Abstract P1-03-03: Multichannel serial imaging of transgenic, preclinical murine models provides the first quantitative analysis of the unusual growth kinetics and lymph-vascular invasion of patient-derived inflammatory breast cancer cells and tumor emboli

Abstract

Background: Lymphovascular invasion (LVI), a critical feature of advanced cancers, is a major route of metastatic dissemination. Of the clinically-distinct types of breast cancer, the most lethal variant is inflammatory breast cancer (IBC) where LVI is a histopathological hallmark. Majority of IBC patients lack a distinct, solid tumor and instead present with diffuse tumor cell clusters cells, also termed as tumor emboli, in the breast and dermal lymphatics, which contributes to the clinical breast-skin symptoms and postulated to provide an efficient path to metastatic spread. The steps in lymphatic dissemination are challenging to study in vitro, and in vivo models are limited. Our goal was to develop new in vivo models that would enable direct visualization and quantitation of local tumor cell growth and tumor-lymphatic vessel interactions for improved understanding of the unique IBC tumor biology and for drug discovery. Methods: mCherry/GFP-expressing non-IBC (4T1, E0771) and IBC cells derived from patient untreated primary tumors (SUM149/basal and SUM190/Her2+) were implanted into the fascia in the center of a window chamber in the skin-fold or mammary fat pad. In order to allow direct study of tumor cell–lymphatic vessel interaction, we generated transgenic nude mice with fluorescent lymphatics [tdTomato fluorophore under control of a Prox1 promoter, encodes a transcription factor (prospero-related homeobox 1) necessary for the formation and maintenance of lymphatic vessels]. Multichannel microscopy was employed to serially quantify tumor growth, tumor spread and lymph-tumor interactions. Using this model, we also optimized implantation and live imaging of pre-formed IBC-cell derived tumor emboli generated by simulating the lymphatic sheer stress in culture. Results: Compared to non-IBC cells, which formed solid tumors that increased in size over a 14 day period, IBC models exhibited a diffuse and disseminating phenotype, within 24h, rather than a centralized tumor mass, similar to the disease presentation in patients. Using a threshold analysis in ImageJ, we quantified tumor area (as a metric of tumor growth), the tumor motility, and the linear density of lymph and blood vessels. Briefly, we binarized each fluorescent channel image such that the tumor or lymph signal was positive; then, an algorithm drew regions of interest (ROIs) around each defined tumor-cell cluster. The ROIs were used to calculate the parameters describing tumor growth and motility (distance of the tumor from the center of mass (CM)), and the area moment (the movement of the clusters relative to the CM) at each timepoint. The linear vascular density was calculated from the distance of segments drawn through the center of large vessels for each mouse, which provided quantitative data describing tumor cell or IBC emboli and lymphovascular interactions, suggesting that there is a specific microenvironment necessary for the unique phenotype and dissemination pattern exhibited by IBC tumor cells. Conclusions: Despite the poor prognosis of IBC, the clinical implications of how and why tumor emboli form and how they survive to migrate into the lymphatics is not defined. These in vivo models provide in-depth imaging and quantitative measurements of locoregional invasion, tumor cell-lymphatic or endothelial vessel interactions of implanted tumor cells or tumor emboli. This model has the potential to be extended to study other cancer cell types exhibiting LVI as well as a screening metric for IBC therapies. Supported by DOD-Breakthrough-W81XWH-17-1-029, IBC Research Foundation, DCI pilot funds, IBC Network Foundation, Duke Surgery Gardner Award. Citation Format: Ashlyn Rickard, Pranalee Patel, Scott J Sauer, Mark W Dewhirst, Gregory M Palmer, Gayathri R Devi. Multichannel serial imaging of transgenic, preclinical murine models provides the first quantitative analysis of the unusual growth kinetics and lymph-vascular invasion of patient-derived inflammatory breast cancer cells and tumor emboli [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P1-03-03.

Influence of Macrophages on Vascular Invasion of Inflammatory Breast Cancer Emboli Measured Using an In Vitro Microfluidic Multi-Cellular Platform

“…Currently, there is a lack of histological differentiations or biomarkers that can effectively distinguish between IBC tumors and non-IBC tumors with a high level of sensitivity and specificity. However, clinical observations of IBC pathologic specimens indicate that they exhibit higher levels of angiogenesis and metastasis, along with increased involvement of lymph nodes compared to non-IBC cases [4] and extensive lymphovascular emboli contributing to the characteristic breast and skin swelling [5]. Recent studies of the tumor microenvironment and breast stroma demonstrate a role for macrophages in IBC presentation and progression [6][7][8].…”