus recommendations were agreed for the management of each B3 lesion. Table 1 illustrates the number of cases in each B3 lesion category that underwent therapeutic surgical excision compared to those that did not following VAB. Table 2 illustrates the upgrade rate to invasive malignancy for each B3 lesion in cases that underwent therapeutic open surgical excision following VAB. Table 3 documents the voting results for each of the B3 lesions and Table 4 shows the overall consensus recommendations for the management of B3 lesions.
In studies analyzing patients on surveillance without surgical treatment following a VAB diagnosis of ADH, long-term upgrade rates to invasive breast cancer of 3-8 % are reported [19,26]. After therapeutic surgical excision of ADH, patients had a fourfold increased risk of developing breast cancer in either breast with a cumulative incidence of 30 % in 25 years [10,11,38,39]. Currently, there is very little data to indicate that lesions smaller than 6 mm completely excised by VAB with less than 2 foci of ADH may safely avoid surgery [14,19,21,26,40] (Table 3).

Consensus recommendation
A lesion containing ADH which is visible on imaging should undergo therapeutic open surgical excision. If a unifocal ADH lesion 1 has been completely removed by VAB, surveillance is justified. Otherwise open surgery is still recommended (Table 4).

Flat epithelial atypia
FEA is defined as a neoplastic proliferation of the terminal ductulo-lobular units (TDLU) by a few layers of cells with low-grade (monomorphic) atypia [2,41,42]. Histopathology of FEA lesions encompasses the proliferation of round and uniform cells (defined as low-grade atypia) exhibiting inconspicuous nuclei [2,41,42]. There is often associated calcification. An FEA lesion lacks secondary architecture such as roman bridges or cellular tufts and exhibits a characteristic immunophenotype of negative low-molecular weight cytokeratins and high regulation of estrogen receptors [2,41,42]. The mammographic appearance of FEA is mostly seen as microcalcifications which are irregular and branching with accompanying marked duct dilatation [2,41,42]. Coexisting lesions both on imaging and on histopathology encompass classical LN, other benign columnar cell lesions, low-grade intraductal proliferations such as ADH/DCIS, or tubular carcinoma [2,41,42]. The risk of developing breast cancer with a diagnosis of FEA is estimated at 1-2 times higher than those without FEA [2,41,42]. ADH and DCIS are the most frequent pathologies found following surgical excision with their incidence varying from 0 to 40 %. Underestimation rates are between 0-20 % if FEA is diagnosed on core biopsy and are very similar if diagnosed on VAB (0-21 %) [43][44][45][46][47][48] (Table 3).

Consensus recommendation
A lesion containing FEA, which is visible on imaging should undergo therapeutic excision with VAB. Thereafter surveillance is justified (Table 4).   [2,12]. The incidence of classical LN has been increasing and varies from 0.5 to 4 %. It can occur at all ages but predominantly in premenopausal women. Most lesions present incidentally without any palpable mass and less than half of classical LN lesions have associated calcification. Published data on the risk of developing breast cancer after diagnosis on CNB or VAB show, a relative risk of 1-2 % per year, 15-17 % after 15 years, and 35 % after 35 years with relatively equal rates of ipsi-and contralateral breast cancer (8.7 and 6.7 %, respectively) [2,[49][50][51][52][53]. The upgrade rate after classical LN diagnosis on CB or VAB is variable in the literature, ranging from 0 to 50 % which can at least partially be explained by variation in study design and inconsistent use of ALH, LCIS, and LN nomenclatures [2,41]. In one study, underestimation was found to be 4 % in classical LN cases when LN was an incidental finding and 18 %, when LN represented the radiologic targets by D'Alfonso et al. [50]. Higher upgrade rates were associated with cases that demonstrated mass lesions and calcification on imaging or with radiopathological discordance. Lower underestimation rates were detected in classical LN cases, where no residual calcification was found after biopsy, calcification was incidental, and there was complete concordance between histological and imaging findings [2,41,50].
The WHO recommends surgical excision after classical LN diagnosis on CNB or VAB if there is another B3 lesion present, if another coexisting lesion warrants excision alone, if there is a mass lesion on imaging, or in any case of radiopathological discordance [2,41] (Table 3).

Consensus recommendation
A lesion containing classical LN lesion, which is visible on imaging should undergo therapeutic excision with VAB. Thereafter surveillance is justified (Table 4).

