Introduction

The number of breast biopsies performed each year in the USA is estimated to reach 1 million. It has been reported that 30% of them are malignant and 70% are non-malignant [1]. Among non-malignant lesions, there is a specific group called high risk [2]. High-risk lesions are considered breast carcinomas precursors and/or markers for developing carcinomas in the future [3,4,5,6]. These lesions are increasing among reported findings of core needle biopsies [7]. It is estimated that up to 10% of the core needle biopsies performed in women in their 40 s contain high-risk pathology [8]. The high-risk lesions include atypical lobular hyperplasia (ALH), atypical ductal hyperplasia (ADH), radial scar/complex sclerosing lesion, lobular carcinoma in situ (LCIS), flat epithelial atypia (FEA) and papilloma. Although high-risk lesions may have a similar pathology, they are important to be differentiated since they may have a different management (conservative versus surgical). Atypical hyperplasia corresponds to an abnormal proliferation of the breast that does not meet the criteria of ductal carcinoma in situ (DCIS) [9]. Atypical hyperplasia has two distinct entities: ALH and ADH. ALH is a rare, occurring in 1% of women [10]. In addition, the upgrade rate to malignancy for ALH ranges from 12 to 33% with an average of 17%. [11]. This review of ALH discusses the pathology, imaging features and its management with other atypical high-risk breast lesions.

Pathology

Lobular neoplasia

ALH is included within the classification of lobular neoplasia (LN), which also includes classic and pleomorphic LCIS [12]. ALH and classic LCIS lesions are thought to represent non-obligate precursors in the pathogenesis of future breast neoplasia in the ipsilateral or contralateral breast. ALH is defined by an epithelial growth within the acini of terminal duct lobular units (TDLU) [3,4,5,6,7,8,9,10,11,12,13]. On histology, ALH consists of small, round uniform cells, that are non-adhesive and display increased nuclear to cytoplasmic ratio with minimal nuclear atypia (Fig. 1). In addition, the E-cadherin stain is negative. A negative E-cadherin is also seen with LCIS and invasive lobular carcinoma (ILC) [6].

Fig. 1
figure 1

Histopathology of ALH (arrows, H&E × 20)

Full size image

The criteria to differentiate ALH from classic LCIS are based on the extension of acini within a lobular unit. [14]. ALH is defined as less than 50% of distended acini. (R1).

Oncogenesis and/or progression of LCIS are defined by mRNA expression levels of MKI67 [15]. In addition, the data favor the hypothesis that classic LCIS is a non-obligate precursor of invasive breast cancer, ductal or lobular [16, 17].

Imaging appearance

Mammogram

The most common presentation of lobular neoplasia is calcification, occurring in 60–90% of LN lesions diagnosed on breast core needle biopsies [18]. The most frequent calcification morphology of lobular neoplasia are pleomorphic (57%), then amorphous (24%), and lastly coarse heterogeneous (19%) [18].

The most common presentation of ALH is grouped amorphous calcifications [18,19,20] (Fig. 2). Less often, the calcifications may be regional in distribution and fine pleomorphic or punctuate in morphology. Infrequently, ALH may present as a mass or architectural distortion. Ibrahim et al. showed among 56 lesions yielding ALH, 87% presented as calcifications, 11% as architectural distortion and 2% as masses [19]. Foster et al. described 14 pure ALH lesions among 6,081 patients. Of the pure ALH lesions, there were 12 calcifications cases, 1 mass case, and 1 developing density case mammographically [21] (Fig. 3).

Fig. 2
figure 2

Mammographic appearance of ALH as calcifications. a 2 cm, grouped, amorphous calcifications in the lateral aspect of the left breast (arrows). Magnification view. b 1 cm, grouped, course heterogeneous and pleomorphic calcifications in the central aspect of the left breast. Pathology yielded ALH with calcifications at both sites (open arrow)

Full size image
Fig. 3
figure 3

Mammographic appearance of ALH as a mass. a Full-field digital mammogram CC image demonstrates a 2 cm, oval mass in the central right breast, posterior depth (arrow). b The mass persisted on tomosynthesis imaging (arrow). It was biopsied under tomographic guidance since there was no definite ultrasound correlate. Pathology yielded ALH

Full size image

Ultrasound

Data regarding sonographic ALH are scant [3, 20]. Ferre et al. reported the US features of ALH displaying the appearance of a hypoechoic mass with an oval shape (67%), circumscribed or lobulated margins (73%), posterior enhancement (73%), and parallel in orientation (53%) [20]. In contrast, Malherbe et al. reported that the ultrasound characteristics most frequently seen for ALH included a mass that was irregular in shape (86%), hypoechoic in echogenicity (88%), having a poorly circumscribed margin (95%), with posterior acoustic enhancement (93%), and absent calcifications (81%) [22] (Fig. 4).

