Keynote Presentation: High-Throughput Interrogation of the Tumor Microenvironment Using Digital Image Analysis

Multiplex immunofluorescence is a powerful tool for phenotyping cells from a single tissue slide, while preserving spatial context. Multiplex immunofluorescence can be used to profile dozens of biomarkers from a single four-micron tissue section, enabling high-throughput biomarker discovery without exhausting precious tissue or disrupting tissue architecture. Accurate and reproducible image analysis is a key aspect to generating multiplex IF datasets. This talk will cover some image analysis strategies for generating data from multiplex IF images, as well as some challenges associated with multiplex IF. In particular, the use of deep-learning based pipelines to perform the segmentation and classification tasks needed for reliable single-cell phenotyping.  Finally, spatial analytics build upon single-cell phenotyping to reveal the complex cell interactions that define the tumor microenvironment. We illustrate two case studies using multiplex IF: characterizing tertiary lymphoid structures in colorectal carcinomas; and detecting cell-cell interactions in the perivascular niche of triple-negative breast cancers. Multiplex IF, when combined with accurate digital image analysis, yields high-quality multiplex immunofluorescence datasets and facilitates deep interrogation of the tumor microenvironment.

Learning Objectives:

1. Summarize the key steps for image analysis of multiplex IF.

2. Review the challenges of multiplex IF analysis and ways to mitigate them.

3. Explain several use cases of multiplex IF technology.