Invasive Lobular Carcinoma of the Breast: The Increasing Importance of This Special Subtype

Amy E. McCart Reed; Lauren Kalinowski; Peter T. Simpson; Sunil R. Lakhani


Breast Cancer Res. 2021;23(6) 

In This Article

Abstract and Introduction


Invasive lobular carcinoma (ILC) is the most common of the breast cancer special types, accounting for up to 15% of all breast cancer cases. ILCs are noted for their lack of E-cadherin function, which underpins their characteristic discohesive growth pattern, with cells arranged in single file and dispersed throughout the stroma. Typically, tumours are luminal in molecular subtype, being oestrogen and progesterone receptor positive, and HER2 negative. Since last reviewing the lobular literature (McCart Reed et al., Breast Cancer Res 17:12, 2015), there has been a considerable increase in research output focused on this tumour type, including studies into the pathology and management of disease, a high-resolution definition of the genomic landscape of tumours as well as the evolution of several potential therapeutic avenues. There abounds a huge amount of new data, which we will review herein.


Invasive lobular carcinoma is the most common 'special' histological subtype of invasive breast carcinoma. From an evolutionary point of view, these tumours arise from a family of non-obligate precursor lesions called atypical lobular hyperplasia (ALH) and lobular carcinoma in situ (LCIS), which may be collectively termed lobular neoplasia (LN). Even within this narrow spectrum of pre-invasive lesions and frank invasive carcinoma, there is significant morphological and biological heterogeneity. The multistep model of breast cancer progression[1,2] contends that although lobular carcinomas arise along the low-grade, ER-positive arm of the pathway (with low-grade, ER-positive ductal lesions), de-differentiation to higher grade lesions can occur through acquisition of alterations in oncogenes such as ERBB2 and TP53, producing a spectrum of heterogenous proliferations (Figure 1).

Figure 1.

Multistep model of the evolution of classic ILC and its morphological variants. A lineage of 'lobular' disease evolves from a normal epithelial cell on a background of a loss of E-cadherin expression and function, and key early somatic alterations involving gain of chromosome 1q, loss of 16q, and mutations in PIK3CA, AKT1, or PTEN. The morphological and molecular diversity of in situ and invasive lobular lesions is likely to be a result of the subsequently arising pattern of molecular alterations that drive progression. Atypical lobular hyperplasia (ALH) is distinguishable from lobular carcinoma in situ (LCIS) based on the extent of proliferation within the lobule. Pleomorphic LCIS (PLCIS) and florid LCIS (FLCIS) can emerge either from ALH (presumably) or from classic LCIS (CLCIS), with an increasing level of genomic complexity and the accumulation of mutations in driver genes such as ERBB2, ERBB3, and TP53. Various morphological variants of ILC have also been described (see also Figure 2), which exhibit either architectural or cytological atypia relative to the classic invasive type, which we imagine being the 'default' pathway of evolution. A number of important points to note: (1) the genomic alterations listed may arise during any stage of progression, though are likely to be acquired at the in situ stage, or earlier (e.g. amplification of 11q13 is evident in the in situ stage); (2) it is assumed FLCIS may progress to alveolar, solid, tubulo-lobular variants, or even the pleomorphic type; (3) it is uncommon for invasive tumours to be of a pure variant morphology, with tumours often also exhibiting classic and/or other variant patterns; (4) a variety of molecular alterations have been associated with some of these morphological variants, but these are not necessarily pathognomonic of the architectural variant; and (5) the interplay between the malignant cells and extracellular matrix may also impact the resulting growth pattern. -, loss; +, gain; dotted line, anticipated route of progression; solid line, demonstrated route of progression

Lobular neoplasia are mostly an incidental finding and comprise neoplastic proliferation of characteristically discohesive cells which fill and distend the terminal duct lobular units. LN encompasses both ALH and LCIS, and the boundary between the two is defined by an arbitrary cut-off using a quantitative measure, depending on the relative extent of involvement of the terminal duct lobular unit (TDLU); if more than 50% of the TDLU is occupied, the lesion is upgraded to LCIS. LN is considered to be a non-obligate precursor of invasive cancer, with ALH associated with a 4–5 times increased relative risk for subsequent cancer, and LCIS an increase of 8–10 times the risk; the risk is bilateral but predominates for the ipsilateral breast.[3] The clinical and morphological features of LCIS and its morphological variants have recently been extensively reviewed elsewhere and will not be covered herein.[4,5]

Classic invasive lobular carcinoma (ILC) typically demonstrates single cell infiltration and a characteristic targetoid pattern of growth with minimal associated stromal response[3] (and reviewed in[6]). This pattern of subtle invasion is such that the size of the tumour often exceeds the imaging findings and obtaining clear surgical margins may be challenging. Although ILCs are generally palpable, a high false-negative mammography rate is possible (in 19–43%; reviewed in[7]). In addition to the classic form of ILC, which is typically histological grade 2, there are special morphological subtypes including Pleomorphic, Solid, Alveolar, and Tubulo-lobular.[3,8–12] These variants are rarely seen as pure forms and are more likely to be present with the classical type. ILC and its subtypes are typified by a loss of cellular adhesion, frequently the result of biallelic inactivation (i.e. gene mutation combined with gene deletion) of the CDH1 gene encoding E-cadherin, although other mechanisms of expression loss also feature. ILCs are normally oestrogen (ER) and progesterone (PR) receptor positive, and as such patients are indicated for hormone therapy. Whilst the biological characteristics of ILC afford patients a good prognosis in the short term, it has become clear that the longer-term prognosis of ILC is frequently worse than for patients with the more commonly diagnosed invasive breast carcinoma of no special type (IBC-NST; invasive ductal carcinoma, IDC) (reviewed in[6]).

The metastatic presentation of ILC has long been considered unique,[13,14] with a predilection for common sites (liver, lung, bone), but also gastrointestinal and gynaecological sites of colonisation;[15,16] recent studies further support this. Inoue et al. showed that lung metastases were less prevalent, but peritoneal metastases are significantly higher in ILC (assumed predominantly classic ILC) compared to ER-positive IBC-NST.[17] A recent study of metastatic spread to gynaecological sites demonstrated an association with ILC and young age at diagnosis and confirmed earlier reports of the wide metastatic colonisation of ILC.[18] Immunophenotyping showed a heterogeneous interplay between hormone receptors and their co-factors during progression, including frequent downregulation of PR expression and variable changes between AR, GATA3, and FOXA1 seen in different metastases within the same patient.[18] Rarer presentations are increasingly being published in the literature, further highlighting the peculiar natural history of ILC. For example, numerous case reports of ILC seeding as orbital metastases appear to suggest these are more likely to arise from an ILC than other types, and in a sole example of a mixed ductal-lobular carcinoma, only the lobular component was found in the orbital metastasis (e.g.[19–22]).

In the last 5 years, an impressive body of work on ILC has amassed. There abounds a huge amount of new data, including studies into the pathology and management of disease, the genomic landscape of ILC and in particular somatic alterations associated with therapy resistance, and the evolution of several potential therapeutic avenues, which we will review herein.