|
Abstract INTRODUCTION:
Recent advances in spatial omics technologies have transformed our ability to study how tissue structure, organization, and function changes throughout the course of disease. By simultaneously measuring the molecular profiles and spatial location of individual cells within their native tissue context, these approaches have been used to gain fundamental insights into how cellular interactions influence tissue homeostasis, infectious disease, neurodegeneration, or cancer progression. In cancer, coordinated interactions between heterogeneous cell types within the tumor microenvironment play a critical role in regulating tumor growth, invasion, metastatic dissemination, and overall treatment effectiveness. Although spatial profiling technologies have improved our ability to investigate the regulatory mechanisms that drive patient outcomes, analyzing and interpreting these data across heterogeneous patient samples in a generalized and scalable way still presents a significant computational challenge.
OBJECTIVES:
This study aims to (1) develop a computational spatial enrichment method that can accurately identify differences in spatial organization across clinical patient groups, (2) benchmark performance across existing tools and platforms, and (3) apply this framework to identify spatial niches that are predictive of metastatic recurrence in triple negative breast cancer (TNBC).
METHODS:
We developed QUICHE (QUantitative InterCellular nicHe Enrichment), a statistical framework designed to discover cellular niches differentially-enriched in patient groups, histological structures, or acellular regions. Unlike previous approaches, QUICHE combines local niche detection with interpretable statistical modeling using graph neighborhoods to detect differentially enriched cellular niches, even when their prevalence is low.
RESULTS:
Using in silico models and spatial proteomic imaging of human tissues, we show that QUICHE can accurately detect condition-specific cellular niches occurring at low frequency and prevalence, outperforming the next best algorithm by 3-fold in niche recovery by prevalence. To validate our approach and understand how tumor structure influences recurrence risk in triple negative breast cancer, we used multiplexed proteomic imaging by time of flight (MIBI-TOF) to comprehensively profile the tumor microenvironment in a multi-center TNBC cohort consisting of 314 primary surgical resections across 5 institutions. We discovered cellular niches that were consistently enriched in key regions of the tumor microenvironment, including the tumor-immune border and extracellular matrix remodeling regions, as well as niches statistically-associated with patient outcomes, including recurrence status and recurrence-free survival. The majority of differential niches (74.2%) were specific to patients that did not relapse and formed a robust interconnected network enriched in monocytes, macrophages, APCs, and CD8T cells with tumor and stroma cells. In contrast, the interaction network for patients that relapsed was notably sparse and enriched in B cells, CD68 macrophages and neutrophils. These findings were validated in two independent cohorts, demonstrating consistent cellular interaction patterns and predictive performance.
CONCLUSION:
This study demonstrates how spatial profiling technologies, local niche detection, and interpretable statistical modeling can uncover cellular niches that lead to differential patient outcomes. Our findings suggest that effective anti-tumor responses in TNBC are not defined by individual cell types, but rather by coordinated, spatially-organized networks involving both innate and adaptive immune cells interacting with the tumor and stroma. |
= Discovery stage. (53.14%, 2025)
= Translation stage. (22.33%, 2025)
= Clinically available. (24.53%, 2025)
