Unlocking Spatial Insights in Colorectal Cancer
Colorectal cancer represents a major health challenge. While surgical intervention is often effective initially, some patients experience recurrent secondary disease with poor prognosis, highlighting the need for advanced therapies such as immunotherapies. Spatial biology techniques, including multiplexed imaging with Cell DIVE, can yield critical insights necessary for developing new treatments. By visualizing a 30-biomarker panel in a colon adenocarcinoma (CAC) sample, multiple biological pathways relevant to cancer progression, such as tumor vascularization, immune cell responses, and cell proliferation, were mapped. AI-powered analysis revealed deep biological relationships between cell types. This method provides an understanding of the molecular and architectural dynamics of cancer progression in this significant cancer subtype.
Explore Cell DIVE Data with Minerva
To showcase Cell DIVE's capabilities, we offer several complete datasets accessible through the Minerva image viewer. Minerva is a lightweight, narrative-based image browser developed by scientists at Harvard University, designed to streamline the demonstration and sharing of histopathology datasets among researchers [1]. This tool enables users to examine full-resolution Cell DIVE datasets directly in their web browser. With guided narrations, users can explore various biological contexts, pinpoint areas of interest, interpret staining patterns, and grasp the rationale behind different biomarkers. These narrations assist in designing a Cell DIVE study, assessing the effectiveness of the imaging solution, and formulating approaches to multiplexed imaging.
Defining Cell Boundaries with AI and Multiplexed Imaging
In this Minerva story, we identify several key findings in a colon adenocarcinoma sample, and demonstrate several analysis methods, usable in software such as Aivia, for getting the most out of spatial biology data. Scientists often leverage image analysis software to define cell boundaries, analyze cell phenotypes, and measure morphological and spatial characteristics of specific cell types. We use a technique known as image segmentation to begin the process of extracting that information. For this task, we use the 2D spatial biology tools available in Aivia, which uses AI-based segmentation to deliver highly accurate segmentation maps of tissue that reflect actual cell boundaries. Once individual cell boundaries are mapped, we can use the relative biomarker signals within those territories to assign biomarker positivity to cells and ultimately identify their individual phenotype. Multiplexed imaging with Cell DIVE allows a higher number of biomarkers to be included in a study, allowing segmentation markers such as NaKATPase to be utilized to aid downstream analysis.
The Minerva image viewer allows researchers to convert Cell DIVE images into quantitative data through analysis. This application not only demonstrates the potential of Cell DIVE imaging but also serves as a vital resource for planning and implementing multiplexed imaging studies.
See for yourself and explore the interactive data set:
