American Association for Cancer Research
Browse
crc-23-0212_fig7.png (1.08 MB)

FIGURE 7 from Peritumoral Immune-suppressive Mechanisms Impede Intratumoral Lymphocyte Infiltration into Colorectal Cancer Liver versus Lung Metastases

Download (1.08 MB)
figure
posted on 2023-10-12, 14:20 authored by Jian Ye, Weihua Guo, Chongkai Wang, Colt A. Egelston, Massimo D'Apuzzo, Geereesh Shankar, Marwan G. Fakih, Peter P. Lee

Higher immune activity in lung metastasis, while more immune suppressive in the liver and peritoneal metastases. A, Representative image of ROI selection in GeoMx digital spatial profiling from a colorectal cancer lung metastasis. Cancer island and stroma within tumor core were identified by the expression of Pan-CK. Cell identity markers from metastatic colorectal cancer tumors (B) and markers highly expressing in the liver (C), lung (D), and peritoneal (E) metastases in GeoMx assay. Paired comparison of gene expression between each two organs in the different histologic regions shown in bottom row. F, GeoMx markers expressing in the LA. Statistical significance was determined by Wilcoxon signed-rank test for B–F.

Funding

HHS | NIH | National Cancer Institute (NCI)

History

ARTICLE ABSTRACT

Patients with microsatellite stable (MSS) colorectal cancer with liver metastases are resistant to immune checkpoint inhibitor (ICI) therapy, while about one-third of patients with colorectal cancer without liver metastases, particularly those with lung-only metastases, respond to ICI. We analyzed primary colorectal cancer tumors and major metastatic sites (liver, lung, peritoneal) using multiplex immunofluorescence and whole-slide spatial analyses to identify variations in immune contexture and regional localization within the tumor microenvironment. While levels of T and B cells within peritumoral regions were similar, their levels were significantly lower within the tumor core of liver and peritoneal metastases compared with lung metastases. In contrast, antigen-presenting cells (APC) and APC–T cell interactions were more abundant in all regions of lung metastases. We also identified an abundance of lymphoid aggregates throughout lung metastases, but these were present only within peritumoral regions of liver and peritoneal metastases. Larger lymphoid aggregates consistent with features of tertiary lymphoid structures were observed within or adjacent to primary tumors, but not metastatic lesions. Our findings were validated using NanoString GeoMx DSP, which further showed that liver metastases had higher expression of immune-suppressive markers, while lung metastases showed higher proinflammatory activity and T-cell activation markers. Peritoneal metastases demonstrated higher expression of cancer-associated fibroblast–related proteins and upregulated PD-1/PD-L1 signaling molecules. Our results demonstrate that functional status and spatial distribution of immune cells vary significantly across different metastatic sites. These findings suggest that metastatic site–dependent immune contexture may underlie discordant responses to ICI therapy in patients with MSS colorectal cancer. Our results demonstrate that functional status and spatial distribution of immune cells vary significantly across different metastatic sites in MSS colorectal cancer. These findings suggest that metastatic site–dependent immune contexture may underlie discordant responses to ICI therapy in patients with MSS colorectal cancer.