Invasive Lobular Carcinoma
Invasive lobular carcinoma (ILC) is characterized by noncohesive cells, usually E-cadherin negative, that are dispersed or arranged in single-file linear pattern, but can show several additional architectural patterns (Tavassoli and Eusebi 2009; Lakhani et al. 2012). ILC is frequently associated with lobular carcinoma in situ (LCIS).
Incidence: ILC is the second most common type of breast cancer and constitutes 5–15% of all invasive breast tumors (Lakhani et al. 2012), although incidence varies according to the criteria applied for diagnosis. The incidence of ILC is increasing, especially among postmenopausal women, probably as consequence of hormonal replacement therapy and alcohol abuse (Rakha and Ellis 2010).
Age: ILC usually affects peri and postmenopausal women aged 50 or more but it can be seen in a wide age range of patients, from young, in their third decade of life, to elderly women, aged more than 80 years (Tavassoli and Eusebi 2009).
Sex: ILC affects more frequently female patients and it is rare in men.
Site: ILC arises in the breast parenchyma, affecting any of the breast quadrants. Bilaterality and multifocality are seen in ILC more frequently than in other breast cancers (Rakha and Ellis 2010). Paget’s disease is not associated with ILC.
ILC may present as a palpable breast mass of variable size. Mammographic features are often difficult to interpret, and this may impair early diagnosis. ILC is the most common type of cancer among mammographically undetected cancers (Tavassoli and Eusebi 2009). Furthermore, Magnetic Resonance Imaging (MRI) is more effective to estimate both the size and the bilaterality of ILC than mammography. Preoperative diagnosis, when based on fine needle aspiration cytology can lead to inadequate results, as a consequence of low cellularity of ILC.
Treatment: Surgery associated with endocrine therapy constitutes the treatment of choice in operable cases; however, adjuvant chemotherapy can be useful in advanced cases.
Outcome: ILC usually shows favorable prognostic profile (low or intermediate nuclear grade, high expression of estrogen (ER) and progesterone receptors (PR), low proliferative index). On the other side, ILC presents at a more advanced stage than other types of carcinoma. Studies focused on ILC prognosis led to conflicting results, failing to demonstrate a better prognosis when ILC is compared to invasive carcinoma of no special type (NST). Distant metastases can occur after many years suggesting a worse long-term prognosis of ILC than of breast carcinoma NST (Rakha and Ellis 2010; Lakhani et al. 2012). ILC metastatic pattern shows a greater tendency toward abdominal dissemination (comprising metastases to gastro-intestinal tract, uterus, and ovary) at presentation. In addition, brain, meninges, bone, pleural and peritoneal serosal membranes may be involved (Tavassoli and Eusebi 2009; Lakhani et al. 2012).
ILC can show a great variety of morphological variants, whose prognostic impact has been poorly studied. According to the data obtained by Orvieto et al. (2008), the classical variant shows a better prognosis than other morphological variants of ILC, due to a lower number of recurrences and metastases.
Histological grade is related to ILC prognosis (Rakha and Ellis 2010). ILC grade 3 shows more frequent aggressive prognostic parameters as advanced T stage, vascular invasion, lymph-node metastases at presentation. As expected, these features are related to poorer outcome, when compared with ILC grade 2.
Macroscopic features vary according to mammography and clinical presentation. About half of the cases present as an ill-defined, firm mass. In addition, ILC can present as multiple minute nodules. Due to the diffuse growth pattern in about 30% of the cases, it is very difficult to detect any nodule, the breast parenchyma being fibrous and firm (Tavassoli and Eusebi 2009).
ILC shows a wide spectrum of different architectural and histological features (Tavassoli and Eusebi 2009) that should be recognized in order to achieve a correct diagnosis.
Architectural histological variants are often intermingled one to the other and the predominant pattern (over 80% of tumor structure) determines the specific tumor type (Tavassoli and Eusebi 2009).
ILC-C neoplastic cells can invade diffusely the mammary gland, in a “spider–web” fashion (Tot 2016). This type of growth indicates poor prognosis. Mitotic count is low and necrosis is rarely seen in this variant.
Mucinous variant (ILC-Muc) is a rare variant of ILC that shows abundant extracellular mucin deposition (Cserni et al. 2017). Mucin deposition can be focal or diffuse. ILC-Muc neoplastic cells are similar to those seen in ILC-C.
Tubulo-lobular variant (ILC-TL) is characterized by a mixture of tubular carcinoma and ILC (Tavassoli and Eusebi 2009).
All these features may manifest within each histologic variants.
Apocrine variant (ILC-A): the neoplastic cells show eosinophilic and granular cytoplasm, central nucleus with evident nucleolus. Diffuse positivity for GCDFP-15 is usually encountered. ILC-A may be associated with LCIS with apocrine features.