Fig. 4
figure 4

Sonographic appearance of ALH. Right breast ultrasound demonstrates a less than 1 cm, irregular, hypoechoic, not circumscribed, not-parallel mass with posterior shadowing (circle). Mammogram not shown. A needle core biopsy was performed under ultrasound-guidance. Pathology revealed ALH. The lesion was removed surgically due to radiology–pathology discordance. On final surgical excision, the pathology yielded invasive lobular carcinoma with ALH

Full size image

MRI

Data regarding MRI findings of ALH are limited [20]. In a study by Ferre et al., they reported that the most common imaging findings of ALH were non-mass enhancement (NME) (71%), and least common a focus (7%) [20] (Fig. 5).

Fig. 5
figure 5

A high-risk, female patient underwent screening MR imaging. Post-gadolinium, axial T1-weighted, subtracted image demonstrates linear non-mass enhancement within the posterior central aspect of the right breast (arrow). This was biopsied under MR-guidance and represented ALH. The area was surgically excised and also yielded ALH

Full size image

To our knowledge, Amos et al. described the largest cohort of LNs seen on MRI [18]. Twenty-eight percent (5/18) of the LN cases with non-incidental concordant imaging findings presented as either a NME or focus [18]. Among these cases, the majority 60% (3/5) were NME demonstrating a linear clumped pattern and 40% (2/5) of cases presented as a focus. In addition, the authors reported that 80% of their ALH cases were seen on preoperative breast MRI in patients with newly diagnosed cancer (18).

Outcome of ALH?

Any high-risk lesion has the risk of upgrade after surgical excision. In addition, the patient is at increased risk of developing breast carcinoma [23].

Premalignant potential of ALH

The first study that demonstrated a risk of developing malignancy in patients with atypia was published by Dupont and Page in 1985 [24]. They showed patients with atypia have a 10 times increased risk of developing breast cancer. More interestingly, they demonstrated that the malignancies most often occurred in the ipsilateral breast diagnosed with ALH, rather than in the contralateral breast [24]. In a study by Myers et al., the authors estimated that ALH is associated with a fourfold to fivefold increased lifetime risk of developing breast neoplasia in either breast [9]. In addition, the risk of developing breast cancer in a patient diagnosed with ALH further increases with a family history of breast carcinoma and an earlier age of diagnosis [11, 23,24,25]. (R2).

The largest cohort of ALH and ADH was published by Hartmann et al. [3] They followed 698 women with atypical (ductal and lobular) hyperplasia for a mean of 12.5 years. They demonstrated that among the women who developed breast carcinoma after the diagnosis of ALH, 87% presented with invasive carcinoma (invasive ductal carcinoma), and of those cancers 53% were grade 2. In addition, they reported that 31% of the patients developed breast cancer within the 5 first years, 24% at 6–10 years, and 49% > 10 years after diagnosis of the high-risk lesion [3]. Moreover, they showed that cancers occurred with a 2:1 ratio. Interestingly, there was no difference in risk between ADH and ALH. In the Mayo study, they showed the cumulative incidence of breast cancer at 25 years was 29%, with cancers developing within 5 years of the biopsy demonstrating atypia [3].

Risk of upgrade

The risk of upgrade to cancer involving ALH is variable in the literature. In the study by Lewin et al., the authors reported the upgrade rates of ALH diagnosed on needle biopsy ranged from 0 to 67%, with an average upgrade rate of 13% [26]. Among 949 excised lesions, Lewin et al. reported that 47 (5.0%) of the lesions were upgraded, including 18 DCIS and 15 invasive mammary carcinomas. In the study by Hartmann et al., they demonstrated that there were additional risk factors in patients with atypia, such as multiple foci of atypia, and younger age of at the time biopsy that increased the risk of upgrade [3]. However, there is an increasing number of publications suggesting that upgrade is much lower (< 5%) with small volume lobular neoplasia [18, 27].

Management

Considering published data with small cohorts, ALH management remains controversial given the risks of over and under treatment. Surgical excision used to be recommended when a needle core biopsy case yielded the pathology of ALH because of its malignant potential and could be detected in the vicinity of an invasive breast cancer. However, the management paradigm is changing [28]. The patient should be referred to a breast surgeon/High-Risk Clinic for further management evaluation and discussion to include possible surgical excision, risk reduction strategies, prophylaxis therapies, and supplemental screening breast MRI.

The American Society of Breast Surgeons recommends surveillance in a patient with ALH if the following criteria are met: [28]

  • Radiological–pathological correlation

  • No associated lesion(s) that would require surgical excision

  • Small-volume lesion

Moreover, MD Anderson recommends surgical excision of ALH for cases of radiological–pathological discordance, targeted versus incidental lesions, in cases with fewer core samples taken, and for a mass lesion [29].

When surgery is deferred, short-term follow-up with imaging (mammogram and a possible breast MRI) is recommended. Some authors have also suggested that imaging follow-up should be considered in cases of MRI-guided biopsies yielding ALH [30].

Conclusion

ALH is a high-risk lesion that can display heterogeneity on imaging. Its management continues to evolve with an emphasis on radiological–pathological correlation and the need for the assessment of a family and personal history of breast cancer in the patient [31]. Thus, there remains the need for larger studies to ensure appropriate management recommendations for this high-risk lesion and for our patients.