Abstract
Tumors are populated by a multitude of immune cell types with varied phenotypic and functional properties, which can either promote or inhibit anti-tumor responses. Appropriate localization and function of these cells within tumors is critical for protective immunity, with CD8 T cell infiltration being a biomarker of disease outcome and therapeutic efficacy. Recent multiplexed imaging approaches have revealed highly complex patterns of localization for these immune cell subsets and the generation of distinct tumor microenvironments (TMEs), which can vary among cancer types, individuals, and within individual tumors. While it is recognized that TMEs play a pivotal role in disease progression, a better understanding of their composition, organization, and heterogeneity, as well as how distinct TMEs are reshaped with immunotherapy, is necessary. Here, we performed spatial analysis using multi-parameter confocal imaging, histocytometry, and CytoMAP to study the microanatomical organization of immune cells in two widely used preclinical cancer models, the MC38 colorectal and KPC pancreatic murine tumors engineered to express human carcinoembryonic antigen (CEA). Immune responses were examined in either unperturbed tumors or after immunotherapy with a CEA T cell bispecific (CEA-TCB) surrogate antibody and anti-PD-L1 treatment. CEA-TCB mono and combination immunotherapy markedly enhanced intra-tumoral cellularity of CD8 T cells, dominantly driven by the expansion of TCF1-PD1+ effector T cells and with more minor increases in TCF1+PD1+ resource CD8 T cells. The majority of infiltrating T cells, particularly resource CD8 T cells, were colocalized with dendritic cells (DCs) or activated MHCII+ macrophages, but largely avoided the deeper tumor nest regions composed of cancer cells and non-activated macrophages. These myeloid cell – T cell aggregates were found in close proximity to tumor blood vessels, generating perivascular immune niches. This perivascular TME was present in untreated samples and markedly increased after CEA-TCB therapy, with its relative abundance positively associated with response to therapy. Together, these studies demonstrate the utility of advanced spatial analysis in cancer research by revealing that blood vessels are key organizational hubs of innate and adaptive immune cells within tumors, and suggesting the likely relevance of the perivascular immune TME in disease outcome.
Highlights
When tumors reached 100-300mm3 in volume, animals were randomized into the following treatment groups: one group received the murine surrogate carcinoembryonic antigen (CEA)-targeted T cell bi-specific antibody (CEATCB), which simultaneously binds to the CEA protein on cancer cells and CD3 on T cells and elicits a T cell mediated attack on CEA-expressing tumors, independent of T cell receptor specificity [50, 79]; one group was treated with the checkpoint inhibitor, anti-PD-L1, and a third group was treated with the combination of CEA-T-cell bispecific (TCB) and aPD-L1 antibody
APD-L1 treatment alone failed to produce a similar magnitude response. These studies confirmed published observations that the CEA-TCB immunotherapy markedly enhances anti-tumor responses by CD8 T cells, and these are further amplified by combination therapy with aPDL1 [79]
We employed high-resolution multiparameter confocal imaging, histocytometry, and computational spatial analysis with CytoMAP to resolve the complexity of the tumor microenvironment (TME) in two preclinical murine cancer models, the MC38 colorectal and KPC pancreatic tumors with and without immunotherapy
Summary
Multiplexed imaging and spatially resolved sequencing technologies have revealed complex cellular organization across tissue types and diverse pathological conditions [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19]. Advanced computational and statistical approaches applied to such datasets allow quantification of various spatial properties, such as cellular distance relationships, preferential cell-cell associations, and organization of tissue microenvironments [16, 20,21,22,23,24,25,26] This in turn allows a data-driven interrogation of how spatial context influences the transcriptional, phenotypic and functional changes within individual cells, and reveals how disorganization of cells can lead to disease pathology. Over a wide variety of samples, specific positional patterns for immune cells have been detected, which correlate with responses to immune therapy This suggests that microscopy-based technologies may be able to parse out complex cellular patterning in highly heterogeneous tissues and have the potential to serve as powerful tools for companion diagnostics or prognostic studies [11, 16, 34,35,36,37]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
