Introduction

Clinical management of high-risk breast lesions identified and determined to be concordant after image-guided core needle biopsy is variable and often debated. Management can include excisional biopsy to exclude upstaging to ductal carcinoma in situ (DCIS) or invasive cancer, or close imaging follow-up in certain instances. In addition, the diagnosis of a high-risk breast lesion may increase the patient’s risk for the development of breast cancer in the future, prompting discussion of risk factors with implementation of appropriate surveillance regimens and risk reduction methods as appropriate. The myriad of pathology in the high-risk breast lesion spectrum, the complexity of management, and the lack of prospective data-guided decision-making encourage a multidisciplinary approach with close clinical, imaging, and pathology correlation.

High-risk lesions are reportedly identified in 5–9.2% of core needle breast biopsies [1,2,3]. The identification of high-risk breast lesions may increase with continued imaging advances as the transition from film-screen mammography to digital mammography has increased detection rates of high-risk lesions such as atypical ductal hyperplasia (ADH), flat epithelial atypia (FEA), and lobular neoplasia (LN) threefold [4]. There is variation among pathologists regarding core needle breast biopsy diagnosis of lesions with atypia. A recent study evaluating individual pathologists’ interpretations compared to expert consensus-derived reference diagnosis revealed 48% agreement for breast lesions with atypia and 84% agreement for ductal carcinoma in situ (DCIS) [5]. This study further emphasizes the importance of radiologic-pathologic correlation with additional tissue sampling when discordant. With the detection of high-risk lesions, there is opportunity to optimize screening methods and educate the patient on prevention measures; however, there is also a risk that women may undergo unnecessary surgical procedures for benign lesions. To offer the best individualized treatment, it is important for future research to identify the clinical, imaging and pathologic predictors of invasive carcinoma. We will review the management of the following high-risk breast lesions identified on core needle biopsy with the assumption that radiologic-pathologic correlation is concordant: ADH, LN, FEA, radial scars/complex sclerosing lesions, and papillomas.

Atypical Ductal Hyperplasia (ADH) and Lobular Neoplasia

Atypical Ductal Hyperplasia

ADH resembles low-grade DCIS microscopically with atypical epithelial cells partially or completely filling less than two duct spaces or occupying less than 2 mm in maximum dimension [6]. ADH is considered a non-obligate precursor to breast cancer; however, common cytogenetic alterations including losses of 16q and 17p exist among ADH, DCIS, and invasive cancer [7]. ADH is identified in approximately 3–4% of core needle biopsies [8].

The upgrade rate for ADH identified on core needle biopsy following excision ranges from 0 to 56% [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27, 28•, 29,30,31]. When reviewing studies that included greater than 100 excisional biopsies, the upgrade rate ranges from 13 to 51% [8, 12, 13, 16, 19, 21,22,23,24, 26, 27, 28•, 29, 31]. A recent large retrospective study with literature review reported an average upgrade rate for ADH of 23%. [28•]. Given the rates of upgrade at excisional biopsy as well as the quantitative nature of the pathologic diagnosis of ADH versus DCIS, excisional biopsy is recommended when ADH is identified on core needle biopsy to ensure appropriate tissue sampling and to exclude in situ or invasive cancer and is supported by the National Comprehensive Cancer Network guidelines [32]. When excisional biopsy is not performed, close imaging follow-up should be pursued.

Several studies have shown that a diagnosis of ADH increases the patient’s relative risk for the future development of breast cancer ranging from 3.1 to 4.7 and this risk is for either breast [33,34,35]. One study reported the risk of breast cancer was 21% at 20 years for women with atypical hyperplasia and risk was related to number of foci of atypia present [34].

Lobular Neoplasia: Atypical Lobular Hyperplasia & Lobular Carcinoma in Situ

Lobular neoplasia (LN) describes a spectrum of lesions that includes atypical lobular hyperplasia (ALH) and lobular carcinoma in situ (LCIS). Lobular neoplasia is considered a non-obligate precursor for the subsequent development of invasive breast cancer. In lobular neoplasia, a monotonous population of neoplastic cells expand and replace the normal epithelial cells of the acini and intralobular ductules expanding the lobular units [36]. The differentiation between ALH and LCIS is quantitative. More than 50% of the acini of a lobular unit must be distended by neoplastic cells for a diagnosis of LCIS, and distension of less than 50% of the acini of a lobular unit would be designated ALH [37]. Pleomorphic LCIS exhibits cells with a greater degree of nuclear pleomorphism and abundant cytoplasm. Lobular neoplasia can be bilateral and multicentric [38] and most often arises in women 40–50 years of age [36]. ALH, LCIS, and invasive lobular carcinoma have been shown to have loss of heterozygosity and mutations in CDH1, the gene encoding E-cadherin [39].