Myoepithelial cell variant (ILC-M): rare cases showing evidence of myoepithelial cell differentiation are on record (Del Vecchio et al. 2005; Tavassoli and Eusebi 2009). The cases described by Del Vecchio et al. (2005) were composed of cells having features intermediate between epithelial-secretory and myoepithelial (therefore the name of “myosecretory” was applied) (Del Vecchio et al. 2005). The myosecretory features were present both in the invasive and in situ components.
Pleomorphic variant (ILC-P): this variant is described in a dedicated chapter (Pleomorphic Invasive Lobular Carcinoma).
All ILC variants are frequently associated with in situ lobular carcinoma (LCIS) (In Situ Lobular Carcinoma) or to in situ duct carcinoma (DCIS) of low nuclear grade. LCIS is frequently multifocal and can be localized within the invasive component or in the surrounding parenchyma, as well as in other quadrants or in the contralateral breast (Foschini et al. 2006).
ILC grading: ILC is usually graded following the Nottingham grading system applied in all breast carcinomas (Rakha and Ellis 2010). ILC grade can be difficult to assess due to the lack of tubule formation and low mitotic index. The vast majority of ILC are grade 2. Grade 3 ILC are more frequently encountered among the special variants, as ILC-S and ILC-P.
Nevertheless, studies based on large cohort of patients, with long-term follow-up information, demonstrate that ILC grading might predict prognosis.
Tumor Infiltrating lymphocytes (TIL) are scanty in ILC. Prominent TIL can be observed in those variants showing more aggressive features, as ILC-P, ILC-S, and in ILC grade 3 cases. On very rare occasions TIL can be so prominent that ILC acquire the features of lymphoepithelioma-like carcinoma (Cristina et al. 2000).
Frozen section interpretation of sentinel node metastases from ILC is difficult, as single neoplastic cells intermingled with lymphocytes can be missed on a quick intra-operative evaluation. To this purpose, the application of rapid cytokeratin immunohistochemical staining can be of help to highlight the neoplastic cells.
Loss of E-cadherin expression is the most typical immunohistochemical feature of ILC neoplastic cells (Dabbs et al. 2013) as to be used in the differential diagnosis between ILC and other breast carcinoma histotypes. In ILC-C, E-cadherin loss is almost complete; no staining is visible along the cell membrane of the neoplastic cells. In some ILC variants, E-cadherin loss can be partial and the neoplastic cells can show a residual, partial positivity along the cell membrane (Canas-Marques and Schnitt 2016).
ILC, in addition to loss of E-cadherin expression, is characterized by altered P120 catenin expression within the cytoplasm instead of the membrane (Canas-Marques and Schnitt 2016). Combined immunohistochemical reactions, showing E-cadherin loss and P120 cytoplasmic expression, can be of help in the correct ILC diagnosis.
ILC cells in addition express low-molecular weight cytokeratins, cytokeratin 34BetaE12, GATA3, mammaglobin, GCDFP-15, Epithelial membrane antigen (EMA), cathepsin D, and cyclin D1 (Rakha and Ellis 2010). Myoepithelial cell markers are usually negative with the exception of the “hybrid” variant (Del Vecchio et al. 2005).
HER2 overexpression or amplification is an extremely rare event in ILC and is almost exclusively seen in the pleomorphic variant of ILC.
Proliferative activity, evaluated immunohistochemically with Ki67 antibody, is lower in ILC than in invasive breast carcinoma NST. High proliferative index is seen in ILC grade 3 or in the special variants with aggressive behavior.
The key molecular alterations detected in ILC are those affecting the CDH1 gene, mapped on 16q22.1, whose protein product is E-cadherin (Dabbs et al. 2013).
A wide spectrum of genetic or epigenetic alterations can affect the CDH1 gene, as deletions, mutations, and methylation. The most frequent alterations, detected in ILC, are 16q loss and gains of 1q and 16p. In addition, several other genetic aberrations have been found, as 8q gain and losses at 8p23-p21, 11q14.1q25, 13q, Xq (Rakha and Ellis 2010).
The E-cadherin transcriptional regulation can be altered by numerous different transcription factors as activation of the TGFβ pathway, SNAIL and SLUG upregulation, and ZEB1 expression (Dabbs et al. 2013).
The different genetic mechanisms leading to E-cadherin inactivation and\or protein loss are reflected by the different patterns of E-cadherin expression detected on immunohistochemistry (Dabbs et al. 2013; Canas-Marques and Schnitt 2016).
CDH1 mutations were enriched in multifocal rather than unifocal tumors, suggesting that these aberrations might be implicated in ILC spreading within the breast stroma. They may be probably related even to the peculiar pattern of metastatic dissemination to organs such as the GI tract, ovaries, and peritoneum due to critical role of cadherin in cell adhesion.
