The Role of E-Cadherin in the Biology of the Tubulolobular Carcinoma of the Breast

Details

Supervisors

Pr Mieke VAN BOCKSTAL

Faculty

Faculté de médecine et médecine dentaire

Degree label

Master [180] en médecine, à finalité spécialisée

Abstract

Tubulolobular carcinoma (TLC) is a rare breast tumor (1% of invasive breast cancers), whose classification remains difficult because of its morphological diversity and its inconsistent E-cadherin (ECAD) expression. The underlying cellular biological mechanism is not yet clear. E-cadherin, a transmembrane tumor suppressor protein, encoded by the CDH1 gene, is involved in the intercellular binding of epithelial cells at junctions while catenin p120 (p120cat) binds it to the cellular cytoskeleton. We studied the expression profile of ECAD/p120cat on TLCs operated at the Cliniques Universitaires Saint-Luc between 2015 and 2020 (9 pure tumors and 3 mixed tumors) and compared their molecular identity established by Next Generation Sequencing (NGS) and shallow Whole Genome Sequencing (sWGS). By immunohistochemistry, we observed in 7 out of 11 analyzable cases, that TLC expresses ECAD in an aberrant way: the localization is cytoplasmic, sometimes mixed (membrane / cytoplasmic), faint or punctiform. The p120cat perfectly follows the ECAD profile: an aberrant ECAD expression leads automatically to p120cat cytoplasmic relocalization. Molecular analysis (10 analyzable TLC cases) revealed that a mutation in the CDH1 gene systematically results in aberrant ECAD expression. We also observed the high frequency of PIK3CA mutations within TLCs (7/10), and their association with CDH1 mutations (6/7), a molecular profile described in invasive lobular carcinomas (ILC). For mixed tumors, we found common CDH1 mutations in a TLC/ILC combination, suggesting that these two morphologically different cell populations have a common clonal ancestor. In a TLC/tubular carcinoma (TubCa) combination, the finding of a common molecular and immunohistochemical profile suggests a mechanism of junction restoration by the p-cadherin (PCAD) pathway recently described in the literature. All our results tend to support a lobular origin of TLC. The lesion should therefore be classified as a subvariant of invasive Lobular Carcinoma, as it is already classified in the WHO Classification of Mammary Tumors. However, the terminology could be revised and the name "Invasive Lobular Carcinoma with Tubular Elements (ILC-TE)" appears to be more appropriate. We also propose the adoption of a new diagnostic algorithm for lesions of mixed lobular/tubular morphology, based on the expression of ECAD. A positive membrane staining in both components indicates a ductal lesion, which could be renamed "Invasive Ductal Carcinoma with Lobular Elements". An aberrant membrane labeling in both components indicates a lobular lesion “Invasive Lobular carcinoma with Tubular Elements” (instead of TLC). And an association of a normal aberrant and tubular lobular marking is in favor of either two different lesions (colliding tumors) either a Mixed Ductal Lobular Carcinoma. The distinction between both requires the use of NGS analysis. NGS could also be used in all cases of lobular morphology with retained membrane labeling (15% of ILCs), to detect a possible mutation in the CDH1 gene responsible for the production of a detected but non-functional ECAD.