Multiple retrospective studies evaluating excisional biopsy of ALH have reported upgrade rates ranging from 0 to 67% [10, 12, 28•, 40,41,42,43,44,45,46,47,48,49,50,51] with a reported average rate of 9% [28•]. Upgrade rates from LCIS to DCIS or invasive carcinoma range from 0 to 60% [10, 12, 28•, 40, 43,44,45,46,47,48, 50,51,52,53] with a reported average of 18% [28•]. Practices vary regarding excisional biopsy after LN diagnosis with imaging concordance. The National Comprehensive Cancer Network supports excisional biopsy of ALH or LCIS diagnosed by core needle biopsy when pleomorphic LCIS is present and when there is multifocal or extensive LCIS involving more than 4 terminal duct lobular units, as this has been shown to have an increased risk for invasive cancer at surgical biopsy [32, 48]. Those patients with LN and concordant imaging who do not undergo excisional biopsy are recommended by the NCCN to have a physical exam with mammography with or without ultrasound every 6–12 months for 1 year [32].

Several studies have shown that a diagnosis of ALH increases the relative risk of breast cancer development 3.1–5.9-fold [33,34,35]. Studies have shown a greater increased risk for the future development of breast cancer in women diagnosed with LCIS 6.9–11.0 [54, 55].

Given the increased rates of future breast cancer development in women with ADH, ALH, and LCIS, once an associated malignancy has been excluded, women should be counseled regarding their increased risk for the development of breast cancer and undergo a risk assessment. Women who are determined to have a familial risk for breast cancer should be referred to a genetic professional. Risk reduction strategies should be discussed and may include but are not limited to lifestyle interventions, active surveillance, chemoprevention, and prophylactic mastectomy.

Lifestyle interventions include exercise, maintaining an ideal body mass index, and lowering the use of alcohol. The direct results of these measures on lowering risk for women with atypia have not been evaluated. Active surveillance includes clinical exam and history every 6–12 months and screening mammography with consideration of tomosynthesis to begin upon diagnosis of ADH or LN, but not prior to 30 years of age. A study comparing patients with a diagnosis of ADH, ALH, or LCIS to matched controls revealed no difference in the sensitivity of mammography but showed a lower specificity in the high-risk group [56]. While the role of breast MRI in patients with atypia needs further exploration, patients should undergo risk assessment as the American Cancer Society screening guidelines recommend breast MRI in women with a 20–25% lifetime risk for the development of breast cancer based on family history models, but report there is insufficient evidence to recommend for or against breast MRI in women with atypia or LCIS [57]. Physical exam is recommended by the National Comprehensive Cancer Network every 6–12 months and this is supported by a study in which 13 of 104 reported cancers were detected on physical exam [57].

Chemoprevention has been shown to reduce breast cancer incidence in women with atypical hyperplasia. The National Surgical Adjuvant Breast and Bowel Project (NSABP) P-1 study revealed an 86% reduction in breast cancer incidence in women with atypical hyperplasia after 5 years of therapy with Tamoxifen [58]. Additionally, the Study of Tamoxifen and Raloxifene (STAR) trial showed that raloxifene had similar risk reduction to tamoxifen with decreased toxicity in post-menopausal high-risk women [59]. Women with a diagnosis of LCIS were also represented in the aforementioned trials representing 6% of women in the NSABP P-1 trial and 9% of women in the STAR trial with a greater than 50% risk reduction for both trials in this subset of participants [58, 59]. The use of chemoprevention measures has been reported to lower the breast cancer incidence in women with a diagnosis of atypical hyperplasia and LCIS from 21 to 7.5% at 10 years [60].

Bilateral prophylactic mastectomy is the most-effective method of breast cancer risk reduction and studies suggest a 95% reduction in the risk of breast cancer in women who undergo this procedure [61]. In the absence of additional risk factors, prophylactic mastectomy is rare in women with LCIS or atypia, and a study of a cohort of participants with LCIS only 5% selected this procedure [62].

