Analysis of ROS1 expression in invasive lobular breast carcinomas
Lobular breast cancer, e-cadherin, ROS1, Crizotinib
The invasive lobular carcinoma, a special type of breast cancer, represents about 10% of this organ carcinomas. Associated to BRCA2 and CDH1 genes mutations, has a discohesive cellular phenotype, as a result of CDH1 gene pathogenic variants, which codifies E-cadherin adhesion molecules. Germline pathogenic variants are highly frequent in patients harboring Hereditary Diffuse Gastric Cancer syndrome and hereditary lobular breast cancer, in a dominant autosomal heritage and incomplete penetrance patternMore than 90% of them express estrogen and progesterone receptors, typically have low Ki67 and do not express HER2, belonging to the Luminal A molecular subtype, exhibiting a good prognosis when treated in the same way as non-special type carcinoma. Although there is currently no specific treatment for lobular carcinomas, there is pre-clinical description of increased ROS1 expression and synthetic lethality between E-cadherin deficiency and ROS1 tyrosine kinase inhibition, using ROS1 tyrosine kinase inhibitors, which results in apoptosis. However, there is no data in the literature correlating ROS1 expression in human tumors with E-cadherin expression and clinical-pathological data. The objective of the study was to analyze the potential of ROS1 expression, by immunohistochemistry, as a prognostic marker and to correlate the findings with "in silico" data from transcriptomes of CLIs in international open databases. This was an experimental cross-sectional study, approved by CEP/FS-UnB, with 79 cases of CLI, from the archives of two private and public institutions in Brasília, between 2015 and 2019. The mean age of the patients was 58.9 years, with data such as laterality, location, and multicentricity similar to those already described in the literature. ROS1, the main biomarker of the study, was expressed in only 3 cases, having H-scores ranging from 15 to 240. All three had Ki67 greater than or equal to 20% (P=0.026) and estrogen receptor expression in less than 20% of neoplastic cells (P=0.000015). No statistical significance was found between the pattern of ROS1 expression and histological type, histological grade, lymphovascular invasion, margin status, lymph node stage, TNM stage, progesterone receptor expression (PR), HER2 overexpression or low expression, or E-cadherin or p120-Catenin expression pattern. Despite the small number of cases with ROS1 expression in the study, suggesting a different cellular homeostasis pattern between humans and rodents, the correlation with higher rates of cell proliferation may correspond to a worse prognosis for patients who present it. The second arm of the study, correlating with "in silico" data, is ongoing and will be presented in a complementary manuscript.