Other four key genes have been reported to be frequently mutated in ILC: PIK3CA (~45%), PTEN (usually LOH or mutations in ~13% of cases), AKT1 (~13%), and HER2 (~5.1%) (Desmedt et al. 2016).
The transcriptomic characterization of 141 cases of ILC based on a standardized molecular assay revealed that 84% of samples were classified as luminal A or luminal B (Ciriello et al. 2015). A more recent study using immunohistochemistry surrogates evaluating 981 primary ILC (Iorfida et al. 2012) confirmed those data, where 41% and 57% of the ILC samples pertained to the luminal A and B subtype, respectively.
Genetic studies based on clonal analyses demonstrate a relation between LCIS and ILC (Morandi et al. 2006). These data have been confirmed recently by an independent study (Sakr et al. 2016), indicating that LCIS and ILC have a similar repertoire of somatic mutations, with PIK3CA and CDH1 being the most frequently mutated genes.
The spectrum of differential diagnoses to be considered in cases of ILC varies according to ILC variant.
Differential diagnosis is more frequently carried out with invasive carcinoma NST, especially in cases of ILC-S. To this purpose, in addition to the morphological features described above, absence or reduced E-cadherin expression is of help. In more difficult cases, cytoplasmic P120 expression can be added.
Differential diagnosis between ILC-C and polymorphous (low-grade) adenocarcinoma (PLGA) is described in a specific chapter (Polymorphous Carcinoma). Shortly, when morphology is difficult to interpret, E-cadherin fragmented expression and reduced CK7 expression as observed in PLGA can help in the differential diagnosis. In addition, BCL2 positivity and ER\AR negativity are frequently seen in PLGA. Epithelioid Myofibroblastoma (Myofibroblastoma) is extremely rare, but may mimic ILC-C, since the epithelioid cells are variously arranged in single cells, single files or nests. These cells are positive for ER and PR and unlike ILC-C are variably positive also for vimentin, desmin, alpha-smooth muscle actin, CD34, CD99, and CD10.
ILC-H should be differentiated from several reactive and neoplastic conditions (Eusebi et al. 1995; Tan et al. 2011). Differential diagnosis can be difficult especially intraoperatively on frozen sections. Focal inflammation, rich in histiocytes, is the most frequent condition to be considered. Differential diagnosis can be especially difficult when ILC-H is associated with prominent TIL. ILC-H usually shows nuclear atypia and it is frequently associated with in situ carcinoma. On permanent sections, the strong positivity on immunohistochemistry for CK7 and GATA3 epithelial markers, together with the high ER, PR, and AR expression, helps in the differential diagnosis.
Differential diagnosis between ILC-H and granular cell tumor is discussed in a specific chapter (Granular Cell Tumor). Shortly, the diagnosis of granular cell tumor should be considered for those lesions lacking nuclear atypia and carcinoma in situ; on permanent sections, granular cell tumor has a specific immunoprofile (S100 positivity) and lacks cytokeratin positivity.
ILC-H and ILC-A should be differentiated from invasive carcinoma with apocrine changes (Apocrine Carcinoma). To this purpose, the absence or decreased expression of E-cadherin together with cytoplasmic P120 are the main hallmarks of ILC.
ILC-SR can be difficult to differentiate from signet ring cell carcinomas arising in different organs especially in cases of ILC-SR presenting with gastric of gastro-intestinal metastases. Breast ILC-SR lesions show strong positivity for ER, PR, and AR, and GATA3, a feature usually lacking in similar carcinomas arising in organs different from breast.
ILC-M is a rare variant that should be differentiated mainly from invasive carcinoma NST (Invasive Carcinoma NST). It has a bland cytology similar to ILC-C, partial loss of E-Cadherin and a concurring expression of epithelial and myoepithelial cell markers (Del Vecchio et al. 2005).
ILC-P is discussed in detail in the dedicated chapter (Pleomorphic Lobular Carcinoma).