FEA

FEA is a columnar cell lesion with low-grade cytologic nuclear atypia involving the terminal duct lobular unit without architectural features of ADH or DCIS [63]. The frequent co-existence of columnar cell lesions, including FEA, LCIS, and tubular carcinoma has been termed the “Rosen Triad” prompting close surveillance at biopsy for histologic evidence of the other two entities [64].

Flat epithelial atypia is a rare lesion reported on 1.3% of breast biopsies [28•]. A recent meta-analysis identified 32 studies revealing a range in upgrade rates from 0-42% with an average upgrade rate of 11% [65•]. Studies have shown residual calcifications after core biopsy have been associated with upgrade rate [28•]. The World Health Organization suggests that surveillance may be appropriate for pure FEA in the absence of residual calcifications and presence of pathologic-radiologic concordance [63]. Currently, the Translational Breast Cancer Research Consortium (TBCRC) has a prospective trial evaluating the incidence of adjacent synchronous ipsilateral invasive carcinoma or DCIS in patients diagnosed with FEA by core needle biopsy.

The Nashville Cohort Study evaluated the long-term risk of breast cancer in patients with FEA and showed a similar relative risk for women with columnar cell lesions without atypia and FEA of 1.5 [66]. The Mayo Clinic Cohort study also showed that women with a diagnosis of FEA alone had a similar risk to women with proliferative lesions without atypia [67].

Radial Scar/Complex Sclerosing Lesion

A radial scar is a composed of proliferating tubules radiating from a central fibroelastotic core, and this term is reserved for lesions less than 1 cm. A complex sclerosing lesion is a larger (> 1 cm) radial sclerosing lesion with more complex epithelial elements. Additional proliferative lesions are frequently associated with these lesions and include papillomas, sclerosing adenosis, and usual ductal hyperplasia.

Radial scars have been reported in 1–2% of core needle biopsies [68, 69]. A recent meta-analysis including 20 studies revealed an upgrade rate of 26% for radial scars with atypia and an upgrade rate of 7.5% for radial scars without atypia at excisional biopsy [69]. Incidental radial scars less than 5 mm have been shown to be less likely to be upgraded at excision [69] suggesting a role for imaging follow-up in these patients.

Papilloma

Solitary intraductal papillomas are intraductal growths of arborizing epithelia with fibrovascular stalks. These lesions may be associated with proliferative change or atypia. Central papillomas arise in a large duct and can be associated with clear or bloody nipple discharge. Peripheral papillomas arise in the small ducts and are more often clinically occult and identified on breast imaging.

A recent meta-analysis of 34 studies reported an upgrade rate of 15.7% to DCIS or invasive cancer following excisional biopsy for a papillary lesion identified on core biopsy [70]. The meta-analysis also showed a higher upgrade rate for atypical papillary lesion (36.9%) as compared to benign papillomas (7.0%) [70]. Several studies have reported upgrade rates from 2.3 to 2.7% associated with benign papillomas identified at core needle biopsy [71•, 72]. These studies suggest that close imaging surveillance may be appropriate in patients with a benign papilloma diagnosed at core needle biopsy with radiologic-pathologic concordance and minimal residual mass after core biopsy. Several factors have been associated with upgrade to malignancy at excisional biopsy after core needle biopsy for benign papillary lesions and include the following: > 54 years of age, lesions greater than 1 cm, and patients with ipsilateral breast cancer [73]. Papillary lesions with atypia as previously mentioned have a higher rate of upgrade to malignancy and should undergo surgical excision to exclude a higher-grade lesion. Currently, the TBCRC is evaluating the incidence of adjacent synchronous ipsilateral invasive carcinoma or DCIS in patients with intraductal papilloma without atypia on core needle biopsy in a prospective trial. Benign papillary lesions confer a similar risk for the future development of breast cancer as other proliferative lesions approximately 2-fold [74,75,76,77].

Conclusions

Clinical management of high-risk breast lesions identified and determined to be concordant after image-guided core needle biopsy varies and prospective data are needed to better guide management decisions. High-risk breast lesions including ADH, LCIS, ALH FEA, radial scar, and papillary lesions require close radiologic-pathologic correlation when diagnosed by image-guided breast core needle biopsy. Excisional biopsy can exclude a higher-grade lesion such as DCIS or invasive cancer; however, in certain cases, close observation with follow-up may be appropriate. Additionally, women who have LCIS, ADH, or ALH are at an increased risk for the future development of breast cancer and should undergo risk assessment and discussion of risk reduction measures. Ongoing and future prospective trials may provide data to better guide these management decisions and optimize patient care.