Abstract 2382: Three-dimensional Spatial Phenotyping of Cellular Landscapes in the Bone Microenvironment of Spontaneous Metastases

Daniel S. Kermany,Weijie Zhang,Jianting Sheng,Matthew Vasquez,Xiang Zhang,Stephen T. Wong

DOI: https://doi.org/10.1158/1538-7445.am2023-2382

IF: 11.2

2023-01-01

Cancer Research

Abstract:Abstract Metastasis to distant organs is the major cause of cancer-related deaths, and bone is the most frequent destination of metastasis in many cancers. Recent studies have shown that the bone microenvironment, not only permits, but augments cancer cells’ invasion and spread within tissue. Recent evidence also suggests that NG2+ cells participate in the initiation of bone metastasis and enhance proliferation and migration via cell-to-cell interactions and often co-localize with early disseminated tumor cells (DTCs). Hence, we developed an imageomics framework to measure and model three-dimensional (3D) spatial relationships within the bone microenvironment using whole slide confocal imaging. To investigate the progression of DTCs, we delivered Lewis Lung Carcinoma (LLC1) GFP+ cancer cells to hind limb bones through intra-iliac artery injection. We did not observe immunogenic rejection of the cells or loss of markers during tumor progression. We performed whole slide confocal imaging to reconstruct high, single-cell resolution, 3D images of femur bones using NG2-creER;ROSA26-LoxP-TdTomato mice. This allowed us to detect and localize single spontaneous DTCs and NG2+ cells within the overall structure of the bone using anti-RFP and anti-GFP fluorescent markers. To quantify 3D spatial relationships in the bone microenvironment, cell positions were labeled within the image by setting a channel-specific minimum intensity threshold to remove background noise and weakly-stained objects, followed by removing all objects measuring <10 μm in more than one dimension. The remaining markers for DTCs were evaluated manually. DTCs were identified based on 3D shape, size, heterogeneity of stain markers, and absence of non-specific staining, while the coordinates (x,y,z) of each DTC were recorded. NG2+ cell positions were determined automatically; a black-hat transformation is used to remove background noise and enhance contrast, followed by thresholding to segment NG2+ cells. For segmented objects within the volume range of an expected cell (5-20 μm radius), the centroid coordinates of the resulting objects were recorded. To evaluate the spatial relationships in the stationary point patterns (SPPs) generated from the DTC and NG2+ staining, we implemented a 3D Ripley’s cross-K function to indicate spatial clustering or dispersion within and between SPPs to quantitatively evaluate spatial relationships at various distances and statistically compare against random distributions. We have developed an imageomics framework for 3D spatial relationships of different cells within the bone microenvironment using whole slide confocal imaging. Our framework can accurately assess the relationship between the two groups and evaluate additional elements within the bone microenvironment, and potentially other cancer-affected organ systems. Citation Format: Daniel S. Kermany, Weijie Zhang, Jianting Sheng, Matthew Vasquez, Xiang Zhang, Stephen T. Wong. Three-dimensional spatial phenotyping of cellular landscapes in the bone microenvironment of spontaneous metastases [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2382.