- Ciriello, G., Gatza, M. L., Beck, A. H., Wilkerson, M. D., Rhie, S. K., Pastore, A., Zhang, H., McLellan, M., Yau, C., Kandoth, C., Bowlby, R., Shen, H., Hayat, S., Fieldhouse, R., Lester, S. C., Tse, G. M. K., Factor, R. E., Collins, L. C., Allison, K. H., Chen, Y. Y., Jensen, K., Johnson, N. B., Oesterreich, S., Mills, G. B., Cherniack, A. D., Robertson, G., Benz, C., Sander, C., Laird, P. W., Hoadley, K. A., King, T. A., Research Network, T. C. G. A., & Perou, C. M. (2015). Comprehensive molecular portraits of invasive lobular breast cancer. Cell, 163(2), 506–519.CrossRefPubMedPubMedCentralGoogle Scholar
- Cserni, G., Floris, G., Koufopoulos, N., Kovács, A., Nonni, A., Regitnig, P., Stahls, A., & Varga, Z. (2017). Invasive lobular carcinoma with extracellular mucin production—a novel pattern of lobular carcinomas of the breast. Clinico-pathological description of eight cases. Virchows Arch, 471(1), 3–12.CrossRefPubMedGoogle Scholar
- Dabbs, D. J., Schnitt, S. J., Geyer, F. C., Weigelt, B., Baehner, F. L., Decker, T., Eusebi, V., Fox, S. B., Ichihara, S., Lakhani, S. R., Palacios, J., Rakha, E., Richardson, A. L., Schmitt, F. C., Tan, P. H., Tse, G. M., Vincent-Salomon, A., Ellis, I. O., Badve, S., & Reis-Filho, J. S. (2013). Lobular neoplasia of the breast revisited with emphasis on the role of E-cadherin immunohistochemistry. Am J Surg Pathol, 37(7), e1–11.CrossRefPubMedGoogle Scholar
- Desmedt, C., Zoppoli, G., Gundem, G., Pruneri, G., Larsimont, D., Fornili, M., Fumagalli, D., Brown, D., Rothé, F., Vincent, D., Kheddoumi, N., Rouas, G., Majjaj, S., Brohée, S., Van Loo, P., Maisonneuve, P., Salgado, R., Van Brussel, T., Lambrechts, D., Bose, R., Metzger, O., Galant, C., Bertucci, F., Piccart-Gebhart, M., Viale, G., Biganzoli, E., Campbell, P. J., & Sotiriou, C. (2016). Genomic characterization of primary invasive lobular breast cancer. J Clin Oncol, 34(16), 1872–1881.CrossRefPubMedGoogle Scholar
- Eusebi, V., Foschini, M. P., Bussolati, G., Rosen, P. P. (1995). Myoblastomatoid (histiocytoid) carcinoma of the breast. A type of apocrine carcinoma. Am J Surg Pathol, 19(5), 553–62.Google Scholar
- Iorfida, M., Maiorano, E., Orvieto, E., Maisonneuve, P., Bottiglieri, L., Rotmensz, N., Montagna, E., Dellapasqua, S., Veronesi, P., Galimberti, V., Luini, A., Goldhirsch, A., Colleoni, M., & Viale, G. (2012). Invasive lobular breast cancer: subtypes and outcome. Breast Cancer Res Treat, 133(2), 713–723.CrossRefPubMedGoogle Scholar
- Lakhani, S. R., Rakha, E. A., & Simpson, P. T. (2012). Invasive lobular carcinoma. In S. R. Lakhani, I. O. Ellis, S. J. Schnitt, P. H. Tan, & M. van de Vijver (Eds.), WHO classification of tumors of the breast (4th ed., pp. 40–42). Lyon: IARC Press.Google Scholar
- Orvieto, E., Maiorano, E., Bottiglieri, L., Maisonneuve, P., Rotmensz, N., Galimberti, V., Luini, A., Brenelli, F., Gatti, G., & Viale, G. (2008). Clinicopathological characteristics of invasive lobular carcinoma of the breast: results of an analysis of 530 cases from a single institution. Cancer, 113(7), 1511–1520.CrossRefPubMedGoogle Scholar
- Sakr, R. A., Schizas, M., Carniello, J. V., Ng, C. K., Piscuoglio, S., Giri, D., Andrade, V. P., De Brot, M., Lim, R. S., Towers, R., Weigelt, B., Reis-Filho, J. S., & King, T. A. (2016). Targeted capture massively parallel sequencing analysis of LCIS and invasive lobular cancer: Repertoire of somatic genetic alterations and clonal relationships. Mol Oncol, 10(2), 360–370.CrossRefPubMedGoogle Scholar
- Tavassoli, F. A., & Eusebi, V. (2009). Invasive lobular carcinoma. In Tumors of the mammary gland, AFIP Atlas of tumor pathology, series 4 (pp. 156–168). Washington, DC: American Registry of Pathology.Google Scholar
- van Deurzen, C. H., Cserni, G., Bianchi, S., Vezzosi, V., Arisio, R., Wesseling, J., Asslaber, M., Foschini, M. P., Sapino, A., Castellano, I., Callagy, G., Faverly, D., Martin-Martinez, M. D., Quinn, C., Amendoeira, I., Kulka, J., Reiner-Concin, A., Cordoba, A., Seldenrijk, C. A., van Diest, P. J. (2010). Nodal-stage classification in invasive lobular breast carcinoma: influence of different interpretations of the pTNM classification. J Clin Oncol, 28(6), 999–1004.Google Scholar