Papillary lesion
PLs represent up to 5 % of all biopsied breast lesions [54][55][56][57]. The term PL comprises a heterogeneous group of epithelial lesions such as intraductal papilloma, intraductal papilloma with ADH, intraductal papilloma with DCIS, papillary DCIS, encapsulated papillary carcinoma, solid papillary carcinoma, and invasive papillary carcinoma [2]. PLs demonstrate intra-lesional heterogeneity and can be associated with small foci of ADH or DCIS within the PL or in the adjacent tissue which may be missed by limited sampling with CNB. When describing PL, only PL without atypia should be considered, as a lesion with atypia should be considered within the higher class lesions (e.g., ADH) and offered therapeutic open surgical excision. Upgrade rates after surgical excision of benign papillomata diagnosed following CNB or VAB vary from 0 to 28 % with atypical cells and from 0 to 20 % for invasive cancer [58][59][60]. Generally, understaging of invasive malignancy is reduced if multiple biopsy cores are taken or if a larger biopsy needles are employed such as in VAB. Only one study by Chang et al. evaluated the accuracy of VAB in PL without atypia by open surgical excision following VAB with no upgrade to malignancy but an upgrade of 18.3 % to atypia [61]. Most studies following up VAB excision of PL without atypia did not observe any upgrade to malignancy with at least 2 years of surveillance [60,62,63]. One recorded a minimal underestimation of 1.4 % [64] and another 3.2 % [65].
The upgrade rate to malignancy following VAB in the MIBB database was 7.7 % for PL without atypia which is higher than in the documented literature. One reason might be the fact that the size of the PL was not recorded, implying that some PL might not have been completely removed. Mosier et al. removed only lesions smaller than 15 mm (range 3-15 mm) to ensure the complete removal of the PL [62]. With this approach, they had no upgrades to malignancy after nearly 9 years. Due to difficulties in excluding malignancy with small tissue samples at CNB, heterogeneity of PLs, and an upgrade rate to carcinoma of up to 20 % [58], the current recommendation is to completely remove PL without atypia, either by surgery or VAB [58][59][60].
If a CNB returned PL on histology 100 % of the participants thought the lesion should be excised. 84 % thought therapeutic VAB excision was acceptable and 11 % thought therapeutic open surgical excision should be performed.
If a VAB returned PL on histology 9 % of the participants thought that therapeutic open surgical excision should be performed and 91 % thought that surveillance was adequate (Table 3).

Consensus recommendation
A PL lesion, which is visible on imaging should undergo therapeutic excision with VAB. Thereafter surveillance is justified (Table 4).

Phyllodes tumor
PTs are rare fibroepithelial neoplasms accounting for less than 1 % of primary breast tumors [2,66]. Histologically, they are classified as benign, borderline, and malignant with the first two subtypes categorized as B3 lesions [67]. The majority of PTs are benign, (63-78 %) with borderline PTs diagnosed in 11-30 % of cases [2]. Incidence is highest in women aged 40-51 years [68,69]. Overlapping clinical, radiological, and histopathological features may make differentiation from benign fibroadenomata challenging at times, however accurate preoperative diagnosis is essential to establish the most appropriate therapeutic approach.

Consensus recommendation
A PT lesion, which is visible on imaging should undergo therapeutic open surgical excision with clear margins. If a VAB shows a benign PT, surveillance is justified, while borderline and malignant PTs require re-excision to obtain clear margins (Table 4).

Radial scar
RS or complex sclerosing lesions (CSL) of the breast are characterized by a stellate-like distortion. The nomenclature depends on the size of the lesion which is defined as radial scar if the focus is less than 1 cm or complex sclerosing lesion if over 1 cm [2,41]. Histopathology of a RS/CSL involves a stellate-like elastosis with or without the presence of associated lobulocentric cysts, usual ductal hyperplasia, adenosis, and microcalcifications. The adenosis may evolve the elastic fibers resulting in entrapped glands, which may mimic a highly differentiated neoplastic glandular proliferation [2,41]. On mammography, RS/CSL mostly appear as a stellate lesion which mimics an invasive carcinoma. The incidence is variable being 4-9 % in population-based pathology databases, but being significantly higher, up to 63 % in sole pathology literature [2,41].

Consensus recommendation
A RS/CSL lesion, which is visible on imaging should undergo therapeutic excision with VAB. Thereafter surveillance is justified (Table 4).

Discussion
The expert consensus panel and participants agreed that for most of the B3 lesions (except for ADH and PT), surgery can be avoided and therapeutic excision with VAB of a lesion which is visible on imaging is an acceptable alternative. However, as data are lacking at present the panel still recommends open surgery in cases of ADH. As more data on the minimally invasive conservative management of B3 lesions become available, a more conservative approach may also be justified in cases of ADH. The outcome from this consensus meeting is a progressive move forward to a more conservative approach to managing these lesions in which open surgery can potentially be avoided. Studies following a diagnosis of low-grade DCIS have shown excellent survival rates of more than 98 % at ten years after diagnosis without surgery [91,92] which have prompted randomized phase III trials for surgery versus no surgery in low-and intermediate-grade DCIS [93,94]. Therefore it is becoming clear, that it is even more reasonable to try to avoid unnecessary open surgery or overtreatment in some women. It is important to emphasize that these recommendations cannot exclude a false-negative diagnosis in every individual patient and each case should be discussed on an individual basis with a multidisciplinary team taking into account the imaging features, lesion size, practicality and technical feasibility of minimally invasive management, patient demographics, and patient preference.