Abstract Study background: Inflammatory Breast Cancer (IBC) is the most metastatic variant of breast cancer. Although IBC is recognized as a distinct variant of breast cancer, the molecular basis for the rapid progression of IBC remains largely undefined, in part due to the lack of preclinical models that recapitulate the human disease as well as a lack of comprehensive analysis of the preclinical models of IBC that are available. Methods: All available 7 pre-clinical models of IBC, including 2 new models, FC-IBC01 and FC-IBC02 developed from pleural effusion, were used to identify genes and molecular pathways that are selectively altered compared to non IBC breast tumor models. Laser capture micro dissection of biopsy tissue from core biopsy and skin punch biopsies were also analyzed by whole transcriptome analysis. Results: Whole transcriptome analysis defined 7 pre-clinical models of IBC as being within either the triple negative or ErbB2/Her-2 expressing subtypes, similar to the prevalence of these subtypes of breast cancer observed in IBC patients. Comparative analysis of the FC-IBC01, FC-IBC02 and Mary-X models of IBC demonstrated that each of these recapitulate the formation of tumor emboli with encircling lymphovasculogenesis. The majority (6/7) of the pre-clinical models of IBC express CDH1, which encodes for E-cadherin, which was associated with a loss of ZEB1, a transcriptional repressor of E-cadherin. The lack of ZEB1 expression was validated in a limited set of 4 skin punch biopsy samples from IBC patients that were isolated by laser capture micro-dissection, demonstrating concordance with loss of ZEB1 in pre-clinical models of IBC. Expression of other transcription factors involved in acquisition of a cancer stem cell phenotype, including Snai1, which encodes for Snail, SNAI2, which encodes for Slug and TWIST1, was retained in pre-clinical models of IBC. Maintenance of E-cadherin in pre-clinical models of IBC was associated with a loss of genes within the transforming growth factor beta (TGFβ signaling pathway, with expression of SMAD6, a known repressor of TGFβ. This is similar to a recent study reporting the persistence of E-cadherin and loss of TGFβ signaling in IBC patient tumors based on gene profiling of 3 independent data sets. Conclusion: The present studies provide first time comparison of gene signatures of 7 pre-clinical models of IBC, including our 2 newly developed pre-clinical models, FC-IBC01 and FC-IBC02, that recapitulate formation of tumor emboli with encircling lymphatic vessels, similar to that observed in biopsy tissues of IBC patients. We demonstrate that E-cadherin expression was associated with both loss of ZEB1 and diminished expression of multiple genes within the TGFβ signaling pathway, with retention of expression of transcription factors and surface markers consistent with maintenance of a cancer stem cell phenotype, as has been reported to be a characteristic of IBC tumors. Collectively, these observations provide first time characterization of the molecular signatures of all available pre-clinical models of IBC, and suggest that IBC has a signature of epithelial plasticity, with characteristics of their ability to undergo the mesenchymal to epithelial reverting transition. The loss of genes within the TGFβ signaling is also consistent with the tight cell: cell aggregation of IBC tumor cells within tumor emboli that exhibit “cohesive invasion”. The new pre-clinical models of IBC that recapitulate the human disease will serve as useful tools to accelerate our understanding of the molecular underpinnings and therapeutic targets of IBC as the most lethal form of breast cancer.
Read